AI chatbots are overconfident - like humans
October 05, 2026
October 05, 2026
BERKELEY, California, Oct. 5 -- The University of California Haas School of Business posted the following news release:
* * *
AI chatbots are overconfident--like humans
*
New UC Berkeley Haas research testing 11 leading large language models--including models from OpenAI, Google, Anthropic, and Meta--found they were overconfident on average: the models reported 88% confidence but were correct only 79% of the time, a nine-point overconfidence ga . . .
* * *
AI chatbots are overconfident--like humans
*
New UC Berkeley Haas research testing 11 leading large language models--including models from OpenAI, Google, Anthropic, and Meta--found they were overconfident on average: the models reported 88% confidence but were correct only 79% of the time, a nine-point overconfidence ga . . .