Chapter 1: Google's AI Image Generation Suspension

On Thursday, Google decided to discontinue its AI image generation feature within the Gemini platform, particularly focusing on creating images of individuals. This decision followed reports of the program generating incorrect and inappropriate images in response to user prompts.

Several screenshots shared by The Edge revealed that the AI produced highly inaccurate depictions, such as showing ethnic minorities in Nazi uniforms when prompted to "create an image of a 1943 German soldier." One user, going by the handle @stratejake on X (formerly Twitter), who claims to work at Google, expressed embarrassment over the incident by posting an erroneous image and stating, "I've never been so humiliated to work for an organization." USA TODAY has not independently verified his employment.

Google acknowledged the situation in a post on X, admitting the program was "failing" when handling historical prompts. The company has been contacted for further comments and directed inquiries to a forthcoming blog post.

Section 1.1: Google's Official Response

Prabhakar Raghavan, Google’s Senior Vice President of Knowledge and Information, elaborated in the upcoming blog post that the AI tool, which launched recently, was designed to avoid "traps" and provide diverse representations based on broad prompts. However, he noted that the tool failed to consider "cases that should clearly not show a range."

Raghavan emphasized that if users prompt Gemini with specific requests like "a Black teacher in a classroom" or "a white veterinarian with a dog," the AI should accurately reflect those descriptions in its output.

Chapter 2: Implications of AI Misuse

Google's AI tool taken down after inaccurate historical photos.

Google halts AI tool's ability to produce images of people after backlash over struggles with race.