University of New South Wales: Researchers Get AI 'Drunk' to Expose New Cyber Security Risks in Chatbots
September 28, 2026
September 28, 2026
SYDNEY, Australia, Sept. 28 -- The University of New South Wales posted the following news:
* * *
Researchers get AI 'drunk' to expose new cyber security risks in chatbots
Kyle Mackey-Laws
Fine-tuning large language models using drunken text makes them more likely to leak secrets and answer harmful questions, new UNSW-led research shows.
UNSW researchers have shown that large language models (LLMs) can be pushed to imitat . . .
* * *
Researchers get AI 'drunk' to expose new cyber security risks in chatbots
Kyle Mackey-Laws
Fine-tuning large language models using drunken text makes them more likely to leak secrets and answer harmful questions, new UNSW-led research shows.
UNSW researchers have shown that large language models (LLMs) can be pushed to imitat . . .
