"I'll have to go without a new LinkedIn profile photo for now!"

Race Matters

While messing with an AI image generator, an Asian-American MIT grad found that the bot's way of professionalizing her headshot was to make her... white.

In an interview with Insider, 24-year-old Rona Wang said prompted the Playground AI image editor to make her headshot seem more "professional" — and instead, it lightened her skin and, well, basically changed her race.

https://twitter.com/ronawang/status/1679867848741765122?s=20

CEO No

Unsurprisingly, the post went viral, garnering a quick response from Playground AI's CEO Suhail Doshi, and soon after, an interview with the Boston Globe about the debacle. Futurism and Insider both reached out to Playground for comment without response, though the CEO did respond to the Globe's interview questions with a few rhetorical ones of his own.

"If I roll a dice just once and get the number 1, does that mean I will always get the number 1?" Doshi quipped. Should I conclude based on a single observation that the dice is biased to the number 1 and was trained to be predisposed to rolling a 1?"

Unbearable Whiteness

Though Wang had been using AI image generators for some time — her commentary about it was even republished on MIT's admissions blog just a few weeks before the Playground incident — this time seemed much more personal.

"I was like, 'Wow, does this thing think I should become white to become more professional?'" she told the Globe, adding that she hopes that the "people who are making software are aware of these biases and thinking about ways to mitigate them."

In that initial interview, Wang said that she didn't "want to jump to conclusions that this AI must be racist," but in her latest comments to Insider, it seems like her views may have changed.

"Racial bias is a recurring issue in AI tools," she said. "I haven't gotten any usable results from AI photo generators or editors yet, so I'll have to go without a new LinkedIn profile photo for now!"

While she's far from alone either in having an image generator turn her white or in getting a racist output in general, this flash-in-the-pan controversy is yet another example of AI messing up hugely — even as companies rush to replace human workers with this very imperfect and imprecise software.

More on AI fails: When AI Is Trained on AI-Generated Data, Strange Things Start to Happen


Share This Article