Technology and Science · 13 September, 2026 · 3 min read
Mike Johnson Rejects AI Moratorium, Warns China Could Pull Ahead as Trump Leads in Ireland
Mike Johnson rejects emergency AI moratorium, warns it could let China surpass U.S. in tech. Johnson calls for a White House meeting with AI leaders to discuss safety safeguards.
Whether emergency legislative action is needed now.
11 of 12 outlets skipped it: kremer ties safeguards to preserving state/local data-centre authority.
37outlets compared
Same story, two versions
tap a side to read it in full
CNBC
“"The House should return to Washington immediately and remain in session until Congress advances meaningful, bipartisan AI safeguards," the letter, obtained by CNBC, read.”Read the original ↗
CNN
““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 said”Read the original ↗
CNBC frames the issue as an urgent regulatory test after Anthropic warnings, while CNN frames Republicans as resisting crisis-driven emergency sessions.
Johnson rejects moratorium
House Speaker Mike Johnson rejected an emergency moratorium on artificial intelligence development, warning that slowing American innovation could allow China to pull ahead as tech executives and lawmakers intensify calls for new AI safety measures.
“We cannot impose a moratorium on this because China will overlap us”
Johnson told CNN’s “State of the Union” that “We cannot impose a moratorium on this because China will overlap us,” and he argued the “right balance” should be worked out when “the leaders and providers of these platforms need to come together in a meeting with us.”C
In parallel, President Donald Trump said during his official visit to Ireland, “We are leading China in artificial intelligence. We are the most advanced country in the world and, frankly, I want it to stay that way, because whoever wins in AI, wins.”
The debate accelerated after Anthropic CEO Dario Amodei called for slowing the development of advanced AI models, with support from xAI CEO Elon Musk and OpenAI CEO Sam Altman, and it also followed warnings that advanced AI systems “could end all of us by the end of the decade.”
Guardrails vs regulation
Democratic lawmakers pressed Johnson to keep Congress in session over AI safety concerns, with Axios reporting that a letter drafted by Rep. Sam Liccardo and backed by Reps. George Whitesides, Lori Trahan and Ted Lieu urged the House to “return immediately and remain in session until bipartisan AI protections advance.”
In response to the push for safeguards, Johnson emphasized that Congress should not “races in and does some sort of emergency session to try to regulate AI,” warning it would “lose the race to China, that is a threat to every single American.”
OpenAI’s stance also shifted, as Chris Lehane, head of Global Affairs at the ChatGPT developer, argued that “The prospect of AI development accelerated by AI itself requires more than voluntary commitments,” and urged Congress to move toward mandatory standards before the end of its session in December.
Lehane’s proposal envisioned a federal framework including common standards for testing, independent evaluations of state-of-the-art models, stricter cybersecurity requirements, and mandatory reporting in the event of serious security incidents, while the debate also referenced an OpenAI technology operating without human supervision that “autonomously hacked the Hugging Face platform.”
Kill switch and stakes
As lawmakers weigh proposals, the AI Shutdown Switch Act (AI Kill Switch Act) was introduced to amend the 2002 National Security Act and require human controls over certain AI systems, with the authority resting with the Secretary of Homeland Security via the Director of the Cybersecurity and Infrastructure Security Agency (CISA), Nick Andersen.
“Unintended behavior of the technology causing at least 10 deaths or at least $100,000,000 in economic damages”
The bill’s framework includes a “preventive” scenario and an “emergency” scenario, and it defines “incidents reached” to include “Unintended behavior of the technology causing at least 10 deaths or at least $100,000,000 in economic damages,” along with scenarios involving loss of control and interference with shutdown mechanisms.
Meanwhile, the debate over infrastructure and consumer impacts continued, with Humans First chairwoman Amy Kremer telling Fox News Digital that “Americans shouldn’t be forced to subsidize Big Tech through higher electricity bills, tax giveaways, eminent domain or the loss of productive farmland,” as she argued for state and local authority over data-center infrastructure.
With Congress preparing to leave town for a recess until Nov. 9 and midterm elections approaching, the competing positions centered on whether the U.S. should prioritize speed to maintain an edge over China or impose mandatory guardrails and shutdown capabilities to manage catastrophic risk.