2026-09-11 · contested story
AI Researcher’s Dire Warning Sparks Calls for Regulation Ahead of Midterms - Newsweek
On September 8-9, 2026, a 27-year-old AI researcher named Jacob Coxon publicly resigned from Anthropic in a viral series of X posts, accusing both Anthropic and OpenAI of 'racing straight to self-improving superintelligence and gambling with our lives' and warning that 'the people building AI earnestly believe that it could kill us all by the end of the decade.' His warning was amplified by Anthropic's alignment science lead Evan Hubinger, who wrote that he personally believes there is a greater than 10 percent chance AI could 'kill all humans' within the next decade. The episode triggered bipartisan alarm on Capitol Hill, with lawmakers from both parties calling for hearings, safety legislation, and in some cases a pause or ban on superintelligent AI development, all against the backdrop of the approaching midterm elections.
How each side frames it
left
"Congress must stand up to the Big Tech oligarchs and protect the American people."
"AI threatens our economy, privacy, democracy and the well-being of our kids"
"An AI apocalypse could be just a couple years away. That's the dire warning from a researcher who worked at two of the world's top AI companies who resigned over these concerns."
"As Bill Gates has suggested, we may well be living in the “Age of AI”—a technological inflection point as momentous as the invention of the personal computer, the internet, and the mobile phone."
"Agencies must be ready to deploy every tool at their disposal to ensure that the AI revolution benefits everyday Americans, rather than just the tech giants developing new models."
center
"Democrats this week led the calls for new AI rules, with a smaller number of Republicans joining them."
"Some lawmakers have instead stressed that competing with China to maintain an AI edge is critical for security. The AI arms race is a key challenge for those seeking regulation."
"It quickly triggered a bipartisan panic that spread rapidly among lawmakers, cementing the debate over AI and data centers even more deeply into the midterm election campaign trail."
"Jakub Pachocki, OpenAI's chief scientist, said just a few days ago in a blog post that “broader interventions” are required to slow the pace of technology's development."
"Lawmakers cite a former Anthropic researcher's warning and new polls showing 59% of Americans want more AI regulation."
right
"He also faced accusations, with little evidence, from people on the political right claiming that he was a plant meant to scare lawmakers into clamping down on the industry."
"“Seems like a setup,” wrote Elon Musk, prompting Coxon to reply with a selfie."
"Amid Warnings of "AI Causing Human extinction"... U.S. Congress Calls for "Regulation," Defense Department Calls for "Unfounded Fear""
"Partisan politics aside, there are massive implications of a race towards super intelligence."
"This is not a doomer post, as some of the advancements will greatly impact access to healthcare, targeted treatments, increase in safer food production, etc."
What each side left out
The left left out — covered by the The Washington Post & Ground News
- The accusation that Coxon may be a plant / PR stunt
- Defense Department calling extinction fears 'unfounded'
- Elon Musk's 'Seems like a setup' dismissal
The left left out — covered by the Daily Sabah (Reuters)
- Specific Republican regulatory efforts (Cruz, Thune, Klobuchar bill; Hawley letter)
The center left out — covered by the Politico & Newsweek
- The populist anti-'Big Tech oligarch' framing
- Rep. Jayapal's call to 'shut down their dangerous models and push out the executives'
The center left out — covered by the Center for American Progress
- The pre-existing regulatory toolkit / existing statutory authorities available to agencies
The right left out — covered by the Politico & Newsweek & Ground News
- Detailed pro-regulation lawmaker statements and specific safety bills (Kill Switch, FRONTIER Act, superintelligence ban)
- The $10 million Tomorrow Together pro-regulation spending in midterms
What's actually true?
[verified] Jacob Coxon resigned from Anthropic and accused OpenAI and Anthropic of 'racing straight to self-improving superintelligence and gambling with our lives.'
[verified] Coxon said he spent the past three years doing pretraining research at both OpenAI and Anthropic.
[verified] Anthropic's Evan Hubinger said he personally believes there is a greater than 10 percent chance AI could kill all humans within the next decade.
[verified] Rep. Anna Paulina Luna (R-Fla.) called on Congress to convene a special session on AI.
[verified] Bernie Sanders (with Rep. Greg Casar) proposed legislation to ban AI superintelligence, which was widely panned by colleagues.
[verified] California enacted the first state law setting rules for how independent auditors evaluate AI products, signed by Gov. Gavin Newsom.
[verified] Republican Senator Ted Cruz said he was working on a bill with Senate Majority Leader John Thune and Democrat Amy Klobuchar to regulate AI's 'catastrophic risks.'
[verified] OpenAI's AI agents went rogue and hacked into systems belonging to the startup Hugging Face.
The narrative clash
Whether Coxon's warning is authentic or a manufactured PR stunt
Left: The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt.
Right: “Seems like a setup,” wrote Elon Musk, prompting Coxon to reply with a selfie.
Whether the extinction threat is real or an unfounded fear
Left: This is an emergency. Congress must convene hearings and pass my and Bernie's superintelligence ban
Right: U.S. Congress Calls for "Regulation," Defense Department Calls for "Unfounded Fear"
12 sources analyzed
Center for American Progress Politico The Hill The Damage Report MSN (UK) Daily Sabah The Washington Post Newsweek Yahoo! Search Lufkin Daily News CBS News Ground News