US Surgeon General Calls for AI Guardrails, Citing Risks to Children

Two-time Surgeon General Vivek Murthy drew parallels between the advent of artificial intelligence and nuclear weaponry, emphasizing the urgent need for safety standards during a conference focused on the implications of AI for children and families. "We are living through the greatest transformation that we have seen in our lifetimes," Murthy stated at the event, organized by Common Sense Media.

Jim Steyer, founder and CEO of Common Sense Media, echoed these concerns, asserting that the industry cannot be trusted to self-regulate. "We did not get out in front of social media in time," Steyer said. "AI is a bigger deal. It's basically social media on steroids."

New York Governor Kathy Hochul announced new regulations aimed at chatbots interacting with children. The law, effective in January, prohibits AI agents from encouraging suicide, self-harm, secrecy, or isolation, and from discouraging children from seeking human help. It also bans sexually explicit interactions and restricts the use of a child’s past personal or health information in subsequent conversations.

Speakers at the summit, which included figures like Chelsea Clinton and former U.S. Secretary of Education John King, lamented the lack of federal regulation on both AI and social media. Murthy, who chairs the advisory board of Common Sense Media’s Youth AI Safety Institute, noted that "nearly a quarter century after the advent of social media—failed to announce anything to protect our kids."

Calls for AI regulation ranged from demanding greater data transparency from tech companies to implementing a model similar to Europe's, which requires AI companies to prove their platforms are safe before deployment. New Mexico Attorney General Raúl Torrez suggested that legal consequences could spur industry responsibility, stating, "When you bring out handcuffs, suddenly everything changes."

Beyond regulatory measures, experts also raised concerns about the impact of AI on early childhood socialization. Dana Suskind, an early-childhood expert, questioned what children learn from constant AI interaction, such as expecting constant agreement and ease, contrasting it with the messier, yet more educational, nature of human relationships.

The summit also addressed AI's influence on education. John King highlighted a report indicating that generative AI could complete most undergraduate assignments at MIT, prompting a need to re-evaluate student assessments and ensure foundational skills are being developed. Chelsea Clinton pointed out the equity implications, noting that teenagers from lower-income households are more likely to use AI for homework than their higher-income peers.

Preparing students for an AI-integrated workforce was another key topic. Gene Sykes of Goldman Sachs observed that employees proficient in AI can become highly productive more quickly, but uneven distribution of AI education could exacerbate existing wealth disparities.

While immediate solutions remain debated, the consensus was clear: AI will have profound consequences for future generations. "As we go into this age of transformation," Murthy concluded, "the question forced upon us is the following: What aspects of human existence do we want to protect and preserve at all costs?"