Ex-Google Safety Chief Warns AI Could Harm Children More Than Social Media

Ex-Google safety chief Tom Siegel has warned that artificial intelligence could cause more harm to children than social media did in the past. Several social media companies have faced massive lawsuits and heavy fines for allegedly harming young users, prompting governments to step in and regulate access. However, Siegel argues that the tech industry is repeating the same mistakes with AI, potentially causing even greater harm to children.

He urged for a slower pace of AI development, warning that the industry is "figuring it out by itself without outside help is unrealistic."

Siegel recently joined Common Sense Media, a non-profit organization that conducts research and reviews the suitability of various media for children. He will serve as the first executive director of the Youth AI Safety Institute, a research organization funded by Anthropic and OpenAI Foundation, to assess kids' use of AI. Siegel plans to develop standards to hold AI firms accountable and work with companies and third-party groups to achieve this goal.

The potential risks of AI to children are significant, according to Siegel. He cited possible scenarios of harm ranging from suicide and psychosis to cognitive offloading, which refers to decreased critical thinking caused by relying on AI for answers. Major AI companies are currently rolling out features targeted at children without implementing sufficient safeguards.

Siegel suggested that these companies could take immediate steps to protect young users, such as enforcing age verification measures and allowing users to turn off AI features on their platforms. While such measures might impact short-term usage or profits, failing to address these concerns could lead to lawsuits and regulatory actions that could harm the companies' businesses in the long run.

Google and OpenAI have responded to Siegel's comments. A Google spokesperson noted that parents have access controls to turn off their child's access to the search engine entirely, and that AI Search features are a useful tool for learning and exploration. An OpenAI spokesperson stated that the company uses account and behavioral signals, as well as self-reported age, to identify underage users and that they continuously test and improve these systems. Anthropic did not immediately respond to a request for comment.