Google Search head encourages innovation despite errors in AI-generated results
Liz Reid, the head of Search at Google, recently addressed the company's employees about the ongoing development of artificial intelligence (AI). During a company-wide meeting, Reid acknowledged that while AI may still make mistakes, it's crucial to continue advancing its development, as reported by CNBC. She encouraged her colleagues not to view these issues as setbacks but as opportunities for improvement. "It is important that we don't hold back features just because there might be occasional problems," Reid said.
Reid's humorous take on AI errors
Reid, who assumed her leadership role in March, has had to address the tendency of Google's AI search to produce nonsensical results. She humorously referred to instances where Google's AI Overview feature gave absurd responses to simple queries, such as suggesting putting glue on pizza or eating rocks. Despite these errors, Reid encouraged risk-taking in the pursuit of innovation. "I don't think we should take away from this that we shouldn't take risks," she said during the meeting.
Google's response to AI errors and user feedback
Google has reportedly implemented patches as well as guardrails to address issues with its AI-generated search results. Despite these measures, some users were still receiving absurd responses from the AI Overview feature until recently. When reached by CNBC, a Google spokesperson defended the tool's performance, stating that the "vast majority" of AI Overview responses were accurate and issues were found in "less than one in every 7 million unique queries on which AI Overviews appeared."
Reid emphasizes continuous improvement
Reid acknowledged that despite extensive testing and efforts to identify vulnerabilities before launching AI products, more work is needed. She highlighted challenges related to understanding the quality and context of webpage content, and urged employees to report issues they encounter. "Anytime you see problems, they can be small, they can be big — please file them," Reid urged her fellow Googlers.
Google's commitment to refining AI products
Google has faced criticism for inaccuracies and biases in AI products, such as its Gemini chatbot and image generation tool. Despite these setbacks, Reid's stance indicates that Google is determined to continue refining its AI offerings based on user and employee input. She noted that the firm has made more than a dozen technical improvements to the AI Overview tool, and is considering additional measures to address intentionally adversarial user queries, and fake AI Overviews created for social engagement.
