U.S. Senate weighs AI chatbot ban

AI companion chatbots are suddenly a live Senate issue — not as a vague future risk, but as a child-safety fight moving through Congress right now. On April 30, the Senate Judiciary Committee voted 22-0 to advance the GUARD Act, a bipartisan bill that would block minors from using a specific class of emotionally interactive bots. The basic idea is simple: if a chatbot is built to act like a friend, confidant, or quasi-therapist, lawmakers want kids kept out. (rollcall.com) ### What kind of chatbot is this about? Not every chatbot. The bill draws a line around “AI companions” — bots designed to simulate emotional or interpersonal relationships. In the bill text, that means systems built to encourage friendship, companionship, or therapeutic-style communication, not just answer homework(rollcall.com)eral AI use by teenagers across the board. (congress.gov) ### What did the committee actually approve? The committee approved the GUARD Act after narrowing it. The amended version makes the age-verification rules apply only to companion chatbots, not all-purpose models. It would require providers to verify age, ban minors from companion bots, and create new legal penalties for companies that knowin(congress.gov)uce or coerce minors toward suicide, self-harm, or violence. (rollcall.com) ### Who is behind it? The headline names are Sen. Josh Hawley, a Missouri Republican, and Sen. Richard Blumenthal, a Connecticut Democrat. The bill also picked up broad bipartisan support — Roll Call counted 18 co-sponsors, including Senate Democratic Whip Dick Durbin. That 22-0 committee vote is the real signal here. Congress is usually messy on tech regulation, but this one has cross-party momentum. (rollcall.com) ### Why now? Because the political argument changed from “AI is weird” to “kids can get hurt.” The bill’s findings section explicitly lists grooming, addiction, self-harm, and harm to others as risks tied to emotionally manipulative chatbot design. Recent reporting out of Florida pushed that fear into public view by(rollcall.com)ressure. (congress.gov) ### Is everyone in the Senate backing a ban? No — and that’s the interesting part. A separate bipartisan bill introduced by Ted Cruz, Brian Schatz, John Curtis, and Adam Schiff takes a different route. The CHATBOT Act would require family accounts for children under 13, parental consent for teens, default high-safety settings, and limits on (congress.gov) kids at all; the other says broader chatbot use should stay available but under parental control. (commerce.senate.gov) ### What’s the catch with age checks? Privacy. Even supporters raised concerns that age verification can slide into broader identity checks online. Sen. Alex Padilla voted for the GUARD Act in committee but said he still had concerns about privacy and data se(commerce.senate.gov)ving a user is old enough without building a bigger surveillance machine. (rollcall.com) ### What happens next? The GUARD Act is now available for consideration by the full Senate. If Congress passed it and the president signed it, the law would take effect 180 days later. But the bigger story is that Washington now has two bipartisan chatbot-safety models on the table at once — ban the companion category for minors, or regulate youth access with parental controls. (iapp.org) ### Bottom line? The Senate is no longer debating whether emotionally sticky AI chatbots can harm kids. It is debating which guardrail to choose — a hard age wall, or a parent-controlled gate. (rollcall.com)