Congressional Testimony
Here's a look at documents involving congressional testimony and member statements
Featured Stories
AI Now Co-Executive Director West Testifies Before Senate Banking, Housing & Urban Affairs Committee
WASHINGTON, June 26 -- The Senate Banking, Housing and Urban Affairs Committee released the following testimony by AI Now Co-Executive Director Sarah Myers West from a June 11, 2026, hearing entitled "AI and the American Dream: Promoting Innovation, Affordability, and American Dominance":
* * *
Chairman Scott, Ranking Member Warren, and esteemed Members of the Committee, thank you for inviting me to testify on this important set of issues. My name is Dr. Sarah Myers West, and I along with my colleague Amba Kak lead the AI Now Institute, a policy research institute founded in 2017 to study artificial ... Show Full Article WASHINGTON, June 26 -- The Senate Banking, Housing and Urban Affairs Committee released the following testimony by AI Now Co-Executive Director Sarah Myers West from a June 11, 2026, hearing entitled "AI and the American Dream: Promoting Innovation, Affordability, and American Dominance": * * * Chairman Scott, Ranking Member Warren, and esteemed Members of the Committee, thank you for inviting me to testify on this important set of issues. My name is Dr. Sarah Myers West, and I along with my colleague Amba Kak lead the AI Now Institute, a policy research institute founded in 2017 to study artificialintelligence technologies. I hold doctoral and master's degrees from the Annenberg School at the University of Southern California, where I studied the political economy of the technology industry. I previously served as a Senior Advisor on Artificial Intelligence at the Federal Trade Commission, where my role focused on the effects of AI and other automated technologies emanating across the broader economy.
There is a tremendous amount of hope articulated in the vision for AI: that it has transformative potential that could lift our economy and improve the quality of life for all of us. Equally, there's a tremendous amount of pessimism within the broader public about what this technology is actually doing in the world, and what it means for their lives and livelihoods. This divide can be attributed to the skewed market incentives that have shaped the development of AI within the corporate labs that now dominate the scene. These incentives are leading to concentrated risk in the economy that could enact tremendous amount of damage to our national and economic
security - whether or not the "AI bubble" bursts. These harms, many of which are already here, could occur especially if things go exactly as the industry intends: we are in a race to the bottom, where the American people absorb all of the downside risks; higher prices, job loss and devaluation; and economic instability.
It's important to situate this in the longer trajectory of artificial intelligence. AI is a field with an almost seventy year history; the term 'AI' has been used to refer to an array of technological capabilities over time. The field and industry driving AI development has been shaped by laws and policies that have, among other outcomes, functioned to concentrate control over the industry in the hands of a small group of Silicon Valley firms.1
I've watched as over the past decade, the impact of AI on the economy at large has largely served to harm, rather than benefit, American workers, consumers, and small businesses - increasing the surveillance and control exerted over their lives, ratcheting up inequality2 and eliminating agency and dignity in their lives.
My message to this Committee is that this does not need to define our technological future: we can put the needs of the public back at the center of the AI economy. Congress has an obligation to intervene before an economic crisis occurs, starting by alleviating the burden already felt on people's pocketbooks, exercising Congressional oversight over AI firms and stopping a bailout of the industry that is arguably already underway.
There are three core areas of concern that I urge this Committee to consider as urgent priorities for intervention:
First, the nature of AI infrastructure investment is increasingly risky and capital intensive.
Growing reliance on complex and opaque financial instruments needs stronger oversight from financial regulators and Congress before the wider economy faces deepened risk.
For years, AI infrastructure investment was largely funded by the revenues of Big Tech firms, emanating from an interest in maintaining their dominant position in the cloud infrastructure market.3 But as AI developers arrived on very large scale AI as their primary focus for development, and thus the need to build more data center infrastructures to train and run inference on these large scale AI models, the industry began to take on debt to finance the astronomical capital expenditures.4 This was a strategic choice shaped by a high degree of concentration in the market and a perceived race between the leading firms to build very powerful artificial general intelligence models. It's become a race to the bottom, one in which these firms have exhibited an extraordinary willingness to court risk that is borne not only by the firms themselves, but by the wider economy and the public.
The justification AI firms offer is that the pursuit of artificial general intelligence will enable solutions to entrenched social problems ranging from curing cancer5 to solving the problem of climate change.6 Setting aside that the industry proffers privatized, AI-shaped solutions to these challenges, the primary constraint they assert is the infrastructure needed to bring AGI into being. This ignores, however, that as the pace of AI expenditures reaches new heights, capital is being pulled away from other potentially productive sectors of the economy, with the turn to credit within the AI sector potentially producing a 'crowding out' effect for other forms of investment and research.7 This could have broad-based impact: capital-intensive industries like manufacturing would feel this pressure acutely at a moment where these industries are already contending with the impact of tariffs.8 Small businesses are already downshifting their plans for capital expenditures.9 Moreover, the flood of capital to AI may undermine the horizon for future innovation as the market places a singular bet on AI.
To paint a picture of just how substantial these expenditures have grown: In the first quarter of 2026, Amazon, Google, Meta, and Microsoft reported spending $130.65 billion in capital expenditures for data centers--71 percent higher than what they spent during the same quarter of 2025./10 These four companies anticipate spending approximately $700 billion this year.11 Morgan Stanley analysts project that combined, Big Tech companies will spend nearly $3 trillion through 2028, but only generate half of that amount in cash.12
To finance this buildout, tech companies are taking on an increased amount of debt: in 2025, big tech firms issued $121 billion in new debt,13 but this number is anticipated to balloon in the years to come: Morgan Stanley is estimating debt issuance will top $500 billion this year.14 Some of this debt is structured through standard corporate bonds. But a growing share is being supplied by increasingly complex, risky, and opaque financial arrangements. Private equity firms are key conduits for the vast amounts of capital going into data centers. In 2025, private equity investment into data center transactions/deals reached $45 billion.15 Private credit firms also make up a growing share of the supply of debt for data center financing.16 This May, the Financial Stability Board warned private credit carries particular vulnerabilities for the broader economy, notably due to a lack of transparency and willingness to engage with borrowers that have lower credit quality and higher leverage.17 Many data center financing arrangements rely on distinct legal entities known as special-purpose vehicles (SPVs) to fund and execute deals. These deals are "off-balance sheet," meaning that the deal--and any debt the SPV has raised--is not listed on Big Tech firms' balance sheets, thus masking a clear picture of a company's true financial health, let alone the risk to the financial system as a whole.18
Over the last year we've seen the emergence of circular spending deals, where AI companies invest in and buy from one another--creating skewed market incentives and distorting decision-making.19
Increasingly these deals hinge on the projected future revenues of just a handful of AI developers, and if those fail to transpire to the tune needed to justify capital expenditures, the house of cards could come falling down.
When only a handful of buyers account for such a large share of the overall market, circular spending deals can drive cascading negative effects in a future market downturn.20 For example, OpenAI is on the hook for over a trillion in deals with other AI firms and chipmakers,21 including a $300 billion deal with Oracle for its compute infrastructure,22 $250 billion with Microsoft for its Azure infrastructure (which includes Microsoft's 26% return ownership share of OpenAI);23 $38 billion to rent access to Amazon Web Services servers, (a deal which could expand under a revised structure),24 $22 billion with CoreWeave for use of its data centers,25 a deal with Google Cloud,26 a "strategic partnership" to deploy 10 gigawatts of AI data centers with Nvidia (in return for Nvidia's $100 billion investment into OpenAI),27 a multi-billion dollar chip deal with AMD (a deal that enables OpenAI to take a 10% ownership stake in AMD),28 and a partnership with Broadcom to develop and deploy chips that OpenAI would design.29 This compounding set of bets assumes that OpenAI's revenue will substantiate its asserted demand, ignoring that if it fails to do so, it will also fail to fulfill its contractual obligations to these companies, many of which also have similar arrangements with each other. This scenario is prompting an increasing number of investors to turn to credit default swaps to hedge the downside risks associated with AI investments, a move that echoes the use of CDS to hedge against the risks of mortgage-backed securities in the lead-up to the housing crisis.30
The revenue needed for the industry to break even is becoming astronomical: a widely cited study from Bain estimates that $2 trillion in new revenue is needed by 2030 to fund the current AI scaling trend.31 Today, while revenues are quickly growing for AI firms, these are under conditions in which the cost of AI models is already subsidized. However, early indicators suggest there are concerns among major customers that they are not seeing sufficient return on investment to justify the amount they are spending even under current subsidies. And it's far from clear that a rising tide would lift all boats; a recent report from Bridgewater Associates found that a material part of the estimated GDP boost from AI would come from the profits of Nvidia and other chipmakers who capture the value without recycling it back into the broader economy.32
Last month, after burning through its entire 2026 AI budget in only four months, Uber's COO asserted the costs of AI were getting harder to justify because they were not translating into useful customer features.33 And Uber is not alone: a June 2026 Bain study found that nearly 40% of companies said their cost reductions from AI were significantly less than expected.34 More troubling, 44% of companies based their next wave of AI investments on previous rounds of savings--savings that have consistently come in below expectations.35 As the study put it, "self-funding the next wave from past returns sounds like discipline. In reality, it is a circular bet with a structural leak."36 Scaling revenue may be further challenged by delays in data center construction: recent reporting in the Financial Times found that 40% of data centers planned for 2026 are delayed, further increasing financial risk at the level of individual firm bets.37 The effects of an AI bubble would be acutely felt by the American people.
An AI bubble burst could wipe out over $20 trillion in American household wealth, three trillion more than the financial crisis.38 This is because the economy at large is already significantly exposed to AI risk across several vectors, ranging from stock market exposure to access to credit to job loss.39
For the roughly 60% of Americans who now hold stock, the fate of the Magnificent Seven stocks, which make up about a third of the market, will be felt more acutely than in prior financial crises.40 These market effects show up in index funds, in people's retirement portfolios, and their pensions. A recent Department of Labor proposal to open up 401ks to private equity, private credit and crypto assets would deepen this exposure and act, as AFR's Oscar Valdes Viera wrote, "a regulatory bailout hiding in plain sight and it would be a transfer of wealth from workers' retirement accounts to backstop funds and structures that are under pressure and hungry for new capital for their poorly performing assets."41
The deflation of stock portfolios can fuel a vicious cycle by affecting consumer spending, with effects felt across the broader economy, and it wouldn't take much to trigger it: even a modest decline in stock prices have produced an outsized drag on consumption: a hypothetical AI-driven selloff that erases $1 trillion in market value could translate into roughly $35 billion in reduced consumer spending, a significant shock to the broader economy.42
Banks are exposed to private credit and other nonbank financial institutions in ways that are largely opaque, and that combination of opacity and exposure to risk could pose particular dangers. Federal Reserve Vice Chair for Supervision Michelle Bowman recently warned that the withdrawal of private credit funding sources could pose a channel for the transmission of risk in the financial system.43
But the impact doesn't stop there: a major source of capital for private credit firms comes from working people. Private credit companies deploy capital that comes from 401(k) accounts, life insurance plans, and pensions; they've made a casino of American workers' financial security.
For example, both New York and Pennsylvania's state pension plans are invested in Blue Owl's $7 billion digital infrastructure fund, which in turn has loaned out money to finance data centers for Meta in Richland Parish, Louisiana.44
If Meta fails to post revenues that justify its planned $125 billion in spending for 2026--a move that led to a 6% decline in its stock attributed to investor anxiety45--the effects will be felt across the country.
We don't have to wait for a financial collapse to see these effects of AI play out across the economy.
Everyday Americans are already struggling with an environment of generalized inflation, in which the cost of essential goods and services is rising much faster than income levels.46 A growing consensus suggests that AI is a contributing factor, driving up prices.47 At least part of this story is directly tied to AI in the utilities sector, where wholesale electricity prices for households near data centers have increased by as much as 267% over the last five years.48 Working class Americans are particularly exposed to an AI-driven market downturn, because the effects they're already feeling could be deepened even further if wealthier Americans reduce their consumption. And there are indicators that the anticipation of AI's effects on the lives and livelihoods of white collar workers is already influencing a consumer demand, due to a fear of job loss and generalized environment of higher prices.49 This could be compounded by incentives for firms to attribute AI as the justification for layoffs, or for paying workers less, because it enables them to narrate retrenchment as a product of technological investment. In 2025, we already saw 55,000 layoffs attributed to AI, which amounts to about 5% of job losses that year.50
Importantly, many of these factors will remain true even in a scenario where there is no bubble burst: the vision for success in AI hinges on the significant displacement and devaluation of jobs across the economy, and because in real time people are making decisions based on future projections of what AI will look like, rather than what it is today.
For this reason, Congress shouldn't wait until an economic crisis is at hand to take steps to manage AI's destabilizing effect on the economy. These effects take multiple forms: a perpetual transfer of wealth from the public to AI firms, resource misallocation that may depress the economy, devaluation and dislocation in the job market, and rising prices that serve to depress consumer demand.
Congress must stop the bailout of the AI sector already underway.
As a first step, Congress should use its oversight authority to investigate the significant measures recently taken by the federal government to derisk the portfolio of AI firms.51
In November of 2025, OpenAI CFO Sarah Friar called for a government backstop for AI investment--a taxpayer funded guarantee that would allow companies like hers to continue their unprecedented spending on data centers and other AI investments.52 She was met with vocal backlash and CEO Sam Altman distanced himself from the assertion. But, in many ways, these firms are quietly getting what they asked for.53
The subsidies that AI firms have received are extraordinary, ranging from federal backing of a $1 billion loan to bring the Three Mile Island nuclear plant back online to power Microsoft's A.I. data centers,54 to offering up $1 billion in AI funding through the Big Beautiful Bill,55 to allocating federal lands for data center construction,56 outlining new 'private public partnerships' at national laboratories that house treasure troves of genomic data that can be leveraged for commercial use,57 to its Export AI initiative, which leverages the apparatus of the federal government in support of deal-brokering on behalf of AI firms.58
Looking at the evidence, a bailout of the AI sector has arguably already begun, before Congress has meaningfully acted to protect the public and the economy from the risks introduced by the industry. The public is already aware of the risks that they're being asked to assume, and this is fueling the generalized antipathy they're expressing toward AI. If the end goal is to promote American innovation and to build an economy that works for everyone, Congress must act. It can start by focusing on measures that alleviate the economic burdens already being felt by the public, shedding light on AI industry financing structures through use of its oversight authorities, and most importantly, by committing to stop the bailout before it deepens any further.
* * *
Original text here: https://www.banking.senate.gov/imo/media/doc/west_testimony_6-11-26.pdf
* * *
Chairman Scott, Ranking Member Warren, and esteemed Members of the Committee, thank you for inviting me to testify on this important set of issues. My name is Dr. Sarah Myers West, and I along with my colleague Amba Kak lead the AI Now Institute, a policy research institute founded in 2017 to study artificial ... Show Full Article WASHINGTON, June 26 -- The Senate Banking, Housing and Urban Affairs Committee released the following testimony by AI Now Co-Executive Director Sarah Myers West from a June 11, 2026, hearing entitled "AI and the American Dream: Promoting Innovation, Affordability, and American Dominance": * * * Chairman Scott, Ranking Member Warren, and esteemed Members of the Committee, thank you for inviting me to testify on this important set of issues. My name is Dr. Sarah Myers West, and I along with my colleague Amba Kak lead the AI Now Institute, a policy research institute founded in 2017 to study artificialintelligence technologies. I hold doctoral and master's degrees from the Annenberg School at the University of Southern California, where I studied the political economy of the technology industry. I previously served as a Senior Advisor on Artificial Intelligence at the Federal Trade Commission, where my role focused on the effects of AI and other automated technologies emanating across the broader economy.
There is a tremendous amount of hope articulated in the vision for AI: that it has transformative potential that could lift our economy and improve the quality of life for all of us. Equally, there's a tremendous amount of pessimism within the broader public about what this technology is actually doing in the world, and what it means for their lives and livelihoods. This divide can be attributed to the skewed market incentives that have shaped the development of AI within the corporate labs that now dominate the scene. These incentives are leading to concentrated risk in the economy that could enact tremendous amount of damage to our national and economic
security - whether or not the "AI bubble" bursts. These harms, many of which are already here, could occur especially if things go exactly as the industry intends: we are in a race to the bottom, where the American people absorb all of the downside risks; higher prices, job loss and devaluation; and economic instability.
It's important to situate this in the longer trajectory of artificial intelligence. AI is a field with an almost seventy year history; the term 'AI' has been used to refer to an array of technological capabilities over time. The field and industry driving AI development has been shaped by laws and policies that have, among other outcomes, functioned to concentrate control over the industry in the hands of a small group of Silicon Valley firms.1
I've watched as over the past decade, the impact of AI on the economy at large has largely served to harm, rather than benefit, American workers, consumers, and small businesses - increasing the surveillance and control exerted over their lives, ratcheting up inequality2 and eliminating agency and dignity in their lives.
My message to this Committee is that this does not need to define our technological future: we can put the needs of the public back at the center of the AI economy. Congress has an obligation to intervene before an economic crisis occurs, starting by alleviating the burden already felt on people's pocketbooks, exercising Congressional oversight over AI firms and stopping a bailout of the industry that is arguably already underway.
There are three core areas of concern that I urge this Committee to consider as urgent priorities for intervention:
First, the nature of AI infrastructure investment is increasingly risky and capital intensive.
Growing reliance on complex and opaque financial instruments needs stronger oversight from financial regulators and Congress before the wider economy faces deepened risk.
For years, AI infrastructure investment was largely funded by the revenues of Big Tech firms, emanating from an interest in maintaining their dominant position in the cloud infrastructure market.3 But as AI developers arrived on very large scale AI as their primary focus for development, and thus the need to build more data center infrastructures to train and run inference on these large scale AI models, the industry began to take on debt to finance the astronomical capital expenditures.4 This was a strategic choice shaped by a high degree of concentration in the market and a perceived race between the leading firms to build very powerful artificial general intelligence models. It's become a race to the bottom, one in which these firms have exhibited an extraordinary willingness to court risk that is borne not only by the firms themselves, but by the wider economy and the public.
The justification AI firms offer is that the pursuit of artificial general intelligence will enable solutions to entrenched social problems ranging from curing cancer5 to solving the problem of climate change.6 Setting aside that the industry proffers privatized, AI-shaped solutions to these challenges, the primary constraint they assert is the infrastructure needed to bring AGI into being. This ignores, however, that as the pace of AI expenditures reaches new heights, capital is being pulled away from other potentially productive sectors of the economy, with the turn to credit within the AI sector potentially producing a 'crowding out' effect for other forms of investment and research.7 This could have broad-based impact: capital-intensive industries like manufacturing would feel this pressure acutely at a moment where these industries are already contending with the impact of tariffs.8 Small businesses are already downshifting their plans for capital expenditures.9 Moreover, the flood of capital to AI may undermine the horizon for future innovation as the market places a singular bet on AI.
To paint a picture of just how substantial these expenditures have grown: In the first quarter of 2026, Amazon, Google, Meta, and Microsoft reported spending $130.65 billion in capital expenditures for data centers--71 percent higher than what they spent during the same quarter of 2025./10 These four companies anticipate spending approximately $700 billion this year.11 Morgan Stanley analysts project that combined, Big Tech companies will spend nearly $3 trillion through 2028, but only generate half of that amount in cash.12
To finance this buildout, tech companies are taking on an increased amount of debt: in 2025, big tech firms issued $121 billion in new debt,13 but this number is anticipated to balloon in the years to come: Morgan Stanley is estimating debt issuance will top $500 billion this year.14 Some of this debt is structured through standard corporate bonds. But a growing share is being supplied by increasingly complex, risky, and opaque financial arrangements. Private equity firms are key conduits for the vast amounts of capital going into data centers. In 2025, private equity investment into data center transactions/deals reached $45 billion.15 Private credit firms also make up a growing share of the supply of debt for data center financing.16 This May, the Financial Stability Board warned private credit carries particular vulnerabilities for the broader economy, notably due to a lack of transparency and willingness to engage with borrowers that have lower credit quality and higher leverage.17 Many data center financing arrangements rely on distinct legal entities known as special-purpose vehicles (SPVs) to fund and execute deals. These deals are "off-balance sheet," meaning that the deal--and any debt the SPV has raised--is not listed on Big Tech firms' balance sheets, thus masking a clear picture of a company's true financial health, let alone the risk to the financial system as a whole.18
Over the last year we've seen the emergence of circular spending deals, where AI companies invest in and buy from one another--creating skewed market incentives and distorting decision-making.19
Increasingly these deals hinge on the projected future revenues of just a handful of AI developers, and if those fail to transpire to the tune needed to justify capital expenditures, the house of cards could come falling down.
When only a handful of buyers account for such a large share of the overall market, circular spending deals can drive cascading negative effects in a future market downturn.20 For example, OpenAI is on the hook for over a trillion in deals with other AI firms and chipmakers,21 including a $300 billion deal with Oracle for its compute infrastructure,22 $250 billion with Microsoft for its Azure infrastructure (which includes Microsoft's 26% return ownership share of OpenAI);23 $38 billion to rent access to Amazon Web Services servers, (a deal which could expand under a revised structure),24 $22 billion with CoreWeave for use of its data centers,25 a deal with Google Cloud,26 a "strategic partnership" to deploy 10 gigawatts of AI data centers with Nvidia (in return for Nvidia's $100 billion investment into OpenAI),27 a multi-billion dollar chip deal with AMD (a deal that enables OpenAI to take a 10% ownership stake in AMD),28 and a partnership with Broadcom to develop and deploy chips that OpenAI would design.29 This compounding set of bets assumes that OpenAI's revenue will substantiate its asserted demand, ignoring that if it fails to do so, it will also fail to fulfill its contractual obligations to these companies, many of which also have similar arrangements with each other. This scenario is prompting an increasing number of investors to turn to credit default swaps to hedge the downside risks associated with AI investments, a move that echoes the use of CDS to hedge against the risks of mortgage-backed securities in the lead-up to the housing crisis.30
The revenue needed for the industry to break even is becoming astronomical: a widely cited study from Bain estimates that $2 trillion in new revenue is needed by 2030 to fund the current AI scaling trend.31 Today, while revenues are quickly growing for AI firms, these are under conditions in which the cost of AI models is already subsidized. However, early indicators suggest there are concerns among major customers that they are not seeing sufficient return on investment to justify the amount they are spending even under current subsidies. And it's far from clear that a rising tide would lift all boats; a recent report from Bridgewater Associates found that a material part of the estimated GDP boost from AI would come from the profits of Nvidia and other chipmakers who capture the value without recycling it back into the broader economy.32
Last month, after burning through its entire 2026 AI budget in only four months, Uber's COO asserted the costs of AI were getting harder to justify because they were not translating into useful customer features.33 And Uber is not alone: a June 2026 Bain study found that nearly 40% of companies said their cost reductions from AI were significantly less than expected.34 More troubling, 44% of companies based their next wave of AI investments on previous rounds of savings--savings that have consistently come in below expectations.35 As the study put it, "self-funding the next wave from past returns sounds like discipline. In reality, it is a circular bet with a structural leak."36 Scaling revenue may be further challenged by delays in data center construction: recent reporting in the Financial Times found that 40% of data centers planned for 2026 are delayed, further increasing financial risk at the level of individual firm bets.37 The effects of an AI bubble would be acutely felt by the American people.
An AI bubble burst could wipe out over $20 trillion in American household wealth, three trillion more than the financial crisis.38 This is because the economy at large is already significantly exposed to AI risk across several vectors, ranging from stock market exposure to access to credit to job loss.39
For the roughly 60% of Americans who now hold stock, the fate of the Magnificent Seven stocks, which make up about a third of the market, will be felt more acutely than in prior financial crises.40 These market effects show up in index funds, in people's retirement portfolios, and their pensions. A recent Department of Labor proposal to open up 401ks to private equity, private credit and crypto assets would deepen this exposure and act, as AFR's Oscar Valdes Viera wrote, "a regulatory bailout hiding in plain sight and it would be a transfer of wealth from workers' retirement accounts to backstop funds and structures that are under pressure and hungry for new capital for their poorly performing assets."41
The deflation of stock portfolios can fuel a vicious cycle by affecting consumer spending, with effects felt across the broader economy, and it wouldn't take much to trigger it: even a modest decline in stock prices have produced an outsized drag on consumption: a hypothetical AI-driven selloff that erases $1 trillion in market value could translate into roughly $35 billion in reduced consumer spending, a significant shock to the broader economy.42
Banks are exposed to private credit and other nonbank financial institutions in ways that are largely opaque, and that combination of opacity and exposure to risk could pose particular dangers. Federal Reserve Vice Chair for Supervision Michelle Bowman recently warned that the withdrawal of private credit funding sources could pose a channel for the transmission of risk in the financial system.43
But the impact doesn't stop there: a major source of capital for private credit firms comes from working people. Private credit companies deploy capital that comes from 401(k) accounts, life insurance plans, and pensions; they've made a casino of American workers' financial security.
For example, both New York and Pennsylvania's state pension plans are invested in Blue Owl's $7 billion digital infrastructure fund, which in turn has loaned out money to finance data centers for Meta in Richland Parish, Louisiana.44
If Meta fails to post revenues that justify its planned $125 billion in spending for 2026--a move that led to a 6% decline in its stock attributed to investor anxiety45--the effects will be felt across the country.
We don't have to wait for a financial collapse to see these effects of AI play out across the economy.
Everyday Americans are already struggling with an environment of generalized inflation, in which the cost of essential goods and services is rising much faster than income levels.46 A growing consensus suggests that AI is a contributing factor, driving up prices.47 At least part of this story is directly tied to AI in the utilities sector, where wholesale electricity prices for households near data centers have increased by as much as 267% over the last five years.48 Working class Americans are particularly exposed to an AI-driven market downturn, because the effects they're already feeling could be deepened even further if wealthier Americans reduce their consumption. And there are indicators that the anticipation of AI's effects on the lives and livelihoods of white collar workers is already influencing a consumer demand, due to a fear of job loss and generalized environment of higher prices.49 This could be compounded by incentives for firms to attribute AI as the justification for layoffs, or for paying workers less, because it enables them to narrate retrenchment as a product of technological investment. In 2025, we already saw 55,000 layoffs attributed to AI, which amounts to about 5% of job losses that year.50
Importantly, many of these factors will remain true even in a scenario where there is no bubble burst: the vision for success in AI hinges on the significant displacement and devaluation of jobs across the economy, and because in real time people are making decisions based on future projections of what AI will look like, rather than what it is today.
For this reason, Congress shouldn't wait until an economic crisis is at hand to take steps to manage AI's destabilizing effect on the economy. These effects take multiple forms: a perpetual transfer of wealth from the public to AI firms, resource misallocation that may depress the economy, devaluation and dislocation in the job market, and rising prices that serve to depress consumer demand.
Congress must stop the bailout of the AI sector already underway.
As a first step, Congress should use its oversight authority to investigate the significant measures recently taken by the federal government to derisk the portfolio of AI firms.51
In November of 2025, OpenAI CFO Sarah Friar called for a government backstop for AI investment--a taxpayer funded guarantee that would allow companies like hers to continue their unprecedented spending on data centers and other AI investments.52 She was met with vocal backlash and CEO Sam Altman distanced himself from the assertion. But, in many ways, these firms are quietly getting what they asked for.53
The subsidies that AI firms have received are extraordinary, ranging from federal backing of a $1 billion loan to bring the Three Mile Island nuclear plant back online to power Microsoft's A.I. data centers,54 to offering up $1 billion in AI funding through the Big Beautiful Bill,55 to allocating federal lands for data center construction,56 outlining new 'private public partnerships' at national laboratories that house treasure troves of genomic data that can be leveraged for commercial use,57 to its Export AI initiative, which leverages the apparatus of the federal government in support of deal-brokering on behalf of AI firms.58
Looking at the evidence, a bailout of the AI sector has arguably already begun, before Congress has meaningfully acted to protect the public and the economy from the risks introduced by the industry. The public is already aware of the risks that they're being asked to assume, and this is fueling the generalized antipathy they're expressing toward AI. If the end goal is to promote American innovation and to build an economy that works for everyone, Congress must act. It can start by focusing on measures that alleviate the economic burdens already being felt by the public, shedding light on AI industry financing structures through use of its oversight authorities, and most importantly, by committing to stop the bailout before it deepens any further.
* * *
Original text here: https://www.banking.senate.gov/imo/media/doc/west_testimony_6-11-26.pdf
Information Technology Industry Council Senior VP Flynn Testifies Before Senate Banking, Housing & Urban Affairs Committee
WASHINGTON, June 26 -- The Senate Banking, Housing and Urban Affairs Committee released the following written testimony by Mike Flynn, senior vice president and counsel at the Information Technology Industry Council, from a June 11, 2026, hearing entitled "AI and the American Dream: Promoting Innovation, Affordability, and American Dominance":
* * *
Chairman Scott, Ranking Member Warren, and members of the Committee, thank you for the opportunity to testify before you today. My name is Mike Flynn, and I am a Senior Vice President and Counsel at the Information Technology Industry Council (ITI).
ITI ... Show Full Article WASHINGTON, June 26 -- The Senate Banking, Housing and Urban Affairs Committee released the following written testimony by Mike Flynn, senior vice president and counsel at the Information Technology Industry Council, from a June 11, 2026, hearing entitled "AI and the American Dream: Promoting Innovation, Affordability, and American Dominance": * * * Chairman Scott, Ranking Member Warren, and members of the Committee, thank you for the opportunity to testify before you today. My name is Mike Flynn, and I am a Senior Vice President and Counsel at the Information Technology Industry Council (ITI). ITIrepresents 80 companies driving the nation's artificial intelligence (AI) development, adoption, and leadership. Our member companies operate at every layer of the AI technology stack: we represent chipmakers and designers, leading frontier AI labs, cloud service providers, and developers building the applications that deliver AI to customers worldwide.
Today, our members are investing hundreds of billions of dollars in the United States to do the frontier AI research, semiconductor fabrication and design, AI model training, and application development that make the American AI stack the global leader.
My testimony will provide an overview of the interconnected and interdependent American AI stack, including how AI is transforming cybersecurity, and examine what is needed to ensure America wins the global AI race. Finally, I will provide recommendations on how Congress can continue working to advance American AI leadership and promote U.S. economic security.
I. The American Artificial Intelligence Stack
AI systems are some of the most complex and technically sophisticated products ever made.
Frontier AI models rely on the work of thousands of highly skilled engineers, advanced semiconductors with features many thousands of times smaller than the width of a human hair, and networking that operates at the speed of light, to name only a few of the vital inputs needed to build and run a frontier AI model./1
These AI models, in turn, are used by billions of people across the world for everything from drafting emails and writing code to cutting-edge scientific research./2
Modern AI models are built by learning patterns from large amounts of text, images, code, or other data. When you ask the model a question, the model predicts the most likely useful response based on learned patterns. By integrating more data, better system architectures, and more compute, AI model developers have been able to rapidly improve AI's capabilities.
1 Gustav Grundin et al., "Issue Brief: AI Infrastructure," McKinsey & Company, February 27, 2026, https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/issue-brief-aiinfrastructure.
2 Microsoft AI Economy Institute, "Global AI Adoption in 2025: A Widening Digital Divide," Microsoft, January 8, 2026, https://www.microsoft.com/en-us/corporate-responsibility/topics/AI-Economy-Institute/reports/Global-AIAdoption-2025/.
* * *
a. The Interconnected and Interdependent AI Stack
When the average American thinks about AI, they often think first about the tools they can see and use. For example, they may think of the chatbot they can use to write a poem or a software assistant built into an email application. While these are the types of AI that most people interact with directly, they are only one kind of AI. AI systems are far more than chatbots, and they do everything from blocking fraudulent credit card transactions to improving the speed and efficiency of manufacturing processes./3
Critically, AI systems do not exist in a void; rather, they are part of the deeply interconnected and interdependent AI technology stack. This AI technology stack consists of upstream and downstream infrastructure and technologies that enable AI and is an interconnected set of five layers: infrastructure, data, model development and operations, applications, and governance, security, and trust./4
Overly broad measures aimed at one layer of the U.S. AI tech stack can have unintended consequences for other parts of the stack.
When considering AI policy in the context of the AI stack, it becomes clear that AI policy is about far more than semiconductors and models, two of the most scrutinized parts of the stack. Instead, AI policy should take into consideration everything from applications and data to critical minerals and semiconductor manufacturing equipment. Each layer of the AI stack raises different questions for policymakers, and effective policymaking requires treating the U.S. AI stack as an interconnected system, with the recognition that regulation in one area will have ripple effects across other layers of the stack.
For example, a regulation on AI models will affect how cloud companies allocate computing resources to customers, plan future data centers to meet demand, and, as a result, impact chip and equipment market dynamics. This means that to promote innovation, affordability, and American AI dominance, Congress must advance proposals for everything from export controls to AI model regulation that embrace the complex reality of the AI stack.
b. AI Innovation & Cybersecurity
American frontier developers have continued to invest in building the best AI models, drastically improving reasoning and context awareness to enable longer tasks and conversations. Due to these advances, frontier AI models have gained sophisticated capabilities-especially in software coding-and have driven a major shift in the cybersecurity ecosystem. The latest models can now identify and remediate software vulnerabilities at an extraordinary scale and speed.
* * *
3 Sheila Chiang, "AI and Machine Learning Helped Visa Combat $40 Billion in Fraud Activity," CNBC, July 26, 2024, https://www.cnbc.com/amp/2024/07/26/ai-and-machine-learning-helped-visa-combat-40-billion-in-fraudactivity.html.
4 Infrastructure includes the semiconductors, data centers, and energy sources necessary to build and run AI. Data includes raw inputs used in AI training. Model development and operations includes tools, platforms, and resources needed to serve AI models at scale. Applications are the part of the AI technology stack that most people interact with during daily life. Governance, security, and trust is a cross-cutting function that permeates every layer of the AI technology stack-from infrastructure to applications. Read more on the AI stack here: https://www.itic.org/documents/artificial-intelligence/ITI_AITechnologyStack.pdf.
* * *
In response to this critical moment for the cybersecurity community, ITI member companies have launched programs to collaborate with key organizations-critical infrastructure firms, essential software developers and providers, and cybersecurity companies-to enable them to use advanced models to get ahead of malicious cyber actors./5
The Trump Administration took a constructive, pro-innovation step to address AI and cybersecurity with its Executive Order, "Promoting Advanced Artificial Intelligence Innovation and Security."/6
This framework for collaboration among AI developers, the federal government, and trusted partners addresses the immediate need for cross-sector coordination on the cybersecurity implications of frontier models. To build on the direction of the Executive Order, Congress should codify and increase funding for the Department of Commerce's Center for AI Standards and Innovation (CAISI), which is doing critical work to assess frontier model capabilities and develop new standards around AI security.
c. Information Sharing and AI Model Distillation
Underpinning the Executive Order is the sharing of cybersecurity information between industry and government, and within industry. The Cybersecurity Information Sharing Act of 2015 (CISA 15) facilitates that sharing. CISA 15 is a foundational element of the U.S. cybersecurity ecosystem, providing critical liability protections to reduce legal risk for sharing information on cyber threat indicators, software vulnerabilities, and defensive measures. Currently, CISA 15 is temporarily reauthorized until September 30, 2026, and Congress must pass a long-term reauthorization to ensure that these essential information-sharing authorities do not lapse at a critical moment for the nation's cybersecurity.
As AI models develop capabilities directly relevant to national security, it is deeply concerning that U.S. adversaries are attempting to distill American frontier closed-source AI models, effectively enabling them to steal the work of U.S. AI labs. Countering distillation campaigns by adversaries will require improved information sharing between industry and government and
* * *
5 Anthropic, "Project Glasswing: Securing Critical Software for the AI Era," 2026, https://www.anthropic.com/glasswing; OpenAI, "Trusted Access for the Next Era of Cyber Defense," 2026, https://openai.com/index/scaling-trusted-access-for-cyber-defense/.
6 The White House, "Promoting Advanced Artificial Intelligence Innovation and Security," June 2, 2026, https://www.whitehouse.gov/presidential-actions/2026/06/promoting-advanced-artificial-intelligence-innovationand-security/.
* * *
within industry. Congress should act to address AI model distillation. Doing so will require consultation with industry to ensure proposals addressing distillation are scoped carefully to avoid unintended impacts on the U.S. AI stack, while applying lessons learned from the cybersecurity space to counter adversarial distillation.
II. Winning the Global AI Competition
a. Global Supply Chains and Markets Fuel American AI Leadership
American AI leadership depends on the U.S. AI stack's ability to participate in complex global supply chains and compete in markets around the world. To support continued U.S. AI leadership, policymakers should work with allies and partners to secure and diversify the supply chains that contribute to the U.S. AI stack. Such efforts could include convening formal supply chain reviews with allies and building upon existing efforts to ensure that market access barriers do not present impediments to global supply chains.
Just as the U.S. AI stack draws on global supply chains to build cutting-edge products, the U.S. AI tech stack must access and compete for customers in the global marketplace. Market access has been a primary driver of U.S. industry success and innovation, as most demand for many U.S. products is outside the United States. The U.S. government should strengthen existing bilateral and plurilateral trade agreements and negotiate new agreements to promote U.S. goods and services in markets abroad. Upholding the World Trade Organization's (WTO) Moratorium on Customs Duties on Electronic Transmissions will also be critical to supporting AI innovation, economic growth, and supply chain resiliency.
Sales abroad help U.S. companies invest in research, development, and manufacturing in the United States, spurring job growth and ensuring that America is first to build next-generation technologies. The numbers demonstrate the importance of overseas sales for U.S. semiconductor producers in particular: "in 2023, sales outside the U.S. comprised about threequarters of total U.S. semiconductor industry sales," and significant revenue from those sales is invested back into the United States./7
To foster U.S. sales abroad, ITI encourages the U.S. government to focus on reducing foreign tariff and non-tariff barriers, eliminating duplicative certification and conformity assessment requirements by foreign governments in areas where international certifications exist, promoting strong protection of intellectual property rights in foreign markets, and seeking commitments from foreign partners that enable digital trade and the deployment of AI across all sectors of the economy.
* * *
7 "State of the U.S. Semiconductor Industry 2024," Semiconductor Industry Association, https://www.semiconductors.org/2024-state-of-the-u-s-semiconductor-industry/.
* * *
U.S. companies face fierce competition in the global marketplace and seek to be partners of choice based on quality, dependability, quantity, cost, timeliness, and other factors. To successfully engage customers around the world, companies need predictability and certainty.
Permanent trade agreements and frameworks, as opposed to potentially ever-changing tariffs, help to create market stability and encourage customers and counterparts in Asia and Europe to be open and productive partners with U.S. companies, helping strengthen supply chains at home and abroad.
To ensure U.S. economic security, U.S. companies must remain partners of choice in the global marketplace. Policymakers should not take U.S. AI leadership for granted, especially as more governments take meaningful steps to foster their domestic AI ecosystems. That is why ITI supports the American AI Exports Program, a whole-of-government approach to facilitating additional U.S. exports to the $235 billion global AI market./8
By promoting the export of American AI technology, the program is inherently aimed at decreasing international adoption of AI technologies developed by countries like China. To support this critical work, Congress should codify the American AI Exports Program by passing H.R. 6996, the Full AI Stack Export Promotion Act.
b. U.S. Economic Security and AI
Maintaining and expanding America's AI lead and ensuring the U.S. stack is adopted across the world is critical to U.S. economic and national security. Congress has done important bipartisan work over the last several years to support American AI leadership, notably passing the CHIPS and Science Act, the Comprehensive Outbound Investment National Security (COINS) Act, and restoring the immediate expensing of domestic research and development costs.
U.S. companies need clarity and certainty to sign contracts with customers, spend billions of dollars on custom chips, and continue the trillion-dollar investments in America that are making the U.S. the undisputed AI leader. This is why ITI has consistently emphasized the importance of stakeholder engagement in developing and finalizing proposals at the intersection of national security, trade, and technology. Absent meaningful engagement with industry and allies, broadly scoped and unilateral measures risk limiting both market opportunities for U.S. companies and the effectiveness of export controls.
As it considers export control proposals, Congress should seek to both promote the American AI stack and protect U.S. economic security. This is not a binary choice; Congress can and should optimize for both by codifying and supporting initiatives like the Department of Commerce's American AI Exports program. At the same time, as Congress considers new export control proposals, it must ensure that these proposals give businesses the certainty and clarity needed to make the American AI stack the global default. Doing so will involve creating clear, implementable programs that provide certainty to both industry and the Department of Commerce's Bureau of Industry and Security (BIS).
* * *
8 Karen Massey and Mariana Fang, "IDC's Worldwide AI and Generative AI Spending-Industry Outlook," IDC, August 21, 2024, https://www.idc.com/resource-center/blog/idcs-worldwide-ai-and-generative-ai-spendingindustry-outlook/.
* * *
Congress created BIS to ensure export controls keep pace with technological advancements.
Congress should not attempt to play BIS's role by setting export control parameters in statute for years and attempting to anticipate technological developments that have yet to unfold.
Rather, Congress must give BIS clear direction, resources, staff, and other tools to accomplish its mission. If Congress sets the expectation that it will act as a regulator for specific technologies, it will undermine BIS and create significant uncertainty for U.S. companies working to ensure that the world runs on American AI.
Congress must remain mindful that the U.S. AI stack is facing increasing scrutiny across the world. We are seeing allies and partners expressing concerns-albeit unfounded ones-about the security, privacy, and reliability of U.S. technology. One example of a proposal that will drive increased scrutiny of the American AI stack is S. 1705/H.R. 3447, the Chip Security Act. The Chip Security Act would compel companies to track exported AI chips by statute. While there are valid commercial reasons for companies to pursue location verification in certain situations, a government tracking mandate would create the impression of deepening U.S. government control over the American AI stack. These fears about American technology are already very real. For example, the EU is considering a proposal driven by fears of a U.S. "kill switch" to assess whether it is too reliant on foreign cloud services./9
Congress must also ensure it works with U.S. partners and allies to develop and pursue plurilateral export control regimes, as doing so increases the effectiveness of controls and reduces the risk of foreign producers' backfilling market share previously held by U.S. companies. A prime example of the need for plurilateral export controls is semiconductor manufacturing equipment (SME). SME underpins much, if not all, of the AI stack as well as other industries that rely on semiconductors. Today, U.S. companies are global leaders in SME but face significant competitive threats from companies in allied jurisdictions with greater commercial opportunities due to more permissive export controls and Chinese firms receiving significant government support./10
* * *
9 Mathieu Pollet, "4 Ways Europe Wants to Wean Off U.S. Tech," POLITICO Pro, May 30, 2026, https://subscriber.politicopro.com/article/2026/05/4-ways-europe-wants-to-wean-off-us-tech-00941058.
10 Jacob Feldgoise and Hanna Dohmen, "Inside Beijing's Chipmaking Offensive: Where Is China Gaining Ground?," Center for Security and Emerging Technology, July 14, 2025, https://cset.georgetown.edu/article/inside-beijingschipmaking-offensive/.
* * *
The disparity in U.S. and allied export control regimes enables companies in allied jurisdictions to sell SME, such as equipment used for deposition, modification, inspection, and deep ultraviolet (DUV) lithography, to U.S. adversaries, even firms on the Entity List. Adversaries then use equipment purchased from U.S. allies to obtain the technology and know-how needed to improve their domestic SME and advanced semiconductor production. For example, China is using allied DUV equipment to enable both domestic production of AI chips and efforts to build cutting-edge extreme ultraviolet (EUV) lithography machines./11
China's use of legally imported, allied-produced SME, including DUV lithography machines and deposition, modification, and inspection equipment, to enable AI chip production and advances in domestic EUV lithography is a prime example of the limited effectiveness of U.S. unilateral controls. The U.S. must work with partners and allies to advance aligned plurilateral SME export controls. We look forward to working with Congress and the administration to pursue SME controls with allies and ensure that allied SME companies are not undermining U.S. economic security.
III. Policy Recommendations
a. Give BIS the Resources to Secure America's Lead
In recent years, BIS's workload has grown considerably, as more export controls have been imposed on technology, and the volume of license applications has increased significantly. The threat of licensing processing delays-whether actual or perceived-can call into question the reliability of U.S. companies and cause foreign customers to reconsider the merits of the U.S. AI stack. To support U.S. economic security, Congress should focus on enabling BIS with resources to accomplish its mission of advancing national security through technology leadership and export controls.
ITI supports several bills aimed at strengthening BIS that have advanced in the House Foreign Affairs Committee and do not yet have Senate companions: H.R. 7003, the BIS STRENGTH Act; H.R. 8284, the Bureau of Industry and Security License Administration Enhancement Act; H.R. 8289, the BIS Licensing Efficiency Act; H.R. 8288, the Strengthening Export Controls Compliance Act; and H.R. 4505, the Export Controls Enforcement Act./12
ITI also appreciates the $44 million increase in BIS's budget that Congress enacted last year, and we strongly support the $215 million budget increase for BIS included in the fiscal year 2027 House Commerce, Justice, Science, and Related Agencies appropriations bill and the president's budget request./13
This 90% budget increase will help ensure that BIS administers and enforces existing U.S. export controls effectively.
* * *
11 Fanny Potkin, "How China Built Its 'Manhattan Project' to Rival the West in AI Chips," Reuters, December 17, 2025, https://www.reuters.com/world/china/how-china-built-its-manhattan-project-rival-west-ai-chips-2025-1217/.
12 Mike Flynn, "Congress Needs to Get Export Controls Right," Information Technology Industry Council, April 20, 2026, https://www.itic.org/news-events/techwonk-blog/congress-needs-to-get-export-controls-right.
* * *
b. Drive Investment in the American AI Stack
To ensure that the U.S. continues to lead in AI, Congress should take the following steps to drive investment in the American AI Stack:
i. Extend and expand the Advanced Manufacturing Investment Credit (AMIC). The AMIC, a powerful incentive for investments in semiconductor production, expires this year. To enhance the AMIC's effectiveness, Congress should extend the December 31, 2026, deadline for qualified investments in facilities that manufacture semiconductors or SME.
Because it takes many years and tens of billions of dollars to build a fab, companies need certainty as they contemplate and plan such large investments. Extending the commence construction deadline would recognize the time-consuming and costly challenges companies can face-including review and permitting processes and workforce shortages-when making large-scale investments a reality. Congress should also expand the AMIC to include research, design, and manufacturing of essential semiconductor materials.
ii. Strengthen U.S. trade opportunities. ITI encourages the U.S. government to focus on reducing foreign tariff and non-tariff barriers and seeking commitments from foreign partners that enable digital trade and the deployment of AI and other technologies across all sectors of the economy. ITI also calls on the administration and Congress to support renewal of the United States-Mexico-Canada Agreement (USMCA). USMCA's high-standard trade commitments on areas such as digital trade remain the gold standard and are crucial to the tech industry's competitiveness and to promoting efforts like the American AI Exports Program.
c. Invest in an AI-Ready Workforce
Every technological revolution throughout history has driven change in the workforce, and AI is no different. While AI is changing the workforce, current evidence indicates it is doing so by shifting and creating jobs, not replacing them./14
For example, ITI members are investing billions of dollars in semiconductor manufacturing facilities to build the chips powering the AI revolution. These facilities create tens of thousands of jobs during construction, and once a semiconductor manufacturing facility comes online, companies often face challenges securing enough workers to operate it./15
Moving up the AI stack, the data center industry is supporting 5.5 million jobs across the U.S./16
These are just a few examples of how the American AI industry is creating new jobs and opportunities across the country.
* * *
13 Nathan James, "Overview of FY2026 Appropriations for Commerce, Justice, Science, and Related Agencies (CJS)," Congressional Research Service, https://www.congress.gov/crs-product/R48643; House Committee on Appropriations, "Committee Approves FY27 Commerce, Justice, Science, and Related Agencies Appropriations Act," May 13, 2026, https://appropriations.house.gov/news/press-releases/committee-approves-fy27-commercejustice-science-and-related-agencies; U.S. Department of Commerce, "FY 2027 President's Budget Request," https://www.commerce.gov/sites/default/files/2026-04/FY2027-Presidents-Budget-Request.pdf.
* * *
To better understand how AI is impacting the workforce, Congress should pass S. 3339, the AI Workforce PREPARE Act, S. 4476, the Workforce Transparency Act, and S. 4046/H.R. 8345, the Economy of the Future Commission Act. These proposals will give Congress and other policymakers the tools they need to better understand how AI is impacting the workforce and provide recommendations to prepare the workforce for AI.
A skilled and AI-ready workforce is a key component of America's continued AI leadership and economic security. Both government and industry can meet the moment by investing in AI education, resources, and research partnerships. ITI member companies are making major investments in workforce development. For example, TSMC is investing over $5 million in an apprenticeship program, while ITI members OpenAI, Google, Intel, and Microsoft are partnering with community colleges on workforce development./17
To match these investments, federal funding for science, technology, engineering, and mathematics (STEM) and computer science education is vital. Investing in STEM education will help build the talent pipeline to meet the growing demand for engineers and scientists driving American AI leadership.
d. Ensure American AI Is Not Compute-Limited
Compute-data centers with chips, power, and connectivity-is foundational to American AI leadership. Without sufficient compute, the U.S. AI stack could fall behind our adversaries.
* * *
14 Dan Shapero, "AI Has Already Added 1.3 Million New Jobs, According to LinkedIn Data," World Economic Forum, January 15, 2026, https://www.weforum.org/stories/2026/01/ai-has-already-added-1-3-million-new-jobsaccording-to-linkedin-data/.
15 TSMC, "TSMC Intends to Expand Its Investment in the United States to US$165 Billion to Power the Future of AI," March 4, 2025, https://pr.tsmc.com/english/news/3210.
16 Data Center Coalition, "2026 Impact Study," Center of Your Digital World, 2026, https://www.centerofyourdigitalworld.org/2026-impact-study.
17 "Governor Katie Hobbs, Joined by TSMC Arizona and Mayor Kate Gallego, Announce New Semiconductor Career Pathways with Expansion of Registered Technician Apprenticeship Program," Office of the Arizona Governor, November 19, 2024, https://azgovernor.gov/office-arizona-governor/news/2024/11/governor-katie-hobbs-joinedtsmc-arizona-and-mayor-kate-gallego; Shalin Jyotishi, "OpenAI, Google, Microsoft, Intel Bet On Community Colleges For AI Talent," Forbes, February 25, 2026, https://www.forbes.com/sites/shalinjyotishi/2026/02/25/openai-google-microsoft-intel-bet-on-communitycolleges-for-ai-talent/.
* * *
While semiconductors are critical to AI compute, they cannot train a cutting-edge AI model without power, cooling, networking, and a data center around them.
Today, the U.S. is the global leader in compute, and over the next four years, 40% of the $7 trillion that will be invested in data centers globally will be in the U.S./18
However, to ensure these investments become reality, policymakers must ensure the data center buildout is not delayed due to a lack of power or other supporting infrastructure. Our adversaries will not face similar delays. For example, in 2024, China added 429 gigawatts (GW) of new power capacity, more than one-third of the entire U.S. grid, and more than half of all global electricity growth./19
Keeping up with China and meeting America's growing electricity demand will require Congress to invest in expanding and modernizing the nation's grid. Investment in new-generation resources, transmission, and grid-enhancing technologies is essential to support the AI buildout and economic growth, while strengthening grid reliability for everyday users and helping catch up with decades of underinvestment in the power grid./20
To invest in the grid and help ensure America can power the next decade of economic growth while supporting more efficient data centers, Congress should pass H.R. 6633, the High-Capacity Grid Act; S. 3269, the Liquid Cooling for AI Act; and S. 3947, the REWIRE Act.
To ensure that investments in the grid are built on time, permitting reform is urgently needed.
Without a more efficient and predictable permitting process, even shovel-ready projects can face years of delay, increasing costs for consumers and businesses while undermining grid reliability. Congress must pass a comprehensive permitting reform proposal to ensure that America's permitting process does not hinder our ability to build the infrastructure we need to win the AI race.
e. Build Secure and Resilient Critical Minerals Supply Chains
Every part of the AI stack-from semiconductors to servers and SME-relies on critical minerals.
Some of these critical minerals are only produced and refined by a few countries, posing a major risk to America's ongoing AI build-out. ITI calls on Congress to work with the administration to promote reliable and secure access to critical minerals through international cooperation and increased U.S. processing and production.
* * *
18 Jesse Noffsinger, Mark Patel, Pankaj Sachdeva, Arjita Bhan, Haley Chang, and Maria Goodpaster, "The Cost of Compute: A $7 Trillion Race to Scale Data Centers," McKinsey & Company, April 28, 2025, https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/the-cost-ofcompute-a-7-trillion-dollar-race-to-scale-data-centers.
19 Christopher Lehane, "OpenAI response to Michael Kratsios, Office of Science and Technology Policy", October 27, 2025, https://cdn.openai.com/pdf/21b88bb5-10a3-4566-919d-f9a6b9c3e632/openai-ostp-rfi-oct-27-2025.pdf.
20 Robinson Meyer, "It's the Age of Electricity and America Isn't Ready," The New York Times, April 27, 2026, https://www.nytimes.com/interactive/2026/04/27/opinion/electricity-power-grid-infrastructure.html.
* * *
Proposals like H.R. 7037, the DOMINANCE Act, which recently advanced out of the House Foreign Affairs Committee unanimously and has a partial Senate companion, S. 4392, the Energy Security Pacts Act, will help the U.S. build resilient critical minerals supply chains with trusted allies and partners. ITI also supports H.R. 8780, the Critical Mineral and Extraction Tax Parity Act, which will further expand and scale U.S. critical minerals production capacity.
ITI strongly supports other U.S. government efforts to promote reliable and secure access to the critical minerals used across the AI stack and other critical sectors. Efforts of note include the Office of the U.S. Trade Representative's (USTR) bilateral action plans and work toward plurilateral agreements and the U.S. Department of State's Pax Silica initiative.
* * *
Original text here: https://www.banking.senate.gov/imo/media/doc/flynn_testimony_6-11-26.pdf
* * *
Chairman Scott, Ranking Member Warren, and members of the Committee, thank you for the opportunity to testify before you today. My name is Mike Flynn, and I am a Senior Vice President and Counsel at the Information Technology Industry Council (ITI).
ITI ... Show Full Article WASHINGTON, June 26 -- The Senate Banking, Housing and Urban Affairs Committee released the following written testimony by Mike Flynn, senior vice president and counsel at the Information Technology Industry Council, from a June 11, 2026, hearing entitled "AI and the American Dream: Promoting Innovation, Affordability, and American Dominance": * * * Chairman Scott, Ranking Member Warren, and members of the Committee, thank you for the opportunity to testify before you today. My name is Mike Flynn, and I am a Senior Vice President and Counsel at the Information Technology Industry Council (ITI). ITIrepresents 80 companies driving the nation's artificial intelligence (AI) development, adoption, and leadership. Our member companies operate at every layer of the AI technology stack: we represent chipmakers and designers, leading frontier AI labs, cloud service providers, and developers building the applications that deliver AI to customers worldwide.
Today, our members are investing hundreds of billions of dollars in the United States to do the frontier AI research, semiconductor fabrication and design, AI model training, and application development that make the American AI stack the global leader.
My testimony will provide an overview of the interconnected and interdependent American AI stack, including how AI is transforming cybersecurity, and examine what is needed to ensure America wins the global AI race. Finally, I will provide recommendations on how Congress can continue working to advance American AI leadership and promote U.S. economic security.
I. The American Artificial Intelligence Stack
AI systems are some of the most complex and technically sophisticated products ever made.
Frontier AI models rely on the work of thousands of highly skilled engineers, advanced semiconductors with features many thousands of times smaller than the width of a human hair, and networking that operates at the speed of light, to name only a few of the vital inputs needed to build and run a frontier AI model./1
These AI models, in turn, are used by billions of people across the world for everything from drafting emails and writing code to cutting-edge scientific research./2
Modern AI models are built by learning patterns from large amounts of text, images, code, or other data. When you ask the model a question, the model predicts the most likely useful response based on learned patterns. By integrating more data, better system architectures, and more compute, AI model developers have been able to rapidly improve AI's capabilities.
1 Gustav Grundin et al., "Issue Brief: AI Infrastructure," McKinsey & Company, February 27, 2026, https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/issue-brief-aiinfrastructure.
2 Microsoft AI Economy Institute, "Global AI Adoption in 2025: A Widening Digital Divide," Microsoft, January 8, 2026, https://www.microsoft.com/en-us/corporate-responsibility/topics/AI-Economy-Institute/reports/Global-AIAdoption-2025/.
* * *
a. The Interconnected and Interdependent AI Stack
When the average American thinks about AI, they often think first about the tools they can see and use. For example, they may think of the chatbot they can use to write a poem or a software assistant built into an email application. While these are the types of AI that most people interact with directly, they are only one kind of AI. AI systems are far more than chatbots, and they do everything from blocking fraudulent credit card transactions to improving the speed and efficiency of manufacturing processes./3
Critically, AI systems do not exist in a void; rather, they are part of the deeply interconnected and interdependent AI technology stack. This AI technology stack consists of upstream and downstream infrastructure and technologies that enable AI and is an interconnected set of five layers: infrastructure, data, model development and operations, applications, and governance, security, and trust./4
Overly broad measures aimed at one layer of the U.S. AI tech stack can have unintended consequences for other parts of the stack.
When considering AI policy in the context of the AI stack, it becomes clear that AI policy is about far more than semiconductors and models, two of the most scrutinized parts of the stack. Instead, AI policy should take into consideration everything from applications and data to critical minerals and semiconductor manufacturing equipment. Each layer of the AI stack raises different questions for policymakers, and effective policymaking requires treating the U.S. AI stack as an interconnected system, with the recognition that regulation in one area will have ripple effects across other layers of the stack.
For example, a regulation on AI models will affect how cloud companies allocate computing resources to customers, plan future data centers to meet demand, and, as a result, impact chip and equipment market dynamics. This means that to promote innovation, affordability, and American AI dominance, Congress must advance proposals for everything from export controls to AI model regulation that embrace the complex reality of the AI stack.
b. AI Innovation & Cybersecurity
American frontier developers have continued to invest in building the best AI models, drastically improving reasoning and context awareness to enable longer tasks and conversations. Due to these advances, frontier AI models have gained sophisticated capabilities-especially in software coding-and have driven a major shift in the cybersecurity ecosystem. The latest models can now identify and remediate software vulnerabilities at an extraordinary scale and speed.
* * *
3 Sheila Chiang, "AI and Machine Learning Helped Visa Combat $40 Billion in Fraud Activity," CNBC, July 26, 2024, https://www.cnbc.com/amp/2024/07/26/ai-and-machine-learning-helped-visa-combat-40-billion-in-fraudactivity.html.
4 Infrastructure includes the semiconductors, data centers, and energy sources necessary to build and run AI. Data includes raw inputs used in AI training. Model development and operations includes tools, platforms, and resources needed to serve AI models at scale. Applications are the part of the AI technology stack that most people interact with during daily life. Governance, security, and trust is a cross-cutting function that permeates every layer of the AI technology stack-from infrastructure to applications. Read more on the AI stack here: https://www.itic.org/documents/artificial-intelligence/ITI_AITechnologyStack.pdf.
* * *
In response to this critical moment for the cybersecurity community, ITI member companies have launched programs to collaborate with key organizations-critical infrastructure firms, essential software developers and providers, and cybersecurity companies-to enable them to use advanced models to get ahead of malicious cyber actors./5
The Trump Administration took a constructive, pro-innovation step to address AI and cybersecurity with its Executive Order, "Promoting Advanced Artificial Intelligence Innovation and Security."/6
This framework for collaboration among AI developers, the federal government, and trusted partners addresses the immediate need for cross-sector coordination on the cybersecurity implications of frontier models. To build on the direction of the Executive Order, Congress should codify and increase funding for the Department of Commerce's Center for AI Standards and Innovation (CAISI), which is doing critical work to assess frontier model capabilities and develop new standards around AI security.
c. Information Sharing and AI Model Distillation
Underpinning the Executive Order is the sharing of cybersecurity information between industry and government, and within industry. The Cybersecurity Information Sharing Act of 2015 (CISA 15) facilitates that sharing. CISA 15 is a foundational element of the U.S. cybersecurity ecosystem, providing critical liability protections to reduce legal risk for sharing information on cyber threat indicators, software vulnerabilities, and defensive measures. Currently, CISA 15 is temporarily reauthorized until September 30, 2026, and Congress must pass a long-term reauthorization to ensure that these essential information-sharing authorities do not lapse at a critical moment for the nation's cybersecurity.
As AI models develop capabilities directly relevant to national security, it is deeply concerning that U.S. adversaries are attempting to distill American frontier closed-source AI models, effectively enabling them to steal the work of U.S. AI labs. Countering distillation campaigns by adversaries will require improved information sharing between industry and government and
* * *
5 Anthropic, "Project Glasswing: Securing Critical Software for the AI Era," 2026, https://www.anthropic.com/glasswing; OpenAI, "Trusted Access for the Next Era of Cyber Defense," 2026, https://openai.com/index/scaling-trusted-access-for-cyber-defense/.
6 The White House, "Promoting Advanced Artificial Intelligence Innovation and Security," June 2, 2026, https://www.whitehouse.gov/presidential-actions/2026/06/promoting-advanced-artificial-intelligence-innovationand-security/.
* * *
within industry. Congress should act to address AI model distillation. Doing so will require consultation with industry to ensure proposals addressing distillation are scoped carefully to avoid unintended impacts on the U.S. AI stack, while applying lessons learned from the cybersecurity space to counter adversarial distillation.
II. Winning the Global AI Competition
a. Global Supply Chains and Markets Fuel American AI Leadership
American AI leadership depends on the U.S. AI stack's ability to participate in complex global supply chains and compete in markets around the world. To support continued U.S. AI leadership, policymakers should work with allies and partners to secure and diversify the supply chains that contribute to the U.S. AI stack. Such efforts could include convening formal supply chain reviews with allies and building upon existing efforts to ensure that market access barriers do not present impediments to global supply chains.
Just as the U.S. AI stack draws on global supply chains to build cutting-edge products, the U.S. AI tech stack must access and compete for customers in the global marketplace. Market access has been a primary driver of U.S. industry success and innovation, as most demand for many U.S. products is outside the United States. The U.S. government should strengthen existing bilateral and plurilateral trade agreements and negotiate new agreements to promote U.S. goods and services in markets abroad. Upholding the World Trade Organization's (WTO) Moratorium on Customs Duties on Electronic Transmissions will also be critical to supporting AI innovation, economic growth, and supply chain resiliency.
Sales abroad help U.S. companies invest in research, development, and manufacturing in the United States, spurring job growth and ensuring that America is first to build next-generation technologies. The numbers demonstrate the importance of overseas sales for U.S. semiconductor producers in particular: "in 2023, sales outside the U.S. comprised about threequarters of total U.S. semiconductor industry sales," and significant revenue from those sales is invested back into the United States./7
To foster U.S. sales abroad, ITI encourages the U.S. government to focus on reducing foreign tariff and non-tariff barriers, eliminating duplicative certification and conformity assessment requirements by foreign governments in areas where international certifications exist, promoting strong protection of intellectual property rights in foreign markets, and seeking commitments from foreign partners that enable digital trade and the deployment of AI across all sectors of the economy.
* * *
7 "State of the U.S. Semiconductor Industry 2024," Semiconductor Industry Association, https://www.semiconductors.org/2024-state-of-the-u-s-semiconductor-industry/.
* * *
U.S. companies face fierce competition in the global marketplace and seek to be partners of choice based on quality, dependability, quantity, cost, timeliness, and other factors. To successfully engage customers around the world, companies need predictability and certainty.
Permanent trade agreements and frameworks, as opposed to potentially ever-changing tariffs, help to create market stability and encourage customers and counterparts in Asia and Europe to be open and productive partners with U.S. companies, helping strengthen supply chains at home and abroad.
To ensure U.S. economic security, U.S. companies must remain partners of choice in the global marketplace. Policymakers should not take U.S. AI leadership for granted, especially as more governments take meaningful steps to foster their domestic AI ecosystems. That is why ITI supports the American AI Exports Program, a whole-of-government approach to facilitating additional U.S. exports to the $235 billion global AI market./8
By promoting the export of American AI technology, the program is inherently aimed at decreasing international adoption of AI technologies developed by countries like China. To support this critical work, Congress should codify the American AI Exports Program by passing H.R. 6996, the Full AI Stack Export Promotion Act.
b. U.S. Economic Security and AI
Maintaining and expanding America's AI lead and ensuring the U.S. stack is adopted across the world is critical to U.S. economic and national security. Congress has done important bipartisan work over the last several years to support American AI leadership, notably passing the CHIPS and Science Act, the Comprehensive Outbound Investment National Security (COINS) Act, and restoring the immediate expensing of domestic research and development costs.
U.S. companies need clarity and certainty to sign contracts with customers, spend billions of dollars on custom chips, and continue the trillion-dollar investments in America that are making the U.S. the undisputed AI leader. This is why ITI has consistently emphasized the importance of stakeholder engagement in developing and finalizing proposals at the intersection of national security, trade, and technology. Absent meaningful engagement with industry and allies, broadly scoped and unilateral measures risk limiting both market opportunities for U.S. companies and the effectiveness of export controls.
As it considers export control proposals, Congress should seek to both promote the American AI stack and protect U.S. economic security. This is not a binary choice; Congress can and should optimize for both by codifying and supporting initiatives like the Department of Commerce's American AI Exports program. At the same time, as Congress considers new export control proposals, it must ensure that these proposals give businesses the certainty and clarity needed to make the American AI stack the global default. Doing so will involve creating clear, implementable programs that provide certainty to both industry and the Department of Commerce's Bureau of Industry and Security (BIS).
* * *
8 Karen Massey and Mariana Fang, "IDC's Worldwide AI and Generative AI Spending-Industry Outlook," IDC, August 21, 2024, https://www.idc.com/resource-center/blog/idcs-worldwide-ai-and-generative-ai-spendingindustry-outlook/.
* * *
Congress created BIS to ensure export controls keep pace with technological advancements.
Congress should not attempt to play BIS's role by setting export control parameters in statute for years and attempting to anticipate technological developments that have yet to unfold.
Rather, Congress must give BIS clear direction, resources, staff, and other tools to accomplish its mission. If Congress sets the expectation that it will act as a regulator for specific technologies, it will undermine BIS and create significant uncertainty for U.S. companies working to ensure that the world runs on American AI.
Congress must remain mindful that the U.S. AI stack is facing increasing scrutiny across the world. We are seeing allies and partners expressing concerns-albeit unfounded ones-about the security, privacy, and reliability of U.S. technology. One example of a proposal that will drive increased scrutiny of the American AI stack is S. 1705/H.R. 3447, the Chip Security Act. The Chip Security Act would compel companies to track exported AI chips by statute. While there are valid commercial reasons for companies to pursue location verification in certain situations, a government tracking mandate would create the impression of deepening U.S. government control over the American AI stack. These fears about American technology are already very real. For example, the EU is considering a proposal driven by fears of a U.S. "kill switch" to assess whether it is too reliant on foreign cloud services./9
Congress must also ensure it works with U.S. partners and allies to develop and pursue plurilateral export control regimes, as doing so increases the effectiveness of controls and reduces the risk of foreign producers' backfilling market share previously held by U.S. companies. A prime example of the need for plurilateral export controls is semiconductor manufacturing equipment (SME). SME underpins much, if not all, of the AI stack as well as other industries that rely on semiconductors. Today, U.S. companies are global leaders in SME but face significant competitive threats from companies in allied jurisdictions with greater commercial opportunities due to more permissive export controls and Chinese firms receiving significant government support./10
* * *
9 Mathieu Pollet, "4 Ways Europe Wants to Wean Off U.S. Tech," POLITICO Pro, May 30, 2026, https://subscriber.politicopro.com/article/2026/05/4-ways-europe-wants-to-wean-off-us-tech-00941058.
10 Jacob Feldgoise and Hanna Dohmen, "Inside Beijing's Chipmaking Offensive: Where Is China Gaining Ground?," Center for Security and Emerging Technology, July 14, 2025, https://cset.georgetown.edu/article/inside-beijingschipmaking-offensive/.
* * *
The disparity in U.S. and allied export control regimes enables companies in allied jurisdictions to sell SME, such as equipment used for deposition, modification, inspection, and deep ultraviolet (DUV) lithography, to U.S. adversaries, even firms on the Entity List. Adversaries then use equipment purchased from U.S. allies to obtain the technology and know-how needed to improve their domestic SME and advanced semiconductor production. For example, China is using allied DUV equipment to enable both domestic production of AI chips and efforts to build cutting-edge extreme ultraviolet (EUV) lithography machines./11
China's use of legally imported, allied-produced SME, including DUV lithography machines and deposition, modification, and inspection equipment, to enable AI chip production and advances in domestic EUV lithography is a prime example of the limited effectiveness of U.S. unilateral controls. The U.S. must work with partners and allies to advance aligned plurilateral SME export controls. We look forward to working with Congress and the administration to pursue SME controls with allies and ensure that allied SME companies are not undermining U.S. economic security.
III. Policy Recommendations
a. Give BIS the Resources to Secure America's Lead
In recent years, BIS's workload has grown considerably, as more export controls have been imposed on technology, and the volume of license applications has increased significantly. The threat of licensing processing delays-whether actual or perceived-can call into question the reliability of U.S. companies and cause foreign customers to reconsider the merits of the U.S. AI stack. To support U.S. economic security, Congress should focus on enabling BIS with resources to accomplish its mission of advancing national security through technology leadership and export controls.
ITI supports several bills aimed at strengthening BIS that have advanced in the House Foreign Affairs Committee and do not yet have Senate companions: H.R. 7003, the BIS STRENGTH Act; H.R. 8284, the Bureau of Industry and Security License Administration Enhancement Act; H.R. 8289, the BIS Licensing Efficiency Act; H.R. 8288, the Strengthening Export Controls Compliance Act; and H.R. 4505, the Export Controls Enforcement Act./12
ITI also appreciates the $44 million increase in BIS's budget that Congress enacted last year, and we strongly support the $215 million budget increase for BIS included in the fiscal year 2027 House Commerce, Justice, Science, and Related Agencies appropriations bill and the president's budget request./13
This 90% budget increase will help ensure that BIS administers and enforces existing U.S. export controls effectively.
* * *
11 Fanny Potkin, "How China Built Its 'Manhattan Project' to Rival the West in AI Chips," Reuters, December 17, 2025, https://www.reuters.com/world/china/how-china-built-its-manhattan-project-rival-west-ai-chips-2025-1217/.
12 Mike Flynn, "Congress Needs to Get Export Controls Right," Information Technology Industry Council, April 20, 2026, https://www.itic.org/news-events/techwonk-blog/congress-needs-to-get-export-controls-right.
* * *
b. Drive Investment in the American AI Stack
To ensure that the U.S. continues to lead in AI, Congress should take the following steps to drive investment in the American AI Stack:
i. Extend and expand the Advanced Manufacturing Investment Credit (AMIC). The AMIC, a powerful incentive for investments in semiconductor production, expires this year. To enhance the AMIC's effectiveness, Congress should extend the December 31, 2026, deadline for qualified investments in facilities that manufacture semiconductors or SME.
Because it takes many years and tens of billions of dollars to build a fab, companies need certainty as they contemplate and plan such large investments. Extending the commence construction deadline would recognize the time-consuming and costly challenges companies can face-including review and permitting processes and workforce shortages-when making large-scale investments a reality. Congress should also expand the AMIC to include research, design, and manufacturing of essential semiconductor materials.
ii. Strengthen U.S. trade opportunities. ITI encourages the U.S. government to focus on reducing foreign tariff and non-tariff barriers and seeking commitments from foreign partners that enable digital trade and the deployment of AI and other technologies across all sectors of the economy. ITI also calls on the administration and Congress to support renewal of the United States-Mexico-Canada Agreement (USMCA). USMCA's high-standard trade commitments on areas such as digital trade remain the gold standard and are crucial to the tech industry's competitiveness and to promoting efforts like the American AI Exports Program.
c. Invest in an AI-Ready Workforce
Every technological revolution throughout history has driven change in the workforce, and AI is no different. While AI is changing the workforce, current evidence indicates it is doing so by shifting and creating jobs, not replacing them./14
For example, ITI members are investing billions of dollars in semiconductor manufacturing facilities to build the chips powering the AI revolution. These facilities create tens of thousands of jobs during construction, and once a semiconductor manufacturing facility comes online, companies often face challenges securing enough workers to operate it./15
Moving up the AI stack, the data center industry is supporting 5.5 million jobs across the U.S./16
These are just a few examples of how the American AI industry is creating new jobs and opportunities across the country.
* * *
13 Nathan James, "Overview of FY2026 Appropriations for Commerce, Justice, Science, and Related Agencies (CJS)," Congressional Research Service, https://www.congress.gov/crs-product/R48643; House Committee on Appropriations, "Committee Approves FY27 Commerce, Justice, Science, and Related Agencies Appropriations Act," May 13, 2026, https://appropriations.house.gov/news/press-releases/committee-approves-fy27-commercejustice-science-and-related-agencies; U.S. Department of Commerce, "FY 2027 President's Budget Request," https://www.commerce.gov/sites/default/files/2026-04/FY2027-Presidents-Budget-Request.pdf.
* * *
To better understand how AI is impacting the workforce, Congress should pass S. 3339, the AI Workforce PREPARE Act, S. 4476, the Workforce Transparency Act, and S. 4046/H.R. 8345, the Economy of the Future Commission Act. These proposals will give Congress and other policymakers the tools they need to better understand how AI is impacting the workforce and provide recommendations to prepare the workforce for AI.
A skilled and AI-ready workforce is a key component of America's continued AI leadership and economic security. Both government and industry can meet the moment by investing in AI education, resources, and research partnerships. ITI member companies are making major investments in workforce development. For example, TSMC is investing over $5 million in an apprenticeship program, while ITI members OpenAI, Google, Intel, and Microsoft are partnering with community colleges on workforce development./17
To match these investments, federal funding for science, technology, engineering, and mathematics (STEM) and computer science education is vital. Investing in STEM education will help build the talent pipeline to meet the growing demand for engineers and scientists driving American AI leadership.
d. Ensure American AI Is Not Compute-Limited
Compute-data centers with chips, power, and connectivity-is foundational to American AI leadership. Without sufficient compute, the U.S. AI stack could fall behind our adversaries.
* * *
14 Dan Shapero, "AI Has Already Added 1.3 Million New Jobs, According to LinkedIn Data," World Economic Forum, January 15, 2026, https://www.weforum.org/stories/2026/01/ai-has-already-added-1-3-million-new-jobsaccording-to-linkedin-data/.
15 TSMC, "TSMC Intends to Expand Its Investment in the United States to US$165 Billion to Power the Future of AI," March 4, 2025, https://pr.tsmc.com/english/news/3210.
16 Data Center Coalition, "2026 Impact Study," Center of Your Digital World, 2026, https://www.centerofyourdigitalworld.org/2026-impact-study.
17 "Governor Katie Hobbs, Joined by TSMC Arizona and Mayor Kate Gallego, Announce New Semiconductor Career Pathways with Expansion of Registered Technician Apprenticeship Program," Office of the Arizona Governor, November 19, 2024, https://azgovernor.gov/office-arizona-governor/news/2024/11/governor-katie-hobbs-joinedtsmc-arizona-and-mayor-kate-gallego; Shalin Jyotishi, "OpenAI, Google, Microsoft, Intel Bet On Community Colleges For AI Talent," Forbes, February 25, 2026, https://www.forbes.com/sites/shalinjyotishi/2026/02/25/openai-google-microsoft-intel-bet-on-communitycolleges-for-ai-talent/.
* * *
While semiconductors are critical to AI compute, they cannot train a cutting-edge AI model without power, cooling, networking, and a data center around them.
Today, the U.S. is the global leader in compute, and over the next four years, 40% of the $7 trillion that will be invested in data centers globally will be in the U.S./18
However, to ensure these investments become reality, policymakers must ensure the data center buildout is not delayed due to a lack of power or other supporting infrastructure. Our adversaries will not face similar delays. For example, in 2024, China added 429 gigawatts (GW) of new power capacity, more than one-third of the entire U.S. grid, and more than half of all global electricity growth./19
Keeping up with China and meeting America's growing electricity demand will require Congress to invest in expanding and modernizing the nation's grid. Investment in new-generation resources, transmission, and grid-enhancing technologies is essential to support the AI buildout and economic growth, while strengthening grid reliability for everyday users and helping catch up with decades of underinvestment in the power grid./20
To invest in the grid and help ensure America can power the next decade of economic growth while supporting more efficient data centers, Congress should pass H.R. 6633, the High-Capacity Grid Act; S. 3269, the Liquid Cooling for AI Act; and S. 3947, the REWIRE Act.
To ensure that investments in the grid are built on time, permitting reform is urgently needed.
Without a more efficient and predictable permitting process, even shovel-ready projects can face years of delay, increasing costs for consumers and businesses while undermining grid reliability. Congress must pass a comprehensive permitting reform proposal to ensure that America's permitting process does not hinder our ability to build the infrastructure we need to win the AI race.
e. Build Secure and Resilient Critical Minerals Supply Chains
Every part of the AI stack-from semiconductors to servers and SME-relies on critical minerals.
Some of these critical minerals are only produced and refined by a few countries, posing a major risk to America's ongoing AI build-out. ITI calls on Congress to work with the administration to promote reliable and secure access to critical minerals through international cooperation and increased U.S. processing and production.
* * *
18 Jesse Noffsinger, Mark Patel, Pankaj Sachdeva, Arjita Bhan, Haley Chang, and Maria Goodpaster, "The Cost of Compute: A $7 Trillion Race to Scale Data Centers," McKinsey & Company, April 28, 2025, https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/the-cost-ofcompute-a-7-trillion-dollar-race-to-scale-data-centers.
19 Christopher Lehane, "OpenAI response to Michael Kratsios, Office of Science and Technology Policy", October 27, 2025, https://cdn.openai.com/pdf/21b88bb5-10a3-4566-919d-f9a6b9c3e632/openai-ostp-rfi-oct-27-2025.pdf.
20 Robinson Meyer, "It's the Age of Electricity and America Isn't Ready," The New York Times, April 27, 2026, https://www.nytimes.com/interactive/2026/04/27/opinion/electricity-power-grid-infrastructure.html.
* * *
Proposals like H.R. 7037, the DOMINANCE Act, which recently advanced out of the House Foreign Affairs Committee unanimously and has a partial Senate companion, S. 4392, the Energy Security Pacts Act, will help the U.S. build resilient critical minerals supply chains with trusted allies and partners. ITI also supports H.R. 8780, the Critical Mineral and Extraction Tax Parity Act, which will further expand and scale U.S. critical minerals production capacity.
ITI strongly supports other U.S. government efforts to promote reliable and secure access to the critical minerals used across the AI stack and other critical sectors. Efforts of note include the Office of the U.S. Trade Representative's (USTR) bilateral action plans and work toward plurilateral agreements and the U.S. Department of State's Pax Silica initiative.
* * *
Original text here: https://www.banking.senate.gov/imo/media/doc/flynn_testimony_6-11-26.pdf
House Transportation & Infrastructure Subcommittee Chairman Nehls Issues Statement at Hearing on National Airspace System
WASHINGTON, June 25 -- Rep. Troy Nehls, R-Texas, chairman of the House Transportation and Infrastructure Subcommittee on Aviation, released the following statement from a June 4, 2026, hearing entitled "Connecting Rural America to the National Airspace System":
* * *
Airports are more than transportation hubs, they are economic engines supporting Americans' livelihoods, connecting communities, and enhancing quality of life.
There are nearly 5,000 public-use airports and heliports across the United States and almost 3,300 of these facilities are on the FAA's National Plan of Integrated Airport ... Show Full Article WASHINGTON, June 25 -- Rep. Troy Nehls, R-Texas, chairman of the House Transportation and Infrastructure Subcommittee on Aviation, released the following statement from a June 4, 2026, hearing entitled "Connecting Rural America to the National Airspace System": * * * Airports are more than transportation hubs, they are economic engines supporting Americans' livelihoods, connecting communities, and enhancing quality of life. There are nearly 5,000 public-use airports and heliports across the United States and almost 3,300 of these facilities are on the FAA's National Plan of Integrated AirportSystems. These numbers are certain to grow as advanced air mobility technologies emerge and become integrated into the aviation system. Yet, of all these facilities, only 64 are large or medium-hub airports. Understandably, they get a lot of attention.
But there are another 450 airports providing critical commercial passenger service across the country. While they are well known within their communities, these airports deserve greater attention and stronger support from Congress. For the 46 million Americans living in rural communities, these facilities provide the same critical access to the National Airspace System (NAS) as larger airports do for urban Americans.
You all are here today to help us better highlight the value of these facilities, the role they play in their communities, and the challenges they face in maintaining and expanding access.
According to a recent study by the Airports Council International - North America, commercial service airports in the United States support over 12.8 million jobs, $618 billion in wages, and $1.8 trillion in economic output. Notably, every additional 1,000 enplanements generate nearly $700,000 in economic output and support six additional jobs. We should be working together to increase enplanements at these airports so all Americans can share these economic benefits.
Fortunately, there are strong federal programs designed to support communities maintaining existing service, expanding service, and attracting new carriers, ultimately increasing enplanements. Essential Air Service (EAS) is the most well-known of these programs. In the FAA Reauthorization Act of 2024, Congress strengthened and reformed EAS to improve accountability, restore fiscal discipline, and better target limited resources to the communities most in need.
The Small Community Air Service Development Program also gives communities the opportunity to identify their own air service needs, recruit carriers, offer revenue guarantees, conduct studies, and market new service. Importantly, this program encourages local business investment. This skin-in-the-game approach incentivizes services that are more likely to succeed long-term without subsidies.
Lastly, the Airport Improvement Program (AIP) distributes resources generated by user fees and fuel taxes by formula to all public-use airports in the National Plan of Integrated Airport Systems (NPIAS).
I would be remiss to not recognize the significant progress the Trump Administration has made in modernizing the system that safely manages America's air traffic.
Under Secretary Duffy's and Administrator Bedford's leadership, more than 2,000 air traffic controllers were hired last year, more than 12,000 applied to become controllers this year, 612 new radar systems are being deployed, half of the outdated copper wiring has already been replaced with modern fiber-optics, 17 control towers transitioned from paper flight strips to electronic systems, and 54 airports now have surface awareness systems. Clearly, the future of American aviation is bright.
* * *
Original text here: https://transportation.house.gov/calendar/eventsingle.aspx?EventID=410069
* * *
Airports are more than transportation hubs, they are economic engines supporting Americans' livelihoods, connecting communities, and enhancing quality of life.
There are nearly 5,000 public-use airports and heliports across the United States and almost 3,300 of these facilities are on the FAA's National Plan of Integrated Airport ... Show Full Article WASHINGTON, June 25 -- Rep. Troy Nehls, R-Texas, chairman of the House Transportation and Infrastructure Subcommittee on Aviation, released the following statement from a June 4, 2026, hearing entitled "Connecting Rural America to the National Airspace System": * * * Airports are more than transportation hubs, they are economic engines supporting Americans' livelihoods, connecting communities, and enhancing quality of life. There are nearly 5,000 public-use airports and heliports across the United States and almost 3,300 of these facilities are on the FAA's National Plan of Integrated AirportSystems. These numbers are certain to grow as advanced air mobility technologies emerge and become integrated into the aviation system. Yet, of all these facilities, only 64 are large or medium-hub airports. Understandably, they get a lot of attention.
But there are another 450 airports providing critical commercial passenger service across the country. While they are well known within their communities, these airports deserve greater attention and stronger support from Congress. For the 46 million Americans living in rural communities, these facilities provide the same critical access to the National Airspace System (NAS) as larger airports do for urban Americans.
You all are here today to help us better highlight the value of these facilities, the role they play in their communities, and the challenges they face in maintaining and expanding access.
According to a recent study by the Airports Council International - North America, commercial service airports in the United States support over 12.8 million jobs, $618 billion in wages, and $1.8 trillion in economic output. Notably, every additional 1,000 enplanements generate nearly $700,000 in economic output and support six additional jobs. We should be working together to increase enplanements at these airports so all Americans can share these economic benefits.
Fortunately, there are strong federal programs designed to support communities maintaining existing service, expanding service, and attracting new carriers, ultimately increasing enplanements. Essential Air Service (EAS) is the most well-known of these programs. In the FAA Reauthorization Act of 2024, Congress strengthened and reformed EAS to improve accountability, restore fiscal discipline, and better target limited resources to the communities most in need.
The Small Community Air Service Development Program also gives communities the opportunity to identify their own air service needs, recruit carriers, offer revenue guarantees, conduct studies, and market new service. Importantly, this program encourages local business investment. This skin-in-the-game approach incentivizes services that are more likely to succeed long-term without subsidies.
Lastly, the Airport Improvement Program (AIP) distributes resources generated by user fees and fuel taxes by formula to all public-use airports in the National Plan of Integrated Airport Systems (NPIAS).
I would be remiss to not recognize the significant progress the Trump Administration has made in modernizing the system that safely manages America's air traffic.
Under Secretary Duffy's and Administrator Bedford's leadership, more than 2,000 air traffic controllers were hired last year, more than 12,000 applied to become controllers this year, 612 new radar systems are being deployed, half of the outdated copper wiring has already been replaced with modern fiber-optics, 17 control towers transitioned from paper flight strips to electronic systems, and 54 airports now have surface awareness systems. Clearly, the future of American aviation is bright.
* * *
Original text here: https://transportation.house.gov/calendar/eventsingle.aspx?EventID=410069
Bureau of Labor Statistics Commissioner Nominee Matsumoto Testifies Before Senate Health, Education, Labor & Pensions Committee
WASHINGTON, June 24 -- The Senate Health, Education, Labor and Pensions Committee released the following testimony by Brett Matsumoto, President Trump's nominee to be commissioner of the U.S. Department of Labor Bureau of Labor Statistics, from a June 10, 2026, confirmation hearing:
* * *
Good morning.
Chairman Cassidy, Ranking Member Sanders, and Members of the Committee, it is an honor to appear before you as the nominee to be Commissioner of the Bureau of Labor Statistics. It is an honor to have been nominated by President Trump for this important position.
Before beginning, I would like ... Show Full Article WASHINGTON, June 24 -- The Senate Health, Education, Labor and Pensions Committee released the following testimony by Brett Matsumoto, President Trump's nominee to be commissioner of the U.S. Department of Labor Bureau of Labor Statistics, from a June 10, 2026, confirmation hearing: * * * Good morning. Chairman Cassidy, Ranking Member Sanders, and Members of the Committee, it is an honor to appear before you as the nominee to be Commissioner of the Bureau of Labor Statistics. It is an honor to have been nominated by President Trump for this important position. Before beginning, I would liketo thank all of the people who are responsible for getting me to where I am today: my parents and my older brothers, as well as other relatives, teachers, and mentors. I would also like to thank everyone who made positive statements regarding my nomination. When President Trump announced the intention to nominate me for this position, there was a lot of media interest. Although reading news stories about myself is something I'm still getting used to, I was humbled by the positive reactions of economists across the political spectrum.
I grew up in Delaware and attended college at the University of Delaware. After graduating with degrees in Philosophy, History, Economics, and a Masters in Economics, I went to the University of North Carolina for a PhD in Economics, where I specialized in Applied Microeconomics. After completing my PhD, I got a job at the Bureau of Labor Statistics.
I have worked for the Bureau of Labor Statistics since 2015 on economic measurement topics. I have published on topics relating to inflation and consumption measurement. I have also worked on projects to support the production offices, mainly the Consumer Price Index program. These include work on alternative data projects, most notably a project to replace traditionally collected data with insurance claims data for the medical price indexes. In September 2024, I was promoted to a supervisory position, where I oversee a small team of economists.
In the first Trump Administration, I did a detail at the Council of Economic Advisors (CEA) where I worked on labor and health topics. In early 2025, I was asked to do another detail by the incoming chairman of the CEA, and I have been in that role since. While working at the CEA, I have gained a great appreciation for the official economic statistics from the perspective of a data user. Much of the analysis done at the CEA would be impossible without the official data and having timely and accurate data is extremely important in the process of developing good economic policies.
I believe strongly in the mission of the BLS to produce "objective, quality measures and analyses of labor market activity, working conditions, price changes, and productivity in the U.S. Economy to support public and private decision making". To achieve this mission, it is important for the public to be confident that decisions at the BLS are being driven by science rather than politics. If confirmed, I will commit to maintain the integrity and independence of the BLS. Also, any changes to data or methods will be approved by career staff at the agency and communicated to the public ahead of time along with research supporting the change.
If confirmed, I would be leading the BLS at a pivotal moment. Our statistical system is built upon surveys of businesses and households. The continued decline in response rates threatens the future reliability of a survey-based statistical system and may already be having impacts on sample sizes, the variability of data, and revisions. My priority as Commissioner would be pursuing efforts to mitigate the decline in response rates while pursuing a research agenda into alternative data sources that can be used to supplement or replace traditional survey data. I will also focus on improving transparency and communication with data users. Finally, I have a great interest in working to improve our measurement of the impact of new technologies such as Artificial Intelligence.
It is a privilege to be here today, and I look forward to your questions.
* * *
Original text here: https://www.help.senate.gov/imo/media/doc/8aa0f266-cf6d-5110-f6a2-e4957bbe6185/Matsumoto%20Testimony_da31b8ff-3581-4b25-8cab-23908cca731b.pdf
* * *
Good morning.
Chairman Cassidy, Ranking Member Sanders, and Members of the Committee, it is an honor to appear before you as the nominee to be Commissioner of the Bureau of Labor Statistics. It is an honor to have been nominated by President Trump for this important position.
Before beginning, I would like ... Show Full Article WASHINGTON, June 24 -- The Senate Health, Education, Labor and Pensions Committee released the following testimony by Brett Matsumoto, President Trump's nominee to be commissioner of the U.S. Department of Labor Bureau of Labor Statistics, from a June 10, 2026, confirmation hearing: * * * Good morning. Chairman Cassidy, Ranking Member Sanders, and Members of the Committee, it is an honor to appear before you as the nominee to be Commissioner of the Bureau of Labor Statistics. It is an honor to have been nominated by President Trump for this important position. Before beginning, I would liketo thank all of the people who are responsible for getting me to where I am today: my parents and my older brothers, as well as other relatives, teachers, and mentors. I would also like to thank everyone who made positive statements regarding my nomination. When President Trump announced the intention to nominate me for this position, there was a lot of media interest. Although reading news stories about myself is something I'm still getting used to, I was humbled by the positive reactions of economists across the political spectrum.
I grew up in Delaware and attended college at the University of Delaware. After graduating with degrees in Philosophy, History, Economics, and a Masters in Economics, I went to the University of North Carolina for a PhD in Economics, where I specialized in Applied Microeconomics. After completing my PhD, I got a job at the Bureau of Labor Statistics.
I have worked for the Bureau of Labor Statistics since 2015 on economic measurement topics. I have published on topics relating to inflation and consumption measurement. I have also worked on projects to support the production offices, mainly the Consumer Price Index program. These include work on alternative data projects, most notably a project to replace traditionally collected data with insurance claims data for the medical price indexes. In September 2024, I was promoted to a supervisory position, where I oversee a small team of economists.
In the first Trump Administration, I did a detail at the Council of Economic Advisors (CEA) where I worked on labor and health topics. In early 2025, I was asked to do another detail by the incoming chairman of the CEA, and I have been in that role since. While working at the CEA, I have gained a great appreciation for the official economic statistics from the perspective of a data user. Much of the analysis done at the CEA would be impossible without the official data and having timely and accurate data is extremely important in the process of developing good economic policies.
I believe strongly in the mission of the BLS to produce "objective, quality measures and analyses of labor market activity, working conditions, price changes, and productivity in the U.S. Economy to support public and private decision making". To achieve this mission, it is important for the public to be confident that decisions at the BLS are being driven by science rather than politics. If confirmed, I will commit to maintain the integrity and independence of the BLS. Also, any changes to data or methods will be approved by career staff at the agency and communicated to the public ahead of time along with research supporting the change.
If confirmed, I would be leading the BLS at a pivotal moment. Our statistical system is built upon surveys of businesses and households. The continued decline in response rates threatens the future reliability of a survey-based statistical system and may already be having impacts on sample sizes, the variability of data, and revisions. My priority as Commissioner would be pursuing efforts to mitigate the decline in response rates while pursuing a research agenda into alternative data sources that can be used to supplement or replace traditional survey data. I will also focus on improving transparency and communication with data users. Finally, I have a great interest in working to improve our measurement of the impact of new technologies such as Artificial Intelligence.
It is a privilege to be here today, and I look forward to your questions.
* * *
Original text here: https://www.help.senate.gov/imo/media/doc/8aa0f266-cf6d-5110-f6a2-e4957bbe6185/Matsumoto%20Testimony_da31b8ff-3581-4b25-8cab-23908cca731b.pdf
Do No Harm Chief Medical Officer Miceli Testifies Before Senate Health, Education, Labor & Pensions Committee
WASHINGTON, June 24 -- The Senate Health, Education, Labor and Pensions Committee released the following written testimony by Kurt Miceli, chief medical officer of Do No Harm, Newton Square, Pennsylvania, from a June 3, 2026, hearing entitled "Protecting Our Children: Exposing the Dangers of Irreversible Gender Transition Procedures on Minors":
* * *
Chairman Cassidy, Ranking Member Sanders, and members of the Committee on Health, Education, Labor and Pensions, thank you for the invitation and opportunity to present on the dangers of gender transition procedures on minors.
My name is Kurt Miceli. ... Show Full Article WASHINGTON, June 24 -- The Senate Health, Education, Labor and Pensions Committee released the following written testimony by Kurt Miceli, chief medical officer of Do No Harm, Newton Square, Pennsylvania, from a June 3, 2026, hearing entitled "Protecting Our Children: Exposing the Dangers of Irreversible Gender Transition Procedures on Minors": * * * Chairman Cassidy, Ranking Member Sanders, and members of the Committee on Health, Education, Labor and Pensions, thank you for the invitation and opportunity to present on the dangers of gender transition procedures on minors. My name is Kurt Miceli.I am a psychiatrist and internist, having worked for hospitals and human-service providers in Southeast Pennsylvania. I am now chief medical officer for Do No Harm, a diverse group of over 50,000 healthcare professionals and concerned citizens committed to ensuring that the practice of medicine is driven by scientific evidence. Our mission is to defend the timeless standards of medical excellence against outside influences that would both erode them and undermine patient trust and public health. We advocate for what should be common sense: children should not be subjected to dangerous, life-altering procedures that lack scientific justification.
At Do No Harm, we have documented the scope of these gender transition procedures on minors. Our Stop the Harm database, launched in the fall of 2024 and based on insurance claims data, shows that nearly 14,000 minors received gender transition treatments between 2019 and 2023. This includes over 5,700 insurance claims for gender transition surgeries throughout the United States.1
Comparable to our findings, Landon Hughes et al., in a Research Letter published last year in JAMA Pediatrics, estimated that 1.4 out of every 1,000 female adolescents had received a prescription for cross-sex hormones by the time they turned 18./2 Jason Wright et al., in analyzing the period 2016 - 2020, found that thousands of children had undergone "gender-affirming surgery," with some being as young as 12 years old.3 All of this has taken place with little evidence supporting the efficacy or necessity of gender transition procedures for minors -- and against the backdrop of research suggesting that most children and adolescents will outgrow their dysphoria if allowed to progress through natural puberty into adulthood.4
It is also noteworthy that the gender dysphoria diagnosis has been found to lack specificity and, as the Cass Review found, is "not predictive" of future outcomes.5
This weak evidentiary foundation is further underscored by the fact that systematic reviews -- the gold standard in evidence-based medicine -- consistently rate the quality of research in this area as low or very low.6 Systematic reviews of evidence assess the entire body of evidence in a manner that is transparent and reproducible. Crucially, systematic reviews do not merely summarize what primary research says but evaluate studies for methodological quality to determine their reliability. Importantly, they consider bias, including publication bias.7
These limitations in the evidence base become even more apparent when one examines the very origins of pediatric gender transition, which was launched worldwide on the basis of two Dutch studies. Later systematic reviews that assessed these and subsequent studies, however, found that the research justifying gender transition in minors suffers from serious problems of methodological quality, including selection bias, dubious outcome measures, lack of control groups, and loss to follow-up. For this reason, the quality of evidence for benefit -- or certainty about the reported effects of medical interventions -- has consistently been rated as low or very low.8
Take, for instance, Kellan Baker at al.'s 2021 systematic review, which was conducted to inform the World Professional Association for Transgender Health (WPATH) Standards of Care, eighth edition (SOC-8). Baker et al. noted only three studies focused on adolescents, representing a total sample size of less than 200 individuals. Regarding quality of life (QOL), the systematic review included just one trial of 50 adolescents that showed "no difference in QOL scores after a year of endocrine interventions." And while just three studies on depression showed some improvement, the risk of bias in these studies was determined to be either "moderate" or "serious." Overall, Baker et al.'s systematic review notes a "high risk of bias in study designs, small sample sizes, and confounding with other interventions."9 The evidence of any benefit from gender transition procedures simply isn't there. Describing these interventions as "medically necessary," as WPATH has done, is irresponsible.10 Additionally, Baker et al.'s systematic review offers no evidence that so-called genderaffirming care decreases suicidality. In fact, the review comments on one study, specifically stating, "We cannot draw any conclusions on the basis of this single study about whether hormone therapy affects death by suicide among transgender people."11 No doubt, the death of any young person by suicide is tragic. It is a loss that reverberates through families, schools, and entire communities, and it deserves compassion rather than politicization.
At the same time, acknowledging the gravity of suicide does not require us to accept unsupported claims or to overlook the broader clinical realities that shape risk. Yet, in clinics across the country, parents faced with the decision of consenting to gender transition procedures on their minor child have been asked the rhetorical question, "Would you rather have a dead son or a living daughter?" That question's implicit assertion is simply not supported by the evidence. In comprehensively assessing suicide among adolescents and young adults, Ruuska et al. found that "clinical gender dysphoria does not appear to be predictive of all-cause nor suicide mortality when psychiatric treatment history is accounted for."12 In other words, once underlying psychiatric conditions are taken into full consideration, there is no convincing evidence that gender-referred youth face statistically higher suicide rates than their peers. Furthermore, claims that suicide rates would surge if puberty blockers or cross-sex hormones were made unavailable are likewise unsupported by the evidence. A 2024 independent review from England and Wales makes this clear.13
Furthermore, a recent Finnish nationwide register study looking at over two decades of data assessed the prevalence of severe psychiatric morbidity among adolescents referred for specialized gender services. As the study noted, "[a]mong adolescents who underwent medical gender reassignment, psychiatric morbidity increased markedly during follow-up." Psychiatric needs did not subside after so-called gender-affirming care; instead, they rose, contradicting the proposition that mental health improves following transition-related interventions.14
Systematic reviews published by Jo Taylor et al. are also enlightening. One such review focused on puberty blockers in adolescents and included 50 studies, of which only one was determined to be of high quality. Consequently, the review found "insufficient and/or inconsistent evidence about the effects of puberty suppression on gender-related outcomes, mental and psychosocial health, cognitive development, cardiometabolic risk, and fertility." However, there was moderate-quality evidence that "bone density and height may be compromised during treatment."15 The evidence reveals harms from puberty blockers more than it finds benefit.
Similarly, a systematic review also by Taylor et al. assessing hormonal interventions included 53 studies, of which only one was determined to be of high quality. The result: "a lack of high-quality evidence to support the initiation of hormones for masculinization or feminization in adolescents experiencing gender dysphoria/incongruence."16
These reviews by Taylor et al., along with four more, were a key part of the academic research commissioned by the Cass Review. Released in April 2024, the Cass Review confirmed the weakness of evidence in pediatric medical transition and specifically stated that "we have no good evidence on the long-term outcomes of interventions to manage gender-related distress."17 Indeed, studies related to pediatric medical transition in minors -- outside of systematic reviews, which provide the most reliable evidence -- are often of low quality given small sample sizes, significant attrition of enrollees, and relatively short followup periods. Selection bias, uncontrolled confounding, and the lack of a comparison group also weigh heavily on many of these studies. All this leads to low-quality evidence, which simply should not be used to make clinical decisions -- especially those that are life-altering, as is the case with these procedures.
Additional systematic reviews have found the same. Thompson and colleagues, in their research article "A PRISMA systematic review of adolescent gender dysphoria literature: 3) treatment," found that "[t]he evidence base for the outcomes of gender dysphoria treatment in adolescents is lacking." Further, "[i]t is impossible from the included data to draw definitive conclusions regarding the safety of treatment."18
The pattern is clear: the evidence supporting gender transition procedures in minors is at best of very low quality and at worst nonexistent. However, the harms caused by these interventions cannot be ignored. Take cross-sex hormones, for instance. Outside of the cardiovascular risks and various cancer risks associated with taking exogenous hormones, the risk of infertility is real.19,20,21,22,23,24,25,26,27,28,29,30 It is for this reason that even the Endocrine Society's guidelines recommend that "clinicians inform and counsel all individuals seeking gender-affirming medical treatment regarding options for fertility preservation prior to initiating puberty suppression in adolescents and prior to treating with hormonal therapy of the affirmed gender in both adolescents and adults."31
Likewise, the American Academy of Pediatrics' Policy Statement -- in a footnote -- acknowledges that "the effect of sustained puberty suppression on fertility is unknown." It further states that "when cross-sex hormones are initiated without endogenous hormones, then fertility may be decreased."32 Never mind the additional surgical complications that may arise from these interventions, which, in the case of genital surgeries, range from wound-related to urological, including, but not limited to, infection, "wound dehiscence," "trauma from intercourse/dilation," "change in voiding function," and "urethral stenosis."33 Do No Harm's 2025 report "Hormonal Interventions for Minors with Gender Dysphoria Cause Significant Harm" further addresses the risks associated with cross-sex hormones. As the report makes clear, risks to patients' sexual and reproductive systems include not only infertility but pelvic pain and discomfort during sexual activity.34Cardiovascular risks include an increased incidence of heart attacks among women who receive hormonal interventions,35 as well as a seven-fold higher rate of erythrocytosis (increase in red blood cell concentration, which may increase risk of blood clots, leading to heart attack and stroke) compared to men not taking testosterone.36
Women taking testosterone may experience changes that lead to unsatisfactory or harderto-interpret Pap smear tests, which can make cervical cancer screening far more challenging.37 In cases of severe obesity, endometrial cancer (cancer of the uterine lining) appears to increase with the duration of testosterone use.38
Nor do males who take estrogen avoid their own health risks. In addition to fertility consequences, such men face an increased risk of venous thromboembolisms (clots in veins that can pass to the lung and cause death),39 retinal vein occlusions (blockages in blood flow from the eye),40,41 strokes,42 and invasive breast cancer. The latter is 46 times more likely to occur in men who take estrogen than in men who don't.43 Worse, these individuals face a markedly elevated mortality.44
Similarly harrowing on this topic of harms is the U.S. Department of Health and Human Services' 2025 report "Treatment for Pediatric Gender Dysphoria: Review of Evidence and Best Practices." According to that document, endocrine and surgical interventions on minors carry the risk of "infertility/sterility, sexual dysfunction, impaired bone density accrual, adverse cognitive impacts, cardiovascular disease and metabolic disorders, psychiatric disorders, surgical complications, and regret." Moreover, "there has been inadequate research into the frequency and severity of these harms," leaving a gap that raises significant safety concerns.45 Regarding cognitive impacts, it should be noted that there is evidence from both animal and human research that puberty blockers may arrest brain development.46
Despite these documented concerns, organized medicine in the United States has, for the better part of the last decade, largely endorsed and promoted so-called gender-affirming care -- including social transition, puberty blockers, cross-sex hormones, and surgeries -- for minors experiencing gender dysphoria. Major professional organizations such as the American Academy of Pediatrics (AAP),47 the Endocrine Society,48 and the American Medical Association (AMA)49 have followed WPATH50 in issuing guidelines or policy statements framing these interventions as medically necessary and evidence-based standards of care.
This institutional consensus has helped to shape clinical practice, insurance coverage, medical education, and public messaging, often portraying dissent or calls for caution as ideologically driven rather than grounded in emerging evidence.
In 2018, for example, the AAP released a policy statement entitled "Ensuring Comprehensive Care and Support for Transgender and Gender-Diverse Children and Adolescents."51 This document framed skepticism towards pediatric medical transition as pathologizing, urging an affirmation-only approach. It presented affirmation as settled science while downplaying evidence gaps.
But the AAP did not act alone, as The New York Times and other publications have reported.52
In fact, on the issue of pediatric gender transition, the AAP and other major U.S. medical organizations have largely followed the lead of WPATH, which has positioned itself as the leading authority in transgender healthcare, noting that its standards of care are the "foremost evidence-based guidelines for the provision of [transgender and gender diverse] healthcare."53 Its guidelines and its members have held themselves out as the global experts, exerting outsized influence over clinical practice, policy, and litigation across the United States and internationally. In short, pediatric gender medicine is WPATH, and WPATH is pediatric gender medicine.
Accordingly, WPATH's credibility and authority has been invoked by almost two dozen major medical groups, including the AMA, in legal briefs.54,55 And, most significantly, its recommendations have received the imprimatur of the AAP and the Endocrine Society, whose own guidelines on youth gender transition were either written by WPATH members or heavily influenced by the WPATH approach.
Furthermore, to understand WPATH's broad influence, consider that the Standards of Care eighth edition (SOC-8) are referenced in many insurance policies for gender transition procedures on minors.56,57,58 In some states, such as New York, "health maintenance organizations and health insurers must apply utilization review criteria consistent with version of the World Professional Association for Transgender Health (WPATH) Standards of Care when conducting utilization review of treatment for Gender Dysphoria."59 Similarly, some children's hospitals list WPATH guidelines as educational resources.60,61
The organization's influence has been profound in the realm of pediatric medical transition, as the medical community within the United States, and even worldwide, has historically yielded deference to WPATH as the preeminent authority in this field. Through this widespread adoption -- by public and private insurers, healthcare providers, professional societies, and regulatory bodies -- WPATH's standards became the de facto framework guiding care for gender-dysphoric youth across the United States, propping up the guidelines and statements from the Endocrine Society,62 AAP,63 American Psychiatric Association,64 and American Psychological Association.65 WPATH's continuing medical education courses, offered through its Global Education Institute (GEI) Certified Training Courses, have been widely used and have shaped the medical education of countless doctors across the country and planet in the area of youth gender medicine.66 All of this even as questions mounted about the strength of the underlying evidence.
Indeed, the foundation upholding WPATH's recommendations has effectively collapsed. In 2016, WPATH issued a position statement adjudging as "medically necessary" transgender "treatment, sex reassignment, and insurance coverage in the U.S.A."67 This language appears repeatedly in SOC-8, where "medically necessary gender-affirming medical treatment" is detailed for adolescents, as are "aspects of medical[ly] necessary care intended to promote the well-being and gender-related needs of children."68 Yet, as reported in The New York Times, one SOC-8 author noted how casting gender transition procedures as medically necessary could advantageously be "a tool for our attorneys to use in defending access to care."69 In reality there was no true medical necessity; the phrase was a strategic declaration designed to strengthen legal and insurance defenses rather than a conclusion grounded in robust, high-quality clinical evidence, particularly when it came to the care of minors.
By 2024, internal WPATH communications obtained by the state of Alabama through subpoena further underscored this pattern, revealing evidence of serious misconduct in the development of the SOC-8 guidelines.70,71,72,73,74 In a particularly revealing episode, WPATH commissioned The Johns Hopkins University Evidence-Based Practice Center to conduct multiple systematic reviews to inform SOC-8. Systematic reviews are considered the gold standard in evidence-based medicine because they evaluate the totality of evidence and, crucially, examine the underlying methodological quality of the research, allowing for determinations about the quality of evidence.75
However, as reported in The Economist, "documents show that [WPATH's] leaders interfered with the production of [these] systematic reviews." The report continued, "From early on in the contract negotiations, WPATH expressed a desire to control the results of the Hopkins team's work."76 This is not how independent scientific research is conducted. As a result, even the Evidence-Based Practice Center's ostensibly impartial analyses were ultimately constrained by WPATH's requirement that nothing be used without its sign-off. Further, WPATH adopted a policy giving it the authority to "nip papers in the bud on the basis of their conclusions." The result: only one paper from the Hopkins team was published after this policy took hold, despite the group "having provided WPATH with the material for six systematic reviews," according to The Economist.77
Indeed, as revealed by BMJ Investigation, some WPATH SOC-8 authors worried that "independent appraisals of the evidence would undermine legal efforts to protect affirming interventions from legislative restriction in minors." BMJ continued, "a chapter author wrote, 'Our concerns, echoed by the social justice lawyers we spoke with, is that evidence based review reveals little or no evidence and puts us in an untenable position in terms of affecting policy or winning lawsuits.'"78
By August 2020, the Hopkins team reported to the Agency for Healthcare Research and Quality at the U.S. Department of Health and Human Services (HHS) that they found "little to no evidence about children and adolescents" and that WPATH was "trying to restrict our ability to publish."79 WPATH leadership effectively suppressed manuscripts whose conclusions did not align with the organization's preferred positions, successfully pressuring the researchers not to publish their findings. SOC-8 nevertheless proceeded to advance claims of medical necessity, falsely asserting that a systematic review of adolescent outcomes was "not possible."80 Such actions represent a serious breach of scientific integrity and ethical standards in guideline development.
News of WPATH's misconduct later appeared in the mainstream press, including liberal media. "Research into trans medicine has been manipulated," read a headline in The Economist.81 "The World Professional Association for Transgender Health interfered with systematic reviews it commissioned from a research team at Johns Hopkins University," wrote the editorial board of The Washington Post.82 "This is not how evidence-based medicine is supposed to work," reported The Atlantic.83
WPATH's misgivings did not end there. As reported in The New York Times, President Biden's Assistant Secretary for Health, Admiral Rachel Levine, "had been instrumental in WPATH's mysterious last-minute deletion of the age minimums in SOC-8."84 Released in late 2021, WPATH's draft guidelines proposed specific lower age thresholds than had been in previous versions of its Standards of Care: "14 for hormonal treatments, 15 for mastectomies, 16 for breast augmentation or facial surgeries, and 17 for genital surgeries or hysterectomies."85 However, within hours of publishing the final standards of care in September 2022, WPATH deleted from the document age minimums for all hormonal interventions and most surgeries, issuing a "correction," according to The Economist. The news magazine continued, noting that "the head of the drafting committee, Eli Coleman, said the publisher went ahead 'without approval' before final changes were made."86 Thus, the age thresholds that appeared in the draft -- and initially in the published version -- were swiftly excised from the final document.
As emails revealed, this decision followed sustained pressure from Levine, whose political support WPATH needed, and from the AAP. Specifically, Levine's staff urged WPATH to remove these explicit age limits over concerns that they could fuel political opposition and restrict access to care for transgender youth.87 Ultimately, the corrected SOC-8 omitted nearly all age minimums, stating "no recommendations on specific ages," with one exception: phalloplasty, which "WPATH stressed should not be performed under the age of 18 years owing to its complexity."88 These decisions weren't driven by science but, rather, political pressure.
Even some WPATH leaders were upset by this outside meddling. "I don't know how I feel about allowing U.S. politics to dictate international professional clinical guidelines that went through Delphi," wrote one.89 Delphi refers to the Delphi method, a structured, iterative consensus-building process commonly used in healthcare guideline development when high-quality evidence is limited or mixed.90 SOC-8 specifically notes that "[c]onsensus on the final recommendations was attained using the Delphi process that included all members of the guidelines committee and required that recommendation statements were approved by at least 75% of members."91 Clearly, this episode underscores a deeper tension in the development of SOC-8: the gap between the formal consensus process and the final guidelines that emerged after external intervention. Removing age minimums -- after they had already achieved Delphi consensus -- prioritized political considerations over caution in an area where systematic reviews were finding the evidence base to be weak.
Nevertheless, when the AAP threatened to publicly oppose WPATH's guideline unless they got rid of the age minimums, WPATH relented.92 WPATH's president wrote in an email that it was "disappointing that politics always trumps common sense and what is best for patients."93 A former WPATH president involved in the guideline process wrote that the deletion of age minimums "is a balancing act between what I feel to be true and what we need to say."94
Additionally, in WPATH's development of its guidelines, "opaque conflict of interest management" was also evident,95 with significant deviation from standard practice.
Subpoenaed emails and later depositions revealed how WPATH leaders failed to manage conflicts of interest among guideline authors.96,97,98 Nevertheless, the published version of SOC-8 states that "[n]o conflicts of interest were deemed significant or consequential."99 This lax approach manifested in fundamental procedural failures. For one, conflict-ofinterest disclosure forms were not due until after SOC-8 committee members were selected. This information was not part of the selection process. As Leor Sapir noted in his article "The Deposition of Eli Coleman," Coleman, the lead author and chair of SOC-8, "testified that he knew 'most' members of SOC-8 had conflicts." 100
Take, for example, one of SOC-8's authors, a plastic surgeon who has performed multiple "gender-affirming surger[ies]" on pediatric patients.101 That surgeon is now the incoming president of WPATH.102 He declared no conflict of interest in SOC-8. Similarly, in a legal deposition, Marci Bowers, a gender surgeon and former president of WPATH who also authored its guideline, admitted to making over one million dollars from performing gender surgeries in 2023 alone.103, 104 No conflict of interest was declared in SOC-8.
Nevertheless, Coleman and his SOC-8 co-chairs determined that all of these conflicts were "neither 'significant' nor 'consequential.'" As a result, "they thought, WPATH did not need to specify how, if at all, they were managed, and did not need to disclose them in the relevant section of the published guideline."105
Taken together, these revelations illustrate a broader, deeply concerning pattern: WPATH's SOC-8 was driven by ideological advocacy, legal strategy, and political pressure rather than by transparent, rigorous, and independent science -- undermining the very foundation of evidence-based medicine. This seeming malfeasance is further discussed in a United States v. Skrmetti amicus brief that reads as a startling narrative of message-shaping on the part of WPATH rather than scientific prowess.106
Yet, WPATH's influence throughout the halls of pediatric gender medicine reigned supreme.
In its September 13, 2017, press release, the Endocrine Society issued its guidelines for "gender-affirmation treatment," officially listing WPATH as a co-sponsor.107 In fact, many of the Endocrine Society guideline authors on the same topic were either WPATH members or had other close ties to the organization.108
This interconnected web of guidelines promoting gender transition procedures for minors extended even further. Jo Taylor and colleagues have shown that gender medicine guidelines repeatedly drew upon one another, with many "cit[ing] and draw[ing] on the World Professional Association for Transgender Health (WPATH) guidelines."109 The result was a self-reinforcing cycle in which position statements and guidelines repeatedly referenced one another, creating the appearance of a robust evidentiary foundation in pediatric gender medicine -- one that, through repetition across the medical literature and within organized medicine, came to seem unquestionable.
Under these conditions, the U.S. health system effectively institutionalized gender transition procedures for minors as standard care. Major medical organizations, hospitals, and insurers treated WPATH's SOC-8 and the closely aligned Endocrine Society guidelines as authoritative, leading to widespread adoption of puberty blockers, cross-sex hormones, and surgeries -- often with minimal gatekeeping. This occurred even as rigorous, goldstandard evidence reviews -- including the Cass Review in the UK and multiple European systematic reviews -- concluded that the benefits of these interventions were of very low certainty, while the risks were substantial.
Yet the past makes clear that medicine is not immune to profound error -- and, at times, to grave wrongs. History offers sobering reminders: physician-endorsed smoking,110 the brutality of lobotomies,111 the use of diethylstilbestrol (DES) in pregnancy,112 the Tuskegee Syphilis Experiment,113 and, most recently, opioid overprescribing and the resulting crisis114 -- to name just a few examples. The record of modern medicine is marked by episodes of misjudgment and sometimes serious scandal. Entire books and scholarly analyses have been devoted to understanding how harmful practices are introduced, or why ineffective and unsafe treatments persist long after evidence has shown they should be abandoned. As physicians, we hold the capacity to do immense good -- but also, if we are not vigilant, immense harm.
Our profession is far from infallible. Yet it is not the absence of error that earns the public's trust in medicine. Rather, it is our willingness, as physicians, to confront mistakes honestly, to reassess our assumptions, and to correct course when the evidence demands it.
For this reason, open inquiry, rigorous debate, peer review, critical appraisal of evidence, and independent evaluation are indispensable. These commitments gave rise to evidence-based medicine -- a systematic approach to clinical decision-making grounded in the best available scientific data.
Yet today, in the case of pediatric medical transition, empirical evidence has been systematically sidelined in favor of ideology. Such widespread politicization has prevented these self-corrective processes from operating as intended. Self-appointed experts have broken the chain of trust, and the medical community has been misled by ostensible authorities who have minimized or, in some cases, veiled altogether the gaps in their knowledge.
Gradually, however, the tide has begun to turn. Given the weak quality of evidence and the obvious harms of gender transition procedures on minors, countries throughout the world, as well as 27 states here at home, have taken a stance to protect children by stopping many of these interventions from occurring. In May 2024, England, for example, banned the clinical use of puberty blockers in those under 18 years of age.115 This followed publication of the Cass Review, whose methodology was recently found to be "robust" by the British Medical Association's board of science.116 That ban on puberty blockers had the support of Conservative and Labour governments alike.117 And just a few months ago, BMJ reported that "NHS England has 'paused' the prescribing of gender affirming hormones to treat gender incongruence or dysphoria for young people under 18."118
Likewise, in recent years Sweden and Finland -- among the world's most socially progressive nations -- have imposed limits on gender transition procedures for minors, all of them initiated under left-leaning governments.119 Their actions were driven by evidence -- or, more precisely, by the lack of evidence supporting so-called gender-affirming care.
Systematic reviews prompted these course corrections. In Sweden's review, for example, the evidence for hormone treatment in minors was deemed "insufficient," and the long-term psychosocial effects of hormone therapy were described as "unknown." The authors further classified puberty blockers as an "experimental treatment."120 Today, nearly two dozen systematic reviews have all consistently demonstrated the absence of a credible evidence base.121
An increasing number of medical professionals are also coming to recognize that pediatric gender transition is the medical scandal of our generation. In my role at Do No Harm, I hear this regularly. Over the past year alone, our organization has grown substantially as more and more clinicians have become aware of the risks associated with youth gender
transition. We routinely receive messages from physicians who know this to be a scandal, but who feel unable to voice their concerns within their professional circles for fear of personal or professional reprisal. This climate of silence creates the misleading impression that most physicians support these interventions. They do not. In a survey of Florida physicians, for example, we found that 66 percent supported the state's prohibition on gender transition procedures for minors.122
In fact, in February of this year, the American Society of Plastic Surgeons (ASPS), representing over 11,000 surgeons,123 became the first major U.S. medical society to recognize the grave concerns associated with gender transition procedures on minors.
ASPS's position statement openly acknowledged the lack of supportive evidence for these procedures, explicitly calling out "gender-related endocrine and surgical interventions." It therefore recommended against gender related surgeries in anyone under 19./124
Additionally, ASPS emphasized an often-ignored truth: "the natural course of pediatric gender dysphoria remains poorly understood."125 For one, clinicians cannot reliably predict which children will persist in their distress and which will desist. This uncertainty is not a minor footnote. The 2025 guidelines from Germany recognize this fact, providing no criteria to differentiate between those children with "temporary 'gender non-contentedness'" and "stable/persistent cases."126 ASPS was thereby right to recognize that, when a condition's trajectory is uncertain and the interventions irreversible, medical ethics requires caution, not acceleration. This applies to both surgeries and medications, including puberty blockers and cross-sex hormones, whose long-term physiological effects and side effects cannot be fully undone.
Moreover, ASPS's position statement sharpened the proper role of autonomy in clinical ethics. Patient autonomy does not obligate physicians to provide interventions when the risk-benefit profile is uncertain or unfavorable -- particularly for minors whose decisionmaking capacity is still maturing. Medicine must move away from the "have it your way" model of care that has failed children with gender dysphoria, fast-tracking them to hormones and surgeries despite systematic reviews showing very low evidence of benefit and significant risk of harm.
This is especially true given the ethical concerns raised by the impossibility of informed consent (or even informed assent) on the part of adolescents whose brains have not yet fully developed. As ASPS's position statement points out, "gender-related surgery procedures intervene directly in the processes of identity formation and psychosexual development.
These are areas of ongoing maturation during adolescence that warrant particular ethical caution as surgeons assess adolescent medical decision-making capacity."127 The same can be said for puberty blockers and cross-sex hormones, which similarly disrupt critical windows of brain and bodily development during a period when true informed decisionmaking capacity remains fundamentally limited.
All of this comes on the heels of the comprehensive umbrella review conducted by the U.S. Department of Health and Human Services, which itself has undergone peer review and provides a methodologically sound and transparent assessment of the evidence.
The HHS Review makes it clear: the quality of evidence regarding key domains such as psychological outcomes and quality of life, as well as long-term health, is very low. The benefits simply aren't there. Yet, the evidence for harms clearly exists and can be drawn from "established knowledge about human physiology and the effects and mechanisms of the pharmacological agents used."128 These effects are real and can't be ignored.
In coming to these conclusions, the HHS Review performed a substantial evidence review.
Following the recommendations for overviews of systematic reviews as per the Cochrane Handbook for Systematic Reviews of Interventions, thousands of studies were screened, with ultimately 17 systematic reviews included. As noted in the HHS Review, "two reviewers independently screened titles and abstracts, and then full texts, to determine study eligibility." The Risk of Bias Assessment Tool for Systematic Reviews (ROBIS) was used to assess the included systematic reviews. Evidence synthesis was performed and organized by outcomes. The effect estimates and quality of evidence followed the GRADE methodology.129
The HHS Review, however, didn't stop there. It included evidence from basic science and physiology to help best understand the harms of puberty blockers and cross-sex hormones.
It reviewed international clinical-practice guidelines, utilizing key systematic reviews. It also added a critically important chapter on ethical considerations as well as psychotherapy as a means of caring for children suffering from gender dysphoria.
Having released an early version of their Review in May of last year, the HHS Review team invited skeptical physicians and medical associations to scrutinize their conclusions. It invited critics to offer a peer review, specifically reaching out to groups such as the AAP, the Endocrine Society, and the American Psychiatric Association (APA). Regrettably, all three organizations have strongly supported gender transition procedures on minors, and they vehemently opposed HHS's initial review in May 2025, with the AAP condemning the Review within hours of its May 1 release.130
Yet, despite HHS's attempt to engage them, the AAP and the Endocrine Society declined to provide a peer review. That choice speaks for itself: when invited to present evidence supporting medical interventions for minors, they opted not to participate. If they had strong, well-supported arguments, they would have taken the opportunity to lay them out and challenge the findings in the HHS Review. They didn't.
Only the APA, to its credit, submitted a peer review. Still, the review offered very little meaningful critique. The APA even appeared to overlook key sections of the Review, asserting that HHS had failed to outline its methodology when the Review clearly did so. In the end, the APA identified no substantive flaws in the HHS findings and raised no objections to the Review's treatment of psychotherapy as an alternative to hormonal interventions. At the end of the day, none of the critical reviewers disproved the central conclusions of the HHS Review. Please read those critiques for yourself, and you'll see that the HHS Review withstood peer-review scrutiny, earning its place as a robust umbrella review.
Put simply, the latest HHS Review is the most comprehensive and evidence-based review of gender transition procedures on minors. It lends support to the many states across our country that have taken a stance to keep kids safe from these interventions, for which the evidence of benefit is lacking and the harm is unfortunately lasting and significant.131 Such laws echo common sense and public sentiment. In a New York Times/Ipsos survey from last year, 71 percent of respondents agreed that "no one under age 18 should have access [to puberty-blocking drugs or hormone therapy]" when used for transgender care.
Among Democrats, only 19 percent felt minors as young as 10 should be allowed access to puberty blockers for these purposes, while 54 percent said no one under 18 should have access.132
It is also important to recognize that many children later come to regret their transition and themselves detransition years later. By one study's account, at least 30 percent of youth discontinue medical transition within just a few years, but many of the effects of hormones and surgeries are irreversible.133 While low regret from pediatric medical transition is frequently cited in the popular press, this conclusion is based on flawed studies with narrow windows of time for follow-up, high rates of attrition (i.e., individuals "lost to follow-up" who simply stopped coming to their appointments), and biased samples or those that cannot be generalized.134 In reality, the true rate of regret is not known.135
It is in part for this reason that Do No Harm recently urged the Centers for Disease Control and Prevention to create new medical diagnosis codes (known as ICD-10 codes) for desistance and detransition.136 Currently, the lack of a specific code for detransition makes those who have detransitioned effectively invisible to the healthcare system, obscuring their medical needs and limiting the real-world research, follow-up care, and outcome monitoring that responsible practice requires.
Moreover, many within the current cohort of gender-dysphoric youth differ markedly from earlier ones. In recent years there has been a dramatic surge in adolescents -- predominantly female adolescents -- with "rapid-onset gender dysphoria," as described by Lisa Littman in 2018. These youth, often influenced by peer groups and social media, frequently suffer from significant psychiatric or neurodevelopmental comorbidities in addition to their gender confusion.137,138 It is these co-morbid mental health conditions that should be the focus of treatment.
Instead, the rush has been to treat gender dysphoria with an affirming model inclusive of medications and surgeries. In many cases, supply has unfortunately created its own demand: simply being seen by a gender specialist is itself a strong predictor that a child will proceed with social and medical transition.139
By contrast, a major follow-up study of boys diagnosed with gender identity disorder found that more than 87 percent desisted by early adulthood -- that is, they no longer experienced gender dysphoria.140 Many of these boys later identified as homosexual. Consistent with this, the DSM-5-TR, last published in March 2022, notes that the rate of persistence for childhood gender dysphoria into adolescence or adulthood ranges from only 2.2 percent to 50 percent, meaning the clear majority of cases resolve on their own or desist.141
Desistance is real, and psychotherapy can be genuinely beneficial to youth struggling with gender confusion, particularly when depression, suicidal thoughts, and self-harm coexist.
These children need high-quality behavioral healthcare, not gender transition procedures.
Finland, for example, has recognized this explicitly, recommending that "first-line treatment for gender dysphoria is psychosocial support and, as necessary, psychotherapy and treatment of possible comorbid psychiatric disorders."142
For the sake of our children, it is time to bring an end to gender transition procedures for minors. The evidence base is weak, with systematic reviews finding no reliable data supporting these interventions. It is time for pediatric gender medicine to move beyond WPATH's SOC-8, a document shaped by opaque processes, marked by conflicts of interest, and influenced by ideological advocacy that has sacrificed science for activism, resulting in the reckless promotion of low-evidence, high-risk medical interventions on minors. Instead, we must acknowledge the certainty of harms done by pediatric medical transition alongside the uncertainty of any benefit. No responsible field of medicine would continue to promote interventions where the harms are well-established and the benefits remain unproven.
If one wishes to look beyond the evidence, then please listen to those who have detransitioned; their stories are powerful reminders of the compassion we must have in caring for youth with gender dysphoria, as well as the resolve we need to first and foremost end these experimental gender transition procedures on minors.
To my fellow physicians: Many of us have relied on colleagues regarded as experts in gender medicine. That is common in a field as complex as the medical profession -- no one can master every specialty. But our responsibility is not solely to trust, or defer, but also to verify. Given the documented harms and the vulnerability of the children involved, unquestioning trust is no longer acceptable. I urge my peers to look beyond partisan narratives and examine the evidence directly. If we hope to restore the public's confidence in our profession, we must begin by acknowledging that we have made mistakes -- grave ones - and that the time to correct them is now.
It is time for the medical profession to return to the evidence. It is time for us to stand with the nation in protecting vulnerable children and ending procedures that carry irreversible harm. It is time to enshrine in law a clear commitment to safeguarding children who are struggling with gender dysphoria and related mental-health challenges. They deserve quality therapy and robust psychosocial support -- not irreversible, life-altering medical interventions. Above all, we must remember our duty: First, do no harm.
Thank you for your time and for reading my testimony. I look forward to further discussing this important matter with the Committee.
* * *
Original text here: https://www.help.senate.gov/imo/media/doc/836dcbdb-dc18-2dcd-e758-5033879edf61/Miceli%20Testimony_481e9f75-cdd8-4a38-966b-fbcb2fda282c.pdf
* * *
Chairman Cassidy, Ranking Member Sanders, and members of the Committee on Health, Education, Labor and Pensions, thank you for the invitation and opportunity to present on the dangers of gender transition procedures on minors.
My name is Kurt Miceli. ... Show Full Article WASHINGTON, June 24 -- The Senate Health, Education, Labor and Pensions Committee released the following written testimony by Kurt Miceli, chief medical officer of Do No Harm, Newton Square, Pennsylvania, from a June 3, 2026, hearing entitled "Protecting Our Children: Exposing the Dangers of Irreversible Gender Transition Procedures on Minors": * * * Chairman Cassidy, Ranking Member Sanders, and members of the Committee on Health, Education, Labor and Pensions, thank you for the invitation and opportunity to present on the dangers of gender transition procedures on minors. My name is Kurt Miceli.I am a psychiatrist and internist, having worked for hospitals and human-service providers in Southeast Pennsylvania. I am now chief medical officer for Do No Harm, a diverse group of over 50,000 healthcare professionals and concerned citizens committed to ensuring that the practice of medicine is driven by scientific evidence. Our mission is to defend the timeless standards of medical excellence against outside influences that would both erode them and undermine patient trust and public health. We advocate for what should be common sense: children should not be subjected to dangerous, life-altering procedures that lack scientific justification.
At Do No Harm, we have documented the scope of these gender transition procedures on minors. Our Stop the Harm database, launched in the fall of 2024 and based on insurance claims data, shows that nearly 14,000 minors received gender transition treatments between 2019 and 2023. This includes over 5,700 insurance claims for gender transition surgeries throughout the United States.1
Comparable to our findings, Landon Hughes et al., in a Research Letter published last year in JAMA Pediatrics, estimated that 1.4 out of every 1,000 female adolescents had received a prescription for cross-sex hormones by the time they turned 18./2 Jason Wright et al., in analyzing the period 2016 - 2020, found that thousands of children had undergone "gender-affirming surgery," with some being as young as 12 years old.3 All of this has taken place with little evidence supporting the efficacy or necessity of gender transition procedures for minors -- and against the backdrop of research suggesting that most children and adolescents will outgrow their dysphoria if allowed to progress through natural puberty into adulthood.4
It is also noteworthy that the gender dysphoria diagnosis has been found to lack specificity and, as the Cass Review found, is "not predictive" of future outcomes.5
This weak evidentiary foundation is further underscored by the fact that systematic reviews -- the gold standard in evidence-based medicine -- consistently rate the quality of research in this area as low or very low.6 Systematic reviews of evidence assess the entire body of evidence in a manner that is transparent and reproducible. Crucially, systematic reviews do not merely summarize what primary research says but evaluate studies for methodological quality to determine their reliability. Importantly, they consider bias, including publication bias.7
These limitations in the evidence base become even more apparent when one examines the very origins of pediatric gender transition, which was launched worldwide on the basis of two Dutch studies. Later systematic reviews that assessed these and subsequent studies, however, found that the research justifying gender transition in minors suffers from serious problems of methodological quality, including selection bias, dubious outcome measures, lack of control groups, and loss to follow-up. For this reason, the quality of evidence for benefit -- or certainty about the reported effects of medical interventions -- has consistently been rated as low or very low.8
Take, for instance, Kellan Baker at al.'s 2021 systematic review, which was conducted to inform the World Professional Association for Transgender Health (WPATH) Standards of Care, eighth edition (SOC-8). Baker et al. noted only three studies focused on adolescents, representing a total sample size of less than 200 individuals. Regarding quality of life (QOL), the systematic review included just one trial of 50 adolescents that showed "no difference in QOL scores after a year of endocrine interventions." And while just three studies on depression showed some improvement, the risk of bias in these studies was determined to be either "moderate" or "serious." Overall, Baker et al.'s systematic review notes a "high risk of bias in study designs, small sample sizes, and confounding with other interventions."9 The evidence of any benefit from gender transition procedures simply isn't there. Describing these interventions as "medically necessary," as WPATH has done, is irresponsible.10 Additionally, Baker et al.'s systematic review offers no evidence that so-called genderaffirming care decreases suicidality. In fact, the review comments on one study, specifically stating, "We cannot draw any conclusions on the basis of this single study about whether hormone therapy affects death by suicide among transgender people."11 No doubt, the death of any young person by suicide is tragic. It is a loss that reverberates through families, schools, and entire communities, and it deserves compassion rather than politicization.
At the same time, acknowledging the gravity of suicide does not require us to accept unsupported claims or to overlook the broader clinical realities that shape risk. Yet, in clinics across the country, parents faced with the decision of consenting to gender transition procedures on their minor child have been asked the rhetorical question, "Would you rather have a dead son or a living daughter?" That question's implicit assertion is simply not supported by the evidence. In comprehensively assessing suicide among adolescents and young adults, Ruuska et al. found that "clinical gender dysphoria does not appear to be predictive of all-cause nor suicide mortality when psychiatric treatment history is accounted for."12 In other words, once underlying psychiatric conditions are taken into full consideration, there is no convincing evidence that gender-referred youth face statistically higher suicide rates than their peers. Furthermore, claims that suicide rates would surge if puberty blockers or cross-sex hormones were made unavailable are likewise unsupported by the evidence. A 2024 independent review from England and Wales makes this clear.13
Furthermore, a recent Finnish nationwide register study looking at over two decades of data assessed the prevalence of severe psychiatric morbidity among adolescents referred for specialized gender services. As the study noted, "[a]mong adolescents who underwent medical gender reassignment, psychiatric morbidity increased markedly during follow-up." Psychiatric needs did not subside after so-called gender-affirming care; instead, they rose, contradicting the proposition that mental health improves following transition-related interventions.14
Systematic reviews published by Jo Taylor et al. are also enlightening. One such review focused on puberty blockers in adolescents and included 50 studies, of which only one was determined to be of high quality. Consequently, the review found "insufficient and/or inconsistent evidence about the effects of puberty suppression on gender-related outcomes, mental and psychosocial health, cognitive development, cardiometabolic risk, and fertility." However, there was moderate-quality evidence that "bone density and height may be compromised during treatment."15 The evidence reveals harms from puberty blockers more than it finds benefit.
Similarly, a systematic review also by Taylor et al. assessing hormonal interventions included 53 studies, of which only one was determined to be of high quality. The result: "a lack of high-quality evidence to support the initiation of hormones for masculinization or feminization in adolescents experiencing gender dysphoria/incongruence."16
These reviews by Taylor et al., along with four more, were a key part of the academic research commissioned by the Cass Review. Released in April 2024, the Cass Review confirmed the weakness of evidence in pediatric medical transition and specifically stated that "we have no good evidence on the long-term outcomes of interventions to manage gender-related distress."17 Indeed, studies related to pediatric medical transition in minors -- outside of systematic reviews, which provide the most reliable evidence -- are often of low quality given small sample sizes, significant attrition of enrollees, and relatively short followup periods. Selection bias, uncontrolled confounding, and the lack of a comparison group also weigh heavily on many of these studies. All this leads to low-quality evidence, which simply should not be used to make clinical decisions -- especially those that are life-altering, as is the case with these procedures.
Additional systematic reviews have found the same. Thompson and colleagues, in their research article "A PRISMA systematic review of adolescent gender dysphoria literature: 3) treatment," found that "[t]he evidence base for the outcomes of gender dysphoria treatment in adolescents is lacking." Further, "[i]t is impossible from the included data to draw definitive conclusions regarding the safety of treatment."18
The pattern is clear: the evidence supporting gender transition procedures in minors is at best of very low quality and at worst nonexistent. However, the harms caused by these interventions cannot be ignored. Take cross-sex hormones, for instance. Outside of the cardiovascular risks and various cancer risks associated with taking exogenous hormones, the risk of infertility is real.19,20,21,22,23,24,25,26,27,28,29,30 It is for this reason that even the Endocrine Society's guidelines recommend that "clinicians inform and counsel all individuals seeking gender-affirming medical treatment regarding options for fertility preservation prior to initiating puberty suppression in adolescents and prior to treating with hormonal therapy of the affirmed gender in both adolescents and adults."31
Likewise, the American Academy of Pediatrics' Policy Statement -- in a footnote -- acknowledges that "the effect of sustained puberty suppression on fertility is unknown." It further states that "when cross-sex hormones are initiated without endogenous hormones, then fertility may be decreased."32 Never mind the additional surgical complications that may arise from these interventions, which, in the case of genital surgeries, range from wound-related to urological, including, but not limited to, infection, "wound dehiscence," "trauma from intercourse/dilation," "change in voiding function," and "urethral stenosis."33 Do No Harm's 2025 report "Hormonal Interventions for Minors with Gender Dysphoria Cause Significant Harm" further addresses the risks associated with cross-sex hormones. As the report makes clear, risks to patients' sexual and reproductive systems include not only infertility but pelvic pain and discomfort during sexual activity.34Cardiovascular risks include an increased incidence of heart attacks among women who receive hormonal interventions,35 as well as a seven-fold higher rate of erythrocytosis (increase in red blood cell concentration, which may increase risk of blood clots, leading to heart attack and stroke) compared to men not taking testosterone.36
Women taking testosterone may experience changes that lead to unsatisfactory or harderto-interpret Pap smear tests, which can make cervical cancer screening far more challenging.37 In cases of severe obesity, endometrial cancer (cancer of the uterine lining) appears to increase with the duration of testosterone use.38
Nor do males who take estrogen avoid their own health risks. In addition to fertility consequences, such men face an increased risk of venous thromboembolisms (clots in veins that can pass to the lung and cause death),39 retinal vein occlusions (blockages in blood flow from the eye),40,41 strokes,42 and invasive breast cancer. The latter is 46 times more likely to occur in men who take estrogen than in men who don't.43 Worse, these individuals face a markedly elevated mortality.44
Similarly harrowing on this topic of harms is the U.S. Department of Health and Human Services' 2025 report "Treatment for Pediatric Gender Dysphoria: Review of Evidence and Best Practices." According to that document, endocrine and surgical interventions on minors carry the risk of "infertility/sterility, sexual dysfunction, impaired bone density accrual, adverse cognitive impacts, cardiovascular disease and metabolic disorders, psychiatric disorders, surgical complications, and regret." Moreover, "there has been inadequate research into the frequency and severity of these harms," leaving a gap that raises significant safety concerns.45 Regarding cognitive impacts, it should be noted that there is evidence from both animal and human research that puberty blockers may arrest brain development.46
Despite these documented concerns, organized medicine in the United States has, for the better part of the last decade, largely endorsed and promoted so-called gender-affirming care -- including social transition, puberty blockers, cross-sex hormones, and surgeries -- for minors experiencing gender dysphoria. Major professional organizations such as the American Academy of Pediatrics (AAP),47 the Endocrine Society,48 and the American Medical Association (AMA)49 have followed WPATH50 in issuing guidelines or policy statements framing these interventions as medically necessary and evidence-based standards of care.
This institutional consensus has helped to shape clinical practice, insurance coverage, medical education, and public messaging, often portraying dissent or calls for caution as ideologically driven rather than grounded in emerging evidence.
In 2018, for example, the AAP released a policy statement entitled "Ensuring Comprehensive Care and Support for Transgender and Gender-Diverse Children and Adolescents."51 This document framed skepticism towards pediatric medical transition as pathologizing, urging an affirmation-only approach. It presented affirmation as settled science while downplaying evidence gaps.
But the AAP did not act alone, as The New York Times and other publications have reported.52
In fact, on the issue of pediatric gender transition, the AAP and other major U.S. medical organizations have largely followed the lead of WPATH, which has positioned itself as the leading authority in transgender healthcare, noting that its standards of care are the "foremost evidence-based guidelines for the provision of [transgender and gender diverse] healthcare."53 Its guidelines and its members have held themselves out as the global experts, exerting outsized influence over clinical practice, policy, and litigation across the United States and internationally. In short, pediatric gender medicine is WPATH, and WPATH is pediatric gender medicine.
Accordingly, WPATH's credibility and authority has been invoked by almost two dozen major medical groups, including the AMA, in legal briefs.54,55 And, most significantly, its recommendations have received the imprimatur of the AAP and the Endocrine Society, whose own guidelines on youth gender transition were either written by WPATH members or heavily influenced by the WPATH approach.
Furthermore, to understand WPATH's broad influence, consider that the Standards of Care eighth edition (SOC-8) are referenced in many insurance policies for gender transition procedures on minors.56,57,58 In some states, such as New York, "health maintenance organizations and health insurers must apply utilization review criteria consistent with version of the World Professional Association for Transgender Health (WPATH) Standards of Care when conducting utilization review of treatment for Gender Dysphoria."59 Similarly, some children's hospitals list WPATH guidelines as educational resources.60,61
The organization's influence has been profound in the realm of pediatric medical transition, as the medical community within the United States, and even worldwide, has historically yielded deference to WPATH as the preeminent authority in this field. Through this widespread adoption -- by public and private insurers, healthcare providers, professional societies, and regulatory bodies -- WPATH's standards became the de facto framework guiding care for gender-dysphoric youth across the United States, propping up the guidelines and statements from the Endocrine Society,62 AAP,63 American Psychiatric Association,64 and American Psychological Association.65 WPATH's continuing medical education courses, offered through its Global Education Institute (GEI) Certified Training Courses, have been widely used and have shaped the medical education of countless doctors across the country and planet in the area of youth gender medicine.66 All of this even as questions mounted about the strength of the underlying evidence.
Indeed, the foundation upholding WPATH's recommendations has effectively collapsed. In 2016, WPATH issued a position statement adjudging as "medically necessary" transgender "treatment, sex reassignment, and insurance coverage in the U.S.A."67 This language appears repeatedly in SOC-8, where "medically necessary gender-affirming medical treatment" is detailed for adolescents, as are "aspects of medical[ly] necessary care intended to promote the well-being and gender-related needs of children."68 Yet, as reported in The New York Times, one SOC-8 author noted how casting gender transition procedures as medically necessary could advantageously be "a tool for our attorneys to use in defending access to care."69 In reality there was no true medical necessity; the phrase was a strategic declaration designed to strengthen legal and insurance defenses rather than a conclusion grounded in robust, high-quality clinical evidence, particularly when it came to the care of minors.
By 2024, internal WPATH communications obtained by the state of Alabama through subpoena further underscored this pattern, revealing evidence of serious misconduct in the development of the SOC-8 guidelines.70,71,72,73,74 In a particularly revealing episode, WPATH commissioned The Johns Hopkins University Evidence-Based Practice Center to conduct multiple systematic reviews to inform SOC-8. Systematic reviews are considered the gold standard in evidence-based medicine because they evaluate the totality of evidence and, crucially, examine the underlying methodological quality of the research, allowing for determinations about the quality of evidence.75
However, as reported in The Economist, "documents show that [WPATH's] leaders interfered with the production of [these] systematic reviews." The report continued, "From early on in the contract negotiations, WPATH expressed a desire to control the results of the Hopkins team's work."76 This is not how independent scientific research is conducted. As a result, even the Evidence-Based Practice Center's ostensibly impartial analyses were ultimately constrained by WPATH's requirement that nothing be used without its sign-off. Further, WPATH adopted a policy giving it the authority to "nip papers in the bud on the basis of their conclusions." The result: only one paper from the Hopkins team was published after this policy took hold, despite the group "having provided WPATH with the material for six systematic reviews," according to The Economist.77
Indeed, as revealed by BMJ Investigation, some WPATH SOC-8 authors worried that "independent appraisals of the evidence would undermine legal efforts to protect affirming interventions from legislative restriction in minors." BMJ continued, "a chapter author wrote, 'Our concerns, echoed by the social justice lawyers we spoke with, is that evidence based review reveals little or no evidence and puts us in an untenable position in terms of affecting policy or winning lawsuits.'"78
By August 2020, the Hopkins team reported to the Agency for Healthcare Research and Quality at the U.S. Department of Health and Human Services (HHS) that they found "little to no evidence about children and adolescents" and that WPATH was "trying to restrict our ability to publish."79 WPATH leadership effectively suppressed manuscripts whose conclusions did not align with the organization's preferred positions, successfully pressuring the researchers not to publish their findings. SOC-8 nevertheless proceeded to advance claims of medical necessity, falsely asserting that a systematic review of adolescent outcomes was "not possible."80 Such actions represent a serious breach of scientific integrity and ethical standards in guideline development.
News of WPATH's misconduct later appeared in the mainstream press, including liberal media. "Research into trans medicine has been manipulated," read a headline in The Economist.81 "The World Professional Association for Transgender Health interfered with systematic reviews it commissioned from a research team at Johns Hopkins University," wrote the editorial board of The Washington Post.82 "This is not how evidence-based medicine is supposed to work," reported The Atlantic.83
WPATH's misgivings did not end there. As reported in The New York Times, President Biden's Assistant Secretary for Health, Admiral Rachel Levine, "had been instrumental in WPATH's mysterious last-minute deletion of the age minimums in SOC-8."84 Released in late 2021, WPATH's draft guidelines proposed specific lower age thresholds than had been in previous versions of its Standards of Care: "14 for hormonal treatments, 15 for mastectomies, 16 for breast augmentation or facial surgeries, and 17 for genital surgeries or hysterectomies."85 However, within hours of publishing the final standards of care in September 2022, WPATH deleted from the document age minimums for all hormonal interventions and most surgeries, issuing a "correction," according to The Economist. The news magazine continued, noting that "the head of the drafting committee, Eli Coleman, said the publisher went ahead 'without approval' before final changes were made."86 Thus, the age thresholds that appeared in the draft -- and initially in the published version -- were swiftly excised from the final document.
As emails revealed, this decision followed sustained pressure from Levine, whose political support WPATH needed, and from the AAP. Specifically, Levine's staff urged WPATH to remove these explicit age limits over concerns that they could fuel political opposition and restrict access to care for transgender youth.87 Ultimately, the corrected SOC-8 omitted nearly all age minimums, stating "no recommendations on specific ages," with one exception: phalloplasty, which "WPATH stressed should not be performed under the age of 18 years owing to its complexity."88 These decisions weren't driven by science but, rather, political pressure.
Even some WPATH leaders were upset by this outside meddling. "I don't know how I feel about allowing U.S. politics to dictate international professional clinical guidelines that went through Delphi," wrote one.89 Delphi refers to the Delphi method, a structured, iterative consensus-building process commonly used in healthcare guideline development when high-quality evidence is limited or mixed.90 SOC-8 specifically notes that "[c]onsensus on the final recommendations was attained using the Delphi process that included all members of the guidelines committee and required that recommendation statements were approved by at least 75% of members."91 Clearly, this episode underscores a deeper tension in the development of SOC-8: the gap between the formal consensus process and the final guidelines that emerged after external intervention. Removing age minimums -- after they had already achieved Delphi consensus -- prioritized political considerations over caution in an area where systematic reviews were finding the evidence base to be weak.
Nevertheless, when the AAP threatened to publicly oppose WPATH's guideline unless they got rid of the age minimums, WPATH relented.92 WPATH's president wrote in an email that it was "disappointing that politics always trumps common sense and what is best for patients."93 A former WPATH president involved in the guideline process wrote that the deletion of age minimums "is a balancing act between what I feel to be true and what we need to say."94
Additionally, in WPATH's development of its guidelines, "opaque conflict of interest management" was also evident,95 with significant deviation from standard practice.
Subpoenaed emails and later depositions revealed how WPATH leaders failed to manage conflicts of interest among guideline authors.96,97,98 Nevertheless, the published version of SOC-8 states that "[n]o conflicts of interest were deemed significant or consequential."99 This lax approach manifested in fundamental procedural failures. For one, conflict-ofinterest disclosure forms were not due until after SOC-8 committee members were selected. This information was not part of the selection process. As Leor Sapir noted in his article "The Deposition of Eli Coleman," Coleman, the lead author and chair of SOC-8, "testified that he knew 'most' members of SOC-8 had conflicts." 100
Take, for example, one of SOC-8's authors, a plastic surgeon who has performed multiple "gender-affirming surger[ies]" on pediatric patients.101 That surgeon is now the incoming president of WPATH.102 He declared no conflict of interest in SOC-8. Similarly, in a legal deposition, Marci Bowers, a gender surgeon and former president of WPATH who also authored its guideline, admitted to making over one million dollars from performing gender surgeries in 2023 alone.103, 104 No conflict of interest was declared in SOC-8.
Nevertheless, Coleman and his SOC-8 co-chairs determined that all of these conflicts were "neither 'significant' nor 'consequential.'" As a result, "they thought, WPATH did not need to specify how, if at all, they were managed, and did not need to disclose them in the relevant section of the published guideline."105
Taken together, these revelations illustrate a broader, deeply concerning pattern: WPATH's SOC-8 was driven by ideological advocacy, legal strategy, and political pressure rather than by transparent, rigorous, and independent science -- undermining the very foundation of evidence-based medicine. This seeming malfeasance is further discussed in a United States v. Skrmetti amicus brief that reads as a startling narrative of message-shaping on the part of WPATH rather than scientific prowess.106
Yet, WPATH's influence throughout the halls of pediatric gender medicine reigned supreme.
In its September 13, 2017, press release, the Endocrine Society issued its guidelines for "gender-affirmation treatment," officially listing WPATH as a co-sponsor.107 In fact, many of the Endocrine Society guideline authors on the same topic were either WPATH members or had other close ties to the organization.108
This interconnected web of guidelines promoting gender transition procedures for minors extended even further. Jo Taylor and colleagues have shown that gender medicine guidelines repeatedly drew upon one another, with many "cit[ing] and draw[ing] on the World Professional Association for Transgender Health (WPATH) guidelines."109 The result was a self-reinforcing cycle in which position statements and guidelines repeatedly referenced one another, creating the appearance of a robust evidentiary foundation in pediatric gender medicine -- one that, through repetition across the medical literature and within organized medicine, came to seem unquestionable.
Under these conditions, the U.S. health system effectively institutionalized gender transition procedures for minors as standard care. Major medical organizations, hospitals, and insurers treated WPATH's SOC-8 and the closely aligned Endocrine Society guidelines as authoritative, leading to widespread adoption of puberty blockers, cross-sex hormones, and surgeries -- often with minimal gatekeeping. This occurred even as rigorous, goldstandard evidence reviews -- including the Cass Review in the UK and multiple European systematic reviews -- concluded that the benefits of these interventions were of very low certainty, while the risks were substantial.
Yet the past makes clear that medicine is not immune to profound error -- and, at times, to grave wrongs. History offers sobering reminders: physician-endorsed smoking,110 the brutality of lobotomies,111 the use of diethylstilbestrol (DES) in pregnancy,112 the Tuskegee Syphilis Experiment,113 and, most recently, opioid overprescribing and the resulting crisis114 -- to name just a few examples. The record of modern medicine is marked by episodes of misjudgment and sometimes serious scandal. Entire books and scholarly analyses have been devoted to understanding how harmful practices are introduced, or why ineffective and unsafe treatments persist long after evidence has shown they should be abandoned. As physicians, we hold the capacity to do immense good -- but also, if we are not vigilant, immense harm.
Our profession is far from infallible. Yet it is not the absence of error that earns the public's trust in medicine. Rather, it is our willingness, as physicians, to confront mistakes honestly, to reassess our assumptions, and to correct course when the evidence demands it.
For this reason, open inquiry, rigorous debate, peer review, critical appraisal of evidence, and independent evaluation are indispensable. These commitments gave rise to evidence-based medicine -- a systematic approach to clinical decision-making grounded in the best available scientific data.
Yet today, in the case of pediatric medical transition, empirical evidence has been systematically sidelined in favor of ideology. Such widespread politicization has prevented these self-corrective processes from operating as intended. Self-appointed experts have broken the chain of trust, and the medical community has been misled by ostensible authorities who have minimized or, in some cases, veiled altogether the gaps in their knowledge.
Gradually, however, the tide has begun to turn. Given the weak quality of evidence and the obvious harms of gender transition procedures on minors, countries throughout the world, as well as 27 states here at home, have taken a stance to protect children by stopping many of these interventions from occurring. In May 2024, England, for example, banned the clinical use of puberty blockers in those under 18 years of age.115 This followed publication of the Cass Review, whose methodology was recently found to be "robust" by the British Medical Association's board of science.116 That ban on puberty blockers had the support of Conservative and Labour governments alike.117 And just a few months ago, BMJ reported that "NHS England has 'paused' the prescribing of gender affirming hormones to treat gender incongruence or dysphoria for young people under 18."118
Likewise, in recent years Sweden and Finland -- among the world's most socially progressive nations -- have imposed limits on gender transition procedures for minors, all of them initiated under left-leaning governments.119 Their actions were driven by evidence -- or, more precisely, by the lack of evidence supporting so-called gender-affirming care.
Systematic reviews prompted these course corrections. In Sweden's review, for example, the evidence for hormone treatment in minors was deemed "insufficient," and the long-term psychosocial effects of hormone therapy were described as "unknown." The authors further classified puberty blockers as an "experimental treatment."120 Today, nearly two dozen systematic reviews have all consistently demonstrated the absence of a credible evidence base.121
An increasing number of medical professionals are also coming to recognize that pediatric gender transition is the medical scandal of our generation. In my role at Do No Harm, I hear this regularly. Over the past year alone, our organization has grown substantially as more and more clinicians have become aware of the risks associated with youth gender
transition. We routinely receive messages from physicians who know this to be a scandal, but who feel unable to voice their concerns within their professional circles for fear of personal or professional reprisal. This climate of silence creates the misleading impression that most physicians support these interventions. They do not. In a survey of Florida physicians, for example, we found that 66 percent supported the state's prohibition on gender transition procedures for minors.122
In fact, in February of this year, the American Society of Plastic Surgeons (ASPS), representing over 11,000 surgeons,123 became the first major U.S. medical society to recognize the grave concerns associated with gender transition procedures on minors.
ASPS's position statement openly acknowledged the lack of supportive evidence for these procedures, explicitly calling out "gender-related endocrine and surgical interventions." It therefore recommended against gender related surgeries in anyone under 19./124
Additionally, ASPS emphasized an often-ignored truth: "the natural course of pediatric gender dysphoria remains poorly understood."125 For one, clinicians cannot reliably predict which children will persist in their distress and which will desist. This uncertainty is not a minor footnote. The 2025 guidelines from Germany recognize this fact, providing no criteria to differentiate between those children with "temporary 'gender non-contentedness'" and "stable/persistent cases."126 ASPS was thereby right to recognize that, when a condition's trajectory is uncertain and the interventions irreversible, medical ethics requires caution, not acceleration. This applies to both surgeries and medications, including puberty blockers and cross-sex hormones, whose long-term physiological effects and side effects cannot be fully undone.
Moreover, ASPS's position statement sharpened the proper role of autonomy in clinical ethics. Patient autonomy does not obligate physicians to provide interventions when the risk-benefit profile is uncertain or unfavorable -- particularly for minors whose decisionmaking capacity is still maturing. Medicine must move away from the "have it your way" model of care that has failed children with gender dysphoria, fast-tracking them to hormones and surgeries despite systematic reviews showing very low evidence of benefit and significant risk of harm.
This is especially true given the ethical concerns raised by the impossibility of informed consent (or even informed assent) on the part of adolescents whose brains have not yet fully developed. As ASPS's position statement points out, "gender-related surgery procedures intervene directly in the processes of identity formation and psychosexual development.
These are areas of ongoing maturation during adolescence that warrant particular ethical caution as surgeons assess adolescent medical decision-making capacity."127 The same can be said for puberty blockers and cross-sex hormones, which similarly disrupt critical windows of brain and bodily development during a period when true informed decisionmaking capacity remains fundamentally limited.
All of this comes on the heels of the comprehensive umbrella review conducted by the U.S. Department of Health and Human Services, which itself has undergone peer review and provides a methodologically sound and transparent assessment of the evidence.
The HHS Review makes it clear: the quality of evidence regarding key domains such as psychological outcomes and quality of life, as well as long-term health, is very low. The benefits simply aren't there. Yet, the evidence for harms clearly exists and can be drawn from "established knowledge about human physiology and the effects and mechanisms of the pharmacological agents used."128 These effects are real and can't be ignored.
In coming to these conclusions, the HHS Review performed a substantial evidence review.
Following the recommendations for overviews of systematic reviews as per the Cochrane Handbook for Systematic Reviews of Interventions, thousands of studies were screened, with ultimately 17 systematic reviews included. As noted in the HHS Review, "two reviewers independently screened titles and abstracts, and then full texts, to determine study eligibility." The Risk of Bias Assessment Tool for Systematic Reviews (ROBIS) was used to assess the included systematic reviews. Evidence synthesis was performed and organized by outcomes. The effect estimates and quality of evidence followed the GRADE methodology.129
The HHS Review, however, didn't stop there. It included evidence from basic science and physiology to help best understand the harms of puberty blockers and cross-sex hormones.
It reviewed international clinical-practice guidelines, utilizing key systematic reviews. It also added a critically important chapter on ethical considerations as well as psychotherapy as a means of caring for children suffering from gender dysphoria.
Having released an early version of their Review in May of last year, the HHS Review team invited skeptical physicians and medical associations to scrutinize their conclusions. It invited critics to offer a peer review, specifically reaching out to groups such as the AAP, the Endocrine Society, and the American Psychiatric Association (APA). Regrettably, all three organizations have strongly supported gender transition procedures on minors, and they vehemently opposed HHS's initial review in May 2025, with the AAP condemning the Review within hours of its May 1 release.130
Yet, despite HHS's attempt to engage them, the AAP and the Endocrine Society declined to provide a peer review. That choice speaks for itself: when invited to present evidence supporting medical interventions for minors, they opted not to participate. If they had strong, well-supported arguments, they would have taken the opportunity to lay them out and challenge the findings in the HHS Review. They didn't.
Only the APA, to its credit, submitted a peer review. Still, the review offered very little meaningful critique. The APA even appeared to overlook key sections of the Review, asserting that HHS had failed to outline its methodology when the Review clearly did so. In the end, the APA identified no substantive flaws in the HHS findings and raised no objections to the Review's treatment of psychotherapy as an alternative to hormonal interventions. At the end of the day, none of the critical reviewers disproved the central conclusions of the HHS Review. Please read those critiques for yourself, and you'll see that the HHS Review withstood peer-review scrutiny, earning its place as a robust umbrella review.
Put simply, the latest HHS Review is the most comprehensive and evidence-based review of gender transition procedures on minors. It lends support to the many states across our country that have taken a stance to keep kids safe from these interventions, for which the evidence of benefit is lacking and the harm is unfortunately lasting and significant.131 Such laws echo common sense and public sentiment. In a New York Times/Ipsos survey from last year, 71 percent of respondents agreed that "no one under age 18 should have access [to puberty-blocking drugs or hormone therapy]" when used for transgender care.
Among Democrats, only 19 percent felt minors as young as 10 should be allowed access to puberty blockers for these purposes, while 54 percent said no one under 18 should have access.132
It is also important to recognize that many children later come to regret their transition and themselves detransition years later. By one study's account, at least 30 percent of youth discontinue medical transition within just a few years, but many of the effects of hormones and surgeries are irreversible.133 While low regret from pediatric medical transition is frequently cited in the popular press, this conclusion is based on flawed studies with narrow windows of time for follow-up, high rates of attrition (i.e., individuals "lost to follow-up" who simply stopped coming to their appointments), and biased samples or those that cannot be generalized.134 In reality, the true rate of regret is not known.135
It is in part for this reason that Do No Harm recently urged the Centers for Disease Control and Prevention to create new medical diagnosis codes (known as ICD-10 codes) for desistance and detransition.136 Currently, the lack of a specific code for detransition makes those who have detransitioned effectively invisible to the healthcare system, obscuring their medical needs and limiting the real-world research, follow-up care, and outcome monitoring that responsible practice requires.
Moreover, many within the current cohort of gender-dysphoric youth differ markedly from earlier ones. In recent years there has been a dramatic surge in adolescents -- predominantly female adolescents -- with "rapid-onset gender dysphoria," as described by Lisa Littman in 2018. These youth, often influenced by peer groups and social media, frequently suffer from significant psychiatric or neurodevelopmental comorbidities in addition to their gender confusion.137,138 It is these co-morbid mental health conditions that should be the focus of treatment.
Instead, the rush has been to treat gender dysphoria with an affirming model inclusive of medications and surgeries. In many cases, supply has unfortunately created its own demand: simply being seen by a gender specialist is itself a strong predictor that a child will proceed with social and medical transition.139
By contrast, a major follow-up study of boys diagnosed with gender identity disorder found that more than 87 percent desisted by early adulthood -- that is, they no longer experienced gender dysphoria.140 Many of these boys later identified as homosexual. Consistent with this, the DSM-5-TR, last published in March 2022, notes that the rate of persistence for childhood gender dysphoria into adolescence or adulthood ranges from only 2.2 percent to 50 percent, meaning the clear majority of cases resolve on their own or desist.141
Desistance is real, and psychotherapy can be genuinely beneficial to youth struggling with gender confusion, particularly when depression, suicidal thoughts, and self-harm coexist.
These children need high-quality behavioral healthcare, not gender transition procedures.
Finland, for example, has recognized this explicitly, recommending that "first-line treatment for gender dysphoria is psychosocial support and, as necessary, psychotherapy and treatment of possible comorbid psychiatric disorders."142
For the sake of our children, it is time to bring an end to gender transition procedures for minors. The evidence base is weak, with systematic reviews finding no reliable data supporting these interventions. It is time for pediatric gender medicine to move beyond WPATH's SOC-8, a document shaped by opaque processes, marked by conflicts of interest, and influenced by ideological advocacy that has sacrificed science for activism, resulting in the reckless promotion of low-evidence, high-risk medical interventions on minors. Instead, we must acknowledge the certainty of harms done by pediatric medical transition alongside the uncertainty of any benefit. No responsible field of medicine would continue to promote interventions where the harms are well-established and the benefits remain unproven.
If one wishes to look beyond the evidence, then please listen to those who have detransitioned; their stories are powerful reminders of the compassion we must have in caring for youth with gender dysphoria, as well as the resolve we need to first and foremost end these experimental gender transition procedures on minors.
To my fellow physicians: Many of us have relied on colleagues regarded as experts in gender medicine. That is common in a field as complex as the medical profession -- no one can master every specialty. But our responsibility is not solely to trust, or defer, but also to verify. Given the documented harms and the vulnerability of the children involved, unquestioning trust is no longer acceptable. I urge my peers to look beyond partisan narratives and examine the evidence directly. If we hope to restore the public's confidence in our profession, we must begin by acknowledging that we have made mistakes -- grave ones - and that the time to correct them is now.
It is time for the medical profession to return to the evidence. It is time for us to stand with the nation in protecting vulnerable children and ending procedures that carry irreversible harm. It is time to enshrine in law a clear commitment to safeguarding children who are struggling with gender dysphoria and related mental-health challenges. They deserve quality therapy and robust psychosocial support -- not irreversible, life-altering medical interventions. Above all, we must remember our duty: First, do no harm.
Thank you for your time and for reading my testimony. I look forward to further discussing this important matter with the Committee.
* * *
Original text here: https://www.help.senate.gov/imo/media/doc/836dcbdb-dc18-2dcd-e758-5033879edf61/Miceli%20Testimony_481e9f75-cdd8-4a38-966b-fbcb2fda282c.pdf