Men Can't Stop Faking How Much They Know About AI

A large Harvard Kennedy School survey of more than 5,400 Americans documents a consistent gender gap in confidence about artificial intelligence: men report recognizing AI leaders and claim familiarity with the field at substantially higher rates than women, even when some named figures are fictional. For example, 73% of men said they knew OpenAI CEO Sam Altman and about 53% said they knew people listed as AI executives - yet two of those names were invented while Mira Murati, the real former OpenAI CTO, was among the few genuine figures. Men use AI more, answer factual AI questions correctly more often, and express greater optimism, but they also guess rather than admit uncertainty: men declined to say “not sure” on average 2.3 times out of 12 questions versus 3.3 for women, while overall error rates were similar across genders.

The study connects those confidence differences to how AI gets covered and discussed: men appear in AI headlines more than 14 times as often as women, and 75% of quoted subject experts were male even though the AI workforce is roughly 29% women. Both men and women list concern as their top emotion about AI, but women register higher fear and worry while men report more curiosity and enthusiasm. The researchers argue that male overconfidence crowds out voices willing to acknowledge uncertainty, and they recommend elevating more women and more “not sure” perspectives to balance public debate and reporting on AI’s future.