Google CEO Bowed His Head: "Gemini AI Errors Are Unacceptable"

Sundar Pichai, CEO of Google, recently addressed the image generation malfunction in the company's AI model Gemini, calling it "completely unacceptable" and stating, "It is our fault."

[Image source=AFP Yonhap News]

According to the economic media CNBC on the 28th (local time), CEO Pichai said in an email sent to employees the previous afternoon, "We want to resolve the recent issues with Gemini's text and image generation," emphasizing this point. He acknowledged, "We know it caused discomfort to users and revealed biases," and stressed, "To be clear, this is completely unacceptable and it is our fault."

He continued, "Our team is working around the clock to fix these problems," adding, "There is no perfect AI at the early stages, but we know the standards are high. No matter how long it takes, we will continue to uphold these standards." Google suspended the image generation feature of Gemini on the 22nd, just about 20 days after adding the function.

Previously, Gemini faced criticism for distorting historical facts, such as depicting historical figures like Einstein as people of color or generating German Nazi soldiers as Asian. This backlash quickly spread on social media (SNS), and Alphabet’s stock price plunged on the New York Stock Exchange. Google also faced difficulties last year during the launch of Bard when incorrect answers appeared during demonstrations of its features.

CEO Pichai emphasized, "We have always strived to provide users with helpful, accurate, and unbiased information about our products. That is why people trust us, and this should be our approach to all our products, including our new AI products."

This content was produced with the assistance of AI translation services.