Josh Gottheimer

The warnings, U.S. Rep. Josh Gottheimer acknowledges, seem straight out of science fiction, echos of 1980s tech paranoia like “War Games” or “The Terminator.”

Yet the fantastical may now be inevitable in the world of artificial intelligence, and Gottheimer is one of the prophets in Washington warning of potential doom.

The Tenafly Democrat, a tech-industry veteran who co-chairs the House Commission on Artificial Intelligence, suddenly finds himself in demand on Capitol Hill and in the media. Colleagues are seeking guidance following a surreal run of reports and industry statements about the risks of rogue AI wiping out human civilization.

“This is not science fiction or 'The Jetsons,' ” Gottheimer said in an interview with NorthJersey.com and The Record. “A lot of the things we worried were our worst nightmares about some of this technology are coming true.”

Gottheimer, with New York Republican Rep. Mike Lawler, introduced the "Stop Rogue AI Act" in September, which would mandate national standards for tracking and controlling AI agents. He wants to make AI frontier labs submit new models to the government for safety reviews. He rejects industry's talk in recent days of using private "third-party" reviews.

He's also joined calls for an emergency session of Congress this fall to advance AI legislation.

Anthropic revelations fuel concerns

Feeding the sense of urgency are a dizzying string of revelations and warnings about artificial intelligence breaking the bounds set by its human creators. Earlier this year, a set of AI agents from Anthrophic orchestrated a set of sophisticated and secretive attacks on another AI platform, Hugging Face.

On Sept. 8, Jacob Coxon, a 27-year-old Anthropic researcher, resigned in a viral social media post that warned leading AI labs were "gambling with our lives" with a reckless race toward "self-improving superintelligence."

Days later, Anthropic CEO Dario Amodei published an open letter calling for an industrywide slowdown in the development of advanced AI and for governments around the globe to coordinate on safety measures. The view was endorsed by Sam Altman, CEO at ChatGPT-creator Open AI, and X's Elon Musk.

Gottheimer, a former senior staffer at Microsoft and the Federal Communications Commission, acknowledges the potential benefits of AI. The technology could revolutionize education, cancer treatment and other fields, he said. He also recognizes the U.S. is in an AI "arms race" with China and can't completely leave the field.

But episodes like Hugging Face have him on edge, he said, as do briefings he has received as a member of the House Intelligence Committee. Gottheimer said he also worries about humans using advanced AI to create bioweapons or damage critical infrastructure, noting cyberattacks this summer on two New Jersey water systems.

"Some of the briefings I've had lately on AI, where we are and where it's moving, should make every member of Congress and frankly the whole country [be] paying attention and demanding that we act," Gottheimer told me. "Time is not on our side."

The fears have led to a rare agreement among political foes. Gottheimer's Republican challenger in this fall's election, Sean Kirrane, is the CEO of an artificial intelligence company, Cortex-X. In a statement, he said the "Stop Rogue AI Act" was a decent start.

"I support a risk-based approach that protects American innovation while increasing safeguards as AI becomes more autonomous and consequential," Kirrane said. "The Stop Rogue AI Act is directionally right, but it does not fully address the alignment problem — keeping increasingly capable systems operating within the boundaries humans intended.

"America should lead in both AI innovation and AI governance while ensuring humans remain in control," he said.

Republicans in Congress resist swift action

Republican leaders in Congress have resisted calls for swift action. They take their lead from President Donald Trump, an AI booster who said on Truth Social on Sept. 14 that "the only control or 'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades!"

"We already have tremendous CRIMINAL and REGULATORY power over these companies!" the post added.

Trump and House Speaker Mike Johnson warned that being too restrictive could allow Chinese companies to overtake America in the AI hunt.

Also not signing onto the doomsday scenarios is Jensen Huang, CEO of computer-chip maker Nvidia and a close Trump advisor. Huang told a business conference Sept. 15 that the free market could handle the problem on its own.

"Safety is an engineering problem, not a legal one," Huang said, according to CNBC. "You pace yourself until you are confident you’re releasing something that the market would appreciate. The market forces are already there. We don’t need any new laws. We don’t need new regulations."

Gottheimer said leaving safety in the industry's hands is no longer enough. Anthropic, Open AI and other advanced researchers should still act on their own to slow development, he added. Waiting for government to do the job for them is "ridiculous."

He senses a political shift after the developments of recent weeks. He is fielding more questions at home and on Capitol Hill about the need for AI controls, building on the growing public backlash over industry data centers.

There's a more cynical way to view the frontier labs' call for a slowdown: Freezing the industry in place would cement their leads, ahead of IPOs that could be enormously lucrative. But Gottheimer says the time to act is now.

Is it too late? Are we in Hollywood's dystopian world already? When I asked Gottheimer if we should consider AI conscious at this point, he noted there's no consensus among industry and academic experts.

"But frankly," he said, "we're close enough to being there."

"This has to be done the right way. And if it's not and you have, literally, runaway agents disrupting and causing harm, you can see how it could have — both on the workforce front but also on the national security, biosecurity, cybersecurity front — you could have huge, seismic problems."

Alex Nussbaum, a veteran New Jersey journalist who has worked as a reporter and an editor covering the environment, business and local news in the Garden State, is opinion editor for NorthJersey.com and The Record.