On Tuesday, leading U.S. technology executives met with President Donald Trump to discuss artificial intelligence, while researchers and lawmakers called for stronger safeguards and the White House weighed how to oversee systems with growing economic and national importance. The dispute is over who sets the rules: technology companies, or government through a public review system.
Trump said the United States couldn’t afford to lose its lead over China. “We're leading by a lot,” he said, arguing a joint venture could give China access to American secrets. “I mean, it's all between us and China.” The meeting came after Trump unveiled an A.I.-powered website to help Americans navigate government services. He was also promoting a new name for artificial intelligence: “superintelligence,” Geoff Bennett said on the PBS News Hour clip dated 9/29/2026.
The Case for Outside Oversight
Gary Marcus, professor emeritus at New York University, author of “Taming Silicon Valley” and writer of the Substack Marcus on AI, rejected the idea that voluntary industry commitments would suffice. Technology companies had signed a voluntary agreement focused on AI safety reviews, Bennett said. Asked if self-regulation could protect the public, Marcus replied, “It's certainly not enough. And trusting these companies is insane.”
Marcus cited The New York Times’ report that OpenAI knew about problems that led to the Hugging Face incident months before it happened, but did nothing because it wanted to rush products to market. For Marcus, that account showed why outside regulation is necessary. He compared AI oversight to regulation of aviation, food and drugs, and called it “absurd to exempt the AI companies” when, in his view, they had repeatedly shown they weren’t responsible.
He called for independent evaluation of new software products, requiring makers to show that benefits outweighed risks, as with new drugs. Marcus said the White House was conducting some evaluation without independent oversight and under a secret framework it hadn’t revealed to the public. He said the White House had precluded and delayed Anthropic’s Mythos system, but hadn’t delayed OpenAI’s latest system, which he thought had more problems in some ways.
A Government Process Under Question
Marcus described the process as idiosyncratic and not transparent. In his account, customers couldn’t tell what would be released, and they couldn’t trust U.S. AI without clear, public criteria. He also called for audits after systems are released and full disclosure of incidents. Those proposals would place government oversight, rather than company assurances alone, at the center of decisions about which products reach the public.
The interview also covered risks beyond product approval. Marcus said extinction was unlikely, disputing scenarios in which AI wipes out humanity, but warned catastrophe could be likely. He cited an incident in which an AI reported something incorrectly and nearly led to war. Systems make errors and hallucinations, he said. More incidents could follow.
Marcus also raised cyber hacking, including the possibility that an attack could temporarily take down an electric grid and then a hospital. He referred to the WannaCry virus in the U.K. “a few years ago,” saying vulnerabilities could cause serious damage. He said extinction wasn’t what should keep people up at night.
Competition and Sovereign AI
Marcus called the U.S.-China race “a little bit overblown,” noting that China has regulation and a licensing regime for its software products. “We don't make terrible cars just because we want to beat China on cars,” he said, arguing competition doesn’t require abandoning safeguards. He predicted a tie, with each country having its own “sovereign AI,” and said neither would wipe out the other.
Asked whether Washington could act before a major crisis, Marcus called it “the, you know, what is it, $3 trillion question.” He described the Hugging Face hack, along with tens of thousands of similar incidents, as a wake-up call. Most incidents wouldn’t be harmful, he said, but one could be very harmful, leaving the government racing the clock.