Ex-Google safety chief warns AI could harm children more than social media did

Former Google safety chief Tom Siegel has called for a slowdown in AI development, warning that the technology industry is repeating the same mistakes made by social media companies and potentially causing greater harm to children. Social media companies have faced massive lawsuits and regulatory scrutiny for harming young users, while governments worldwide are working to curb minors' access to these platforms.

However, AI's rapid development has outpaced regulation. Siegel, now the first executive director of the Youth AI Safety Institute at Common Sense Media, expressed concern that AI poses a deeper threat to children than social media did. He highlighted potential risks such as suicide, psychosis, and cognitive offloading, which refers to decreased critical thinking due to reliance on AI for answers.

Siegel suggested that companies like Anthropic and OpenAI Foundation could implement robust age verification measures and disable AI features for underage users. Google and OpenAI have indicated they are taking steps to address these concerns, but Siegel argued that immediate action is necessary to avoid potential lawsuits and stricter regulation.

As Common Sense Media's executive director, Siegel aims to establish independent standards similar to car crash testing to measure AI tool risks and increase collaboration between AI companies and third-party groups.