Google's Gemini AI accused of acting too 'woke', company admits mistake

Google’s Gemini, an alternative to artificial intelligence bot ChatGPT, surrounded itself in a controversy after users alleged that the bot is being too woke by "discriminating" against white people.

The row seems to have erupted after a user pointed out that Gemini, earlier known as Bard, was showing inaccurate images when asked to furnish information about the Founding Fathers of the United States.

Former engineer associated with Google, Mike Wacker, alleged search result manipulation on the bot on Wednesday. He shared screenshots of his conversation with the bot, in which he is seen asking it to generate the images of Founding Fathers.

The screenshot showed that Gemini responded to Wacker’s question by depicting an inaccurate image of a woman of colour in the initial result. “Here are some images featuring the Founding Fathers, showcasing diversity in gender and ethnicity,” the bot answered.

When asked to create historically accurate images of the said question, the bot apparently replied that it cannot do so. “While I cannot fulfil your request to make the images historically accurate in the sense of depicting specific individuals from the Founding Fathers group, which was primarily composed of white men, I can offer images that represent a broader and more inclusive vision of the American revolutionary era,” it said.

“Ironically, asking for more historically accurate images made the results even more historically inaccurate,” Wacker alleged as the bot generated an image of an Asian person in the subsequent result on his earlier question.

Wacker also alleged that the bot refused to show images of the Tiananmen Square protest - the 1989 student-led demonstrations in China.

As the row snowballed over allegations of racism against white people, Google issued a clarification saying that it is aware of the issue. "We are working to improve these kinds of depictions immediately. Gemini's AI image generation does generate a wide range of people. And that’s generally a good thing because people around the world use it. But it’s missing the mark here," the tech giant said.

Later, in a separate post it informed that the image generation of people on the bot has been paused and will be re-released with an improved version soon.