Google apologizes for flawed AI image-generator, acknowledges overcompensation issues

CALIFORNIA: Google issued an apology on Friday for the imperfect launch of its new artificial intelligence image-generator, admitting that the tool, in some instances, would “overcompensate” in seeking a diverse range of people even when it didn’t align sensibly.

This statement followed Google’s decision to temporarily halt its Gemini chatbot from generating images with people after social media users claimed it exhibited an anti-white bias. Users noted the tool’s generation of racially diverse images in response to prompts, often placing people of color in historically inaccurate settings.

Prabhakar Raghavan, a senior vice president at Google, acknowledged the feature’s shortcomings in a blog post, stating, “It’s clear that this feature missed the mark. Some of the images generated are inaccurate or even offensive. We’re grateful for users’ feedback and are sorry the feature didn’t work well.”

Specific examples weren’t mentioned, but social media highlighted instances like depicting a Black woman as a U.S. founding father and portraying Black and Asian individuals as Nazi-era German soldiers.

The new image-generating feature was integrated into the Gemini chatbot about three weeks ago. Originally known as Bard, it was constructed upon Google’s earlier research experiment, Imagen 2.

Generative AI tools, as highlighted in a 2022 technical paper by Imagen’s developers, can pose concerns related to social and cultural exclusion, bias, and misinformation. Google’s decision not to release Imagen publicly was influenced by these considerations.

Raghavan addressed the issues, stating, “When we built this feature in Gemini, we tuned it to ensure it doesn’t fall into some of the traps we’ve seen in the past with image generation technology — such as creating violent or sexually explicit images or depictions of real people.”

Gemini faced criticism for refusing to generate images on certain topics, such as protest movements, with claims of causing “trivialization of sensitive topics.”

Elon Musk, owner of social media platform X (formerly Twitter), amplified the outrage, condemning Google for what he called “insane racist, anti-civilizational programming.” Musk, a vocal critic of AI development, has accused others of bias in the industry.