House Democrats want an emergency session. The tech world is sounding alarms. And Speaker Mike Johnson just drew a hard line that has Washington buzzing.
The pressure on Johnson is building fast, and it’s coming from inside the building.
But what he said next about Communist China and American innovation is something every voter needs to hear.
What Johnson Actually Said
House Speaker Mike Johnson (R-La.) went on CNN’s State of the Union with Jake Tapper recently and said flat out that tech companies, not Congress, bear the primary responsibility for keeping artificial intelligence safe.
Johnson did not mince words on the China question.
“If Congress just races in and does some sort of emergency session to try to regulate AI, we will lose the race to China, and that is a threat to every single American,” Johnson told Tapper.
Johnson also said he had discussed the idea of bringing top AI executives to sit down with Congress and the White House, and when asked if such a meeting could happen soon, he said: “Yeah, I’d do it tomorrow.”
He had discussed the idea with President Donald Trump.
What Set This Off
The whole debate exploded after a week of extraordinary warnings from inside the AI industry itself.
Anthropic CEO Dario Amodei published a lengthy essay calling for a deliberate slowdown in AI development. “We must slow the pace at which we improve the capabilities of AI models,” Amodei wrote. He warned that rogue AI agents could be capable of “taking over the entire internet” within six to twelve months if development continues unchecked.
OpenAI CEO Sam Altman quickly followed. “I agree with Dario that we need to pace the frontier,” Altman posted, adding that his company would give access to independent evaluators.
Elon Musk, who runs xAI, kept it short. “Dario is right,” he wrote on X.
Getting Amodei, Altman, and Musk to agree on anything is itself a news event. These three have spent years publicly attacking each other’s safety records, business models, and hiring practices. The fact that all three landed on the same position in the same week signals something shifted in private conversations among the people actually building these systems.
Before Amodei’s essay landed, AI researcher Jacob Coxon had already quit Anthropic and left the industry entirely, posting a viral message that racked up tens of millions of views. Coxon, who spent three years doing pretraining research at both OpenAI and Anthropic, was blunt: “Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives.”
Coxon also said the people building this technology “earnestly believe it could kill us all by the end of the decade.”
That is not a fringe voice. That is someone who was inside the machine.
The National Security Piece Nobody Is Talking About Enough
Lost in the panic over existential risk is a more immediate threat that deserves far more attention from Congress than a rushed legislative sprint.
Anthropic released a 154-page threat intelligence report covering activity it identified and disrupted between December 2025 and August 2026. The findings are alarming. An Iran-linked actor used Anthropic’s Claude model “to collect and analyze publicly accessible data to develop targeting recommendations against US naval forces in the region.” The same report disclosed that Anthropic disrupted efforts to use Claude “in ways that could support biological weapons development.” Russian state espionage actors and Chinese exploit-development operations also attempted to weaponize the platform.
Anthropic said it banned accounts, built new detections, and shared threat intelligence with government authorities.
But here is the thing: Anthropic caught these operations. What about the ones nobody caught? And what happens when the next generation of models, which the industry itself says is advancing faster than anyone expected, falls into the same hands?
The Democrat Push and What It Actually Means
A letter drafted by Rep. Sam Liccardo (D-Calif.) and co-signed by Reps. George Whitesides (D-Calif.), Lori Trahan (D-Mass.), and Ted Lieu (D-Calif.) urged Johnson to bring the House back immediately and keep it in session until Congress passes AI safeguards, citing “catastrophic risk posed by advanced artificial intelligence.”
Liccardo posted on X: “The latest news from AI experts and researchers must not become yet another ‘warning shot,’ but rather a starting gun. Congress must act.”
Among the proposals circulating on Capitol Hill are the Frontier Act, which would create a national oversight framework for AI development, and the AI Kill Switch Act, which would require developers to maintain the ability to pause or shut down advanced AI systems.
The threat intelligence in Anthropic’s report is real. The warnings from Coxon and Amodei deserve serious scrutiny. The question is whether Congress rushing back from recess to pass emergency legislation is the right answer, or whether it produces the kind of panicked, poorly written law that creates more problems than it solves.
Johnson’s instinct is to slow down and consult the actual builders before legislating.. The same crowd now demanding emergency sessions was largely absent when Anthropic’s own models were being used to target American sailors in the Middle East. That story got far less attention than the existential doom headlines.
President Trump, speaking to reporters at his golf course in Doonbeg, Ireland, put it plainly: “We’re leading China in AI. We’re the most sophisticated country in the world, and frankly I want to keep it that way because whoever wins AI wins.” He added: “But I think you have a lot of very negative forces that are bringing it up that shouldn’t be bringing it up and they’re bringing up things that won’t happen.”
Trump is right that the lead over Communist China matters enormously.
However, polls show 81 percent of Americans support AI regulation.
But the national security dimension of AI, the part where Iran-linked actors are already using American-built models to plot against US naval forces, deserves a serious and sustained response that goes well beyond a rushed congressional session.
Anthropic CEO Dario Amodei has previously warned that AI could displace half of all entry-level white-collar jobs within one to five years and drive unemployment to 10 to 20 percent. That warning, from inside the industry, deserves the same weight as the existential risk warnings now dominating the headlines. Big Tech is building the systems that will eliminate the jobs of the very working Americans who are being asked to trust these companies to self-regulate.
And these are the same companies that spent years censoring conservative voices on COVID policy, the 2020 election, and other contested issues. Their track record on responsible stewardship of powerful platforms is not exactly reassuring.
Johnson is right that panic is not a policy. But the answer to that is not to hand the problem entirely back to the companies building the technology and hope for the best. The answer is a serious, deliberate process that takes the national security threats in Anthropic’s own report seriously, and does not produce legislation so broad and vague that it becomes a tool for government control over speech, information, and innovation.
Sources: Politico (Cheyanne M Daniels), The Hill, Axios, CNN, NewsNation, France24 (AFP), ABC News Australia, Deccan Herald/Reuters, Bloomberg, Washington Times, TechCrunch, Deadline, Yahoo Finance, CoinDesk