Men Can't Stop Fabricating Their AI Credentials

More Than Half of the Men Studied in New Harvard Research Claimed to Know AI Leaders Who Are, in Fact, Entirely Fictional

More than half of the men surveyed in a new Harvard study claimed to know AI leaders who are actually completely fictional.

Say you're a guy who reads WIRED — you know a thing or two about AI. You know the names of famous people who develop artificial intelligence, like Sam Altman, CEO of OpenAI. But do you know who Mira Moratti is? What about Susan Milner? Craig Dalton?

In a new poll of more than 5,400 Americans, 73 percent of male respondents told Harvard researchers who Altman was. Just over half - 53 per cent - said they knew Moratti, Milner and Dalton.

There's a problem. Moratti is a real person: He's the former CTO of OpenAI and current CEO of Thinking Machine Labs. Milner and Dalton, on the other hand, are absent.

Compared to women, "we find that men are more optimistic and confident" about AI, Laura Manley, executive director of the Shorenstein Center at Harvard Kennedy School, tells WIRED. Sometimes, this veers into overconfidence – such as claiming to identify fictional AI executives. (For context, 40 percent of women claimed to know who Milner and Dalton were; 38 percent said they knew Moratti.)

While Manley notes that men are generally more overconfident when it comes to scrutiny, the report, released Wednesday, shows what the AI confidence gap looks like, as well as the consequences of gender gaps.

The researchers found that men use AI more often and, on average, are better able to answer real-world questions about AI. But men and women answer incorrect questions at the same rate. The difference is based on the number of times women said they were unsure of an answer - an average of 3.3 out of 12 questions compared to 2.3 for men.

The researchers wondered if this desire for certainty colored the veneer of artificial intelligence. Over a year of media coverage, men's names were more than 14 times more than women's in AI-related headlines. Because AI CEOs are overwhelmingly male, the researchers decided to re-examine the coverage and look only at those listed as subject matter experts. Three-quarters of these experts were male—a higher proportion than the actual AI talent, which is approximately 29 percent female.

"Maybe there are more women emerging as experts or a platform to express: I'm not sure," says Manley.

In addition to brainstorming, Americans generally agreed that artificial intelligence terrifies them—both men and women rated "concern" as their primary emotional response. However, female respondents were more anxious, fearful, and worried than their male counterparts. Meanwhile, men rated every positive emotion—including curiosity, enthusiasm, and excitement—higher than women.

"In the current discourse," says Manley, "there is a reliance on doom or abundance." "It just silences all the voices that need to say 'I'm not sure.'" Questions about the proper use and future of AI. One gender is simply less likely to accept such uncertainty.

"We need to be able to say 'I'm not sure' more often," says Manley.