Google may have deeply integrated AI into most of its businesses, from AI-powered overviews in search to AI summaries for YouTube videos, but the company’s CEO is warning against “blindly trusting” and solely using AI to obtain information. 

In a new interview with the BBC published on Tuesday, Google CEO Sundar Pichai told the public not to believe everything AI tells them, saying that despite Google’s best efforts, AI models are still “prone to some errors” and should be balanced with other information sources, such as those found in a standard Google search.

“This is why people also use Google search, and we have other products that are more grounded in providing accurate information,” Pichai told the BBC.   

Related: This Is the Biggest Bet Google Is Making, According to Its Chief Investment Officer

Pichai stated that it is important to know which tasks AI models excel in, giving the example of creative writing as something AI is good at.

Google includes disclaimers on its AI tools, such as the one for AI overviews, which indicate that “AI responses may include mistakes.”

When Google first rolled out AI overviews in May 2024 to its 250 million-plus monthly U.S. users, its AI search results were subjected to scrutiny for telling users to eat rocks and make pizza sauce with “non-toxic glue.” 

Anastasia Kotsiubynska, head of SEO at SE Ranking, foreshadowed the hallucinations in a remark shared with Entrepreneur before the AI overviews rollout: “Most likely, there will still be misleading information in search results and hallucinations, and many users will probably use this information without double-checking,” Kotsiubynska said.

Google holds just over 90% of the search engine market share worldwide as of last month, according to StatCounter. 

Related: These Are AI’s ‘Most Obvious’ Risks, According to Google’s Former CEO

Google remains a dominant force in the AI market, leading the way with massive investments and consumer-focused AI products. For example, Google released its latest AI model, Gemini 3 Pro, on Tuesday to power “AI mode” in Google Search and Gemini.

Google Gemini ranks as the third-biggest AI chatbot globally, with a 13.4% market share as of this month, behind OpenAI’s ChatGPT and Microsoft Copilot. Meanwhile, Google’s parent company, Alphabet, set a record $75 billion in capital expenditure in 2025, mainly for AI infrastructure. 

In his BBC interview, Pichai talked about the balance between how quickly new technology is rolling out and how important it is to build in protections to keep people safe. He said that for Google, handling this challenge means moving quickly and taking risks, but always with caution and responsibility in mind.