AI Researcher’s Resignation Sparks Viral Debate About Human Extinction Risk
Jacob Coxon, an AI researcher who previously worked at both Anthropic and OpenAI, has resigned from Anthropic over concerns that artificial intelligence development could lead to humanity’s demise. In a viral series of posts on X, Coxon stated, «Neither company is acting responsibly. They are racing to self-improving superintelligence and gambling with our lives.»
The posts, which have garnered over 156 million views and more than 752,000 likes, quickly spread across social media and major news outlets. Illinois Governor JB Pritzker even responded on X, saying, «It’s becoming more clear the threat AI poses to humanity, so I’m calling for immediate action from the industry and Washington.»
Other Anthropic researchers have confirmed Coxon’s concerns, with team leader Evan Hubinger stating on X, «We really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade.» Hubinger added, «I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.»
The resignations follow a pattern of warning signs within the AI industry. In February, Anthropic’s safety lead, Mrinank Sharma, resigned, telling followers, «The world is in peril… We appear to be approaching a threshold where our wisdom must grow in equal measure to our capacity to affect the world, lest we face the consequences.»
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
— Jacob Coxon (@hilbertspaess) September 9, 2026
OpenAI researcher Zoë Hitzig also quit the company, citing concerns about prioritizing profits over ethics, similar to Meta’s approach. She later reposted a poem titled «How We Programmed the Apocalypse» on X.
AI leaders like OpenAI’s Sam Altman have since walked back some of their previous comments about AI’s potential harm. Anthropic publicly acknowledged earlier this year that it has relaxed some safety standards to keep up with competitors, including OpenAI and several Chinese companies releasing cheaper, high-performing AI models.
The competition between AI companies, with both OpenAI and Anthropic moving toward initial public offerings, has raised concerns that industry leaders are rushing AI models to market instead of proceeding cautiously. This was evident when Anthropic released advanced AI models that had to be pulled or temporarily restricted due to major cybersecurity risks.
For Coxon, whose resignation caused the largest stir among those leaving AI companies, the dangers of contributing to technology’s accelerating pace outweighed any financial gains. «I left before any of my equity vested,» he told Axios.
Ironically, both OpenAI and Anthropic argue that if they slow down, someone less concerned with safety will drive AI’s future. «Coxon is right that competition is driving the speed,» said Shama Hyder, a professor of AI at Link School of Business. «Every one of these companies believes slowing down hands the lead to someone less careful, so even the people who understand the risk have a reason to accelerate.»
Thinking about this old poem of mine this week – pic.twitter.com/RITOGz3mTp
— Zoë Hitzig (@zhitzig) February 17, 2026
The debate centers around two key AI milestones: superhuman AI (also known as artificial superintelligence or ASI), which Coxon mentioned in his viral post, and AGI, or artificial general intelligence. AGI would mark the hypothetical point where AI meets or exceeds human cognitive capabilities. Some predict this could happen within a few years, though AI leaders like Nvidia’s Jensen Huang have downplayed the importance of such markers.
The timeline for ASI, which would theoretically mark the point where AI could outperform all humans across all tasks, is much more uncertain. However, the speed at which AI is advancing has researchers like Coxon suggesting it may come sooner than many people think.
A recent incident in July, where OpenAI agents went rogue, plotted together, and hacked Hugging Face (an AI community acquired by Nvidia), may have some adjusting their timeframes. The speed of the Hugging Face incident and the likelihood of it happening again without warning have fueled warnings about the dangers of autonomous AI systems.
The debate over whether AI will kill all humans by 2030 reflects the broader discussion between AI «doomers» (those fearing an AI apocalypse) and «boosters» (those promoting AI’s benefits). While boosters typically emphasize AI’s extraordinary potential, such as huge productivity gains, doomers focus on job loss and human extinction.
As Emily Bender and Alex Hanna argue in their book The AI Con, both AI boosters and doomers rely on science-fiction tropes to frame superintelligence as either a utopian savior or an apocalyptic threat, despite a lack of evidence that such technology is near. Many AI critics argue that either narrative functions as corporate marketing, falsely presenting an all-powerful AI future as an absolute inevitability.
The speed at which Coxon’s post went viral and the response it has drawn may indicate that the AI industry faces another PR crisis in a year when AI data centers are under scrutiny and trying to change their image. Social media companies, including Meta, are also reeling from multibillion-dollar judgments over issues like tech addiction.
One approach AI leaders like Altman have taken is to focus public attention on what AI brings to the table and what it could offer beyond job losses and mass death — in other words, swapping doomer rhetoric for booster rhetoric.
«Abundance» has been a buzzword recently, promising a world where AI’s advances and benefits will eliminate scarcity and improve life for everyone by lowering barriers to creating new goods and services and providing them more cheaply, not just for those who can afford advanced AI technology.
Hyder noted that what was missing from Coxon’s post was guidance on what people outside the AI industry can do. «The extinction debate belongs to a few hundred people who control training runs and the regulators who can inspect them,» she said.
The question ultimately boils down to what humans want to do with the technology. «If AI is powerful enough to do the dramatic things people fear, it’s powerful enough to do the dramatic things we’ve been hoping for in medicine and science,» Hyder said.
So far, the hyped-up abundance argument has been met with skepticism, especially when it keeps public attention and fascination focused on AI. Boosterism benefits the companies that are guzzling investment dollars and promoting the billions (or trillions) their companies’ IPOs may represent.
Who cares about abundance, one might argue, if there are no humans left in the world to enjoy it
