Google CEO Sundar Pichai addressed the company’s Gemini controversy Tuesday night, calling the company’s AI app’s problematic handling of race issues unacceptable and calling for structural changes to resolve the issue. He made it clear that he would make changes.
Google says its Gemini image creation tool sometimes refuses to depict white people, or inserts photos of women or people of color when asked to create images of Vikings, Nazis, or the Pope. , suspended that tool last week because it produced perplexing and offensive results.
Controversy swirled when Gemini was found to be making questionable text responses, such as equating Elon Musk’s influence on society with that of Adolf Hitler.
These comments drew sharp criticism, especially from conservatives who accused Google of anti-white bias.
Most companies offering AI tools like Gemini have created guardrails to mitigate fraud and avoid bias, especially considering other experiences. For example, image generation tools provided by companies such as OpenAI have been criticized for creating images of primarily white people in professional roles and depicting black people in stereotypical roles.
“We know that some of those responses were offensive and showed bias. Let’s be clear: It’s completely unacceptable and we made the wrong decision.” Pichai said.
Pichai said the company is already making progress in repairing Gemini’s guardrails. “Our teams are working around the clock to address these issues. We are already seeing significant improvements across a wide range of prompts,” he said.
Google has reviewed the memo, and the full text of Pichai’s memo is below.
