Former Google Safety Chief Warns AI Poses Greater Risk to Children Than Social Media

A former top executive at Alphabet's Google has issued a stark warning that the artificial intelligence industry is repeating the mistakes of the social media era, potentially causing even greater harm to children. Tom Siegel, the founder of Google's trust and safety team, called for a slowdown in AI development, stating that the industry's belief it can self-regulate is "unrealistic," according to comments made to Reuters this week.

A Stark Warning from a Tech Veteran

Siegel, who served as Google's vice president of trust and safety from 2004 to 2019, expressed growing concern that AI poses a deeper threat to young people than social media platforms, which are already facing thousands of lawsuits and global regulatory pressure over alleged harm to minors. He argued that major AI companies are deploying advanced features to children without adequate guardrails in place.

"We find ourselves in a really terrible situation in terms of what AI is doing to kids," Siegel told Reuters. He warned of "very, very severe scenarios of harm," citing risks that include suicide, psychosis, and cognitive offloading—a term for decreased critical thinking due to over-reliance on AI.

New Role to Drive Accountability

To address these concerns, Siegel announced on Tuesday he has joined the nonprofit advocacy group Common Sense Media as the first executive director of its new Youth AI Safety Institute. The research organization, backed by industry players including Anthropic and the OpenAI Foundation, aims to develop independent standards to assess the risks of AI tools for children, similar to crash testing for vehicles.

Ad

Siegel said his goal is to collaborate with AI firms and third-party groups to create accountability standards. He left Google (GOOGL) in 2019 after concluding the tech industry was "plateauing" in its ability to ensure online consumer safety.

Industry Response and Proposed Safeguards

Siegel outlined immediate steps AI companies could take to better protect young users, even if it affects short-term usage or profits. He suggested that inaction risks more severe consequences from future lawsuits and regulation.

• Anthropic and OpenAI could implement more robust age verification to prevent underage access to their chatbots.
• Google could add a simple toggle to turn off its AI Overviews and AI Mode in search, similar to existing filters for explicit content.

In response to the comments, a Google spokesperson told Reuters that its parental controls allow for turning off a child's access to its search engine entirely. An OpenAI spokesperson stated the company uses "account and behavioral signals" in addition to self-reported age to identify underage users and is continuously improving its systems. Anthropic did not immediately provide a comment.