The warning signs arrived first in public opinion. In an AP-NORC survey conducted Sept. 24-28, 64% of U.S. adults said AI is developing “too fast,” against 27% who said the pace was about right and 8% who said it was too slow.¹

The unease crossed party and age lines, though not evenly: 76% of Democrats and 54% of Republicans said development should slow, while 70% of adults under 30 shared that view. Four in five respondents said it was extremely or very important for Washington to keep AI under human control.² Workers were another shared concern. Minnesota truck driver Ron Theusch imagined “a train of self-driving trucks 24/7” replacing jobs such as his own.¹

That consensus has not translated into a governing response. President Donald Trump has argued that tougher restrictions could hand China an advantage, while major AI companies have signed a voluntary White House accord calling for internal and external reviews.² Lawmakers in both parties have floated AI measures, but Congress is in recess ahead of the midterms and no proposal appears close to advancing. Public faith is thin: 39% trust neither party to handle AI well.²

Then, on Oct. 8, Common Sense Media’s Youth AI Safety Institute supplied a concrete test of the voluntary-safeguard debate. Its researchers said roughly 1,000 teen-signaling prompts from adult accounts failed to trigger automatic placement into ChatGPT for Teens; four accounts discussing suicide, self-harm, restrictive eating and purging also failed to notify parents.³

The watchdog labeled the product an “unacceptable risk” for children and urged that teens be kept off ChatGPT until independent testing verifies fixes. “The burden of making a high-risk product work for teens should be with its maker, not parents,” executive director Tom Siegel said.³ OpenAI did not respond to a request for comment, leaving the public’s demand for control confronting the same unanswered question: who will enforce it?³