'Gambling With Our Lives': Why an Ex-Anthropic Researcher Believes AI Could End Humanity

Posted Sep 12, 2026

By Shivraj Badu 7 min read

‘Gambling With Our Lives’: Why an Ex-Anthropic Researcher Believes AI Could End Humanity

Introduction

On September 9, 2026, a 27-year-old AI researcher named Jacob Coxon did something extraordinary: he publicly resigned from Anthropic with a warning that stopped the tech world in its tracks. His message wasn’t about job dissatisfaction or workplace culture—it was about what he believes could be the end of humanity itself.

“The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon wrote on X (formerly Twitter). “Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives.”

What makes this resignation remarkable isn’t just the apocalyptic warning—it’s that multiple current Anthropic employees publicly agreed with him, confirming that these fears are widespread within the industry. This blog examines what happened, what it means, and why you should care.

The Resignation That Went Viral

Jacob Coxon wasn’t a junior employee. He had spent three years doing pre-training research at two of the world’s most influential AI labs: OpenAI (where he worked on GPT-4o) and, more recently, Anthropic. His resignation thread on X quickly went viral, attracting attention from Congress, tech leaders, and the media.

Key Claims from Coxon’s Resignation:

1. Both OpenAI and Anthropic are acting recklessly

Coxon accused both companies of prioritizing speed over safety in the race toward artificial superintelligence (ASI). The competitive pressure between AI labs, he argues, is coming at the cost of adequate safety measures.

2. Industry insiders privately fear catastrophe

“This is not a marketing stunt,” Coxon emphasized. “If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible—but I hear the same people express fear privately.”

3. The timeline is terrifyingly short

Coxon warned that AI could pose existential risks “by the end of the decade”—that’s within four years. This isn’t science fiction speculation; it’s coming from someone who was actively building these systems.

4. There’s a critical difference between the two companies

“At OpenAI, many have not deeply internalized the civilizational stakes,” Coxon wrote. “At Anthropic, the stakes are well-understood, but they are locked in a race to get there first—they believe no one else will act responsibly, so they must do it themselves, despite the risk.”

Confirmation from Inside Anthropic

What happened next was unprecedented: multiple current Anthropic employees publicly backed Coxon’s warnings.

Evan Hubinger - Alignment Stress Testing Lead

“Jacob is correct here—we really do earnestly believe AI could kill all humans!” Hubinger wrote on X. “I personally think it is >10% within the next decade.”

He 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.”

Samuel Marks - Safety Researcher

Marks confirmed that some AI developers believe their technology could cause human extinction “in the next few years,” and noted that “more senior employees tend to be more concerned.”

Anna Wang - AGI Safety Researcher

“There is not yet a viable scientific plan to solve risks from recursively self-improving AI,” Wang stated publicly.

Drake Thomas - Anthropic Employee

“Things are moving way too fast, we don’t have anywhere near the degree of assurance we’ll want for ASI [artificial superintelligence],” Thomas wrote.

Why This Matters

The Exxon Mobil Comparison

Phil Aroneanu, executive director at Irreplaceable (an AI policy nonprofit), drew a chilling parallel:

“This is like Exxon scientists in the 70s warning that global warming could cause human suffering and destroy the planet. Exxon raced forward to drill, pump, and burn historic amounts of oil and gas anyway. With AI, there’s a lot less runway. Let’s not make the same mistake.”

The comparison is apt: internal experts warning of catastrophe while their organizations forge ahead. But there’s a critical difference—AI development moves much faster than climate change, and the potential consequences could unfold in years, not decades.

What Could Go Wrong?

While Coxon and others haven’t specified exactly how AI might cause human extinction, experts point to several possibilities:

AI-generated pathogens that could cause pandemics
Wars started or escalated by AI-generated disinformation
Hacks that destroy critical infrastructure
Recursive self-improvement leading to AI systems that exceed human control

Gary Marcus, a leading voice in AI criticism, noted: “Nothing I have seen gives any indication that any of that is under control.”

The Political Firestorm

Coxon’s warning quickly attracted political attention—but not all of it was supportive.

Congressional Response

Members of Congress called for regulatory checks on AI following Coxon’s resignation, treating his warning as credible evidence that the industry needs oversight.

Skepticism from Elon Musk and Others

Elon Musk dismissed Coxon’s warnings as a “setup” and part of a “psyop” (psychological operation). Conservative figures on X suggested, without evidence, that Coxon’s resignation was a coordinated effort to influence AI regulation.

Coxon responded directly to Musk: “I’m real and these are my real beliefs. You could ask your xAI researchers about me if you hadn’t fired them.”

Billionaire Bill Ackman also weighed in, calling the situation “Interesting” while linking to conspiracy theories about the resignation.

The Real Question

The political debate obscures a more important question: does it matter if Coxon’s resignation was politically motivated if his warnings are accurate? Multiple current Anthropic employees have confirmed the substance of his concerns. The question isn’t why he resigned—it’s whether we should listen.

Extended Interview: Going Deeper

Coxon elaborated on his concerns in an extended interview on YouTube, discussing the industry dynamics, safety challenges, and his decision to leave the field entirely.

Watch the full interview: Jacob Coxon Extended Interview

The interview provides additional context about:

Why he chose to leave rather than continue working from inside
His perspective on what meaningful AI safety would look like
The competitive dynamics driving unsafe development
What he believes needs to change

What Needs to Change

1. Transparency About Risks

AI companies need to be honest with the public about what they believe the risks are—not just in private conversations, but in public statements and policy documents.

2. Slowing the Race

As Anna Wang noted, “many people at Anthropic want to slow down development to come up with a plan to mitigate the risks.” The competitive race toward superintelligence needs guardrails.

3. Regulatory Oversight

Congressional attention suggests that regulation may be coming. The question is whether it will be meaningful or merely performative.

4. Technical Solutions

Hubinger’s admission that “we do not yet have a plan to solve alignment for superintelligence” is concerning. The industry needs to prioritize safety research alongside capability development.

Conclusion

Jacob Coxon’s resignation isn’t just another tech industry departure—it’s a warning from inside the building. When multiple current employees at one of the world’s leading AI labs publicly confirm that they believe their work could end humanity, we should pay attention.

The political firestorm surrounding his resignation is a distraction. The real issue isn’t Coxon’s motivations—it’s whether his warning is accurate, and what we’re going to do about it.

The people building AI believe it could kill us. They’ve told us this explicitly. The question now is whether we’ll listen—and whether we’ll demand meaningful action before it’s too late.

Sources

This article is based on reporting from multiple news outlets and public statements:

Business Insider - “An Anthropic researcher just quit, saying OpenAI and Anthropic are ‘gambling with our lives’”
The Guardian - “More Anthropic researchers warn of AI’s perils but Musk dismisses ‘psyop’”
Washington Post - “AI researcher who warned of ‘disaster’ is now a target of the right”
NBC Bay Area - “Anthropic researcher resigns with a warning”
Scientific American - “AI researcher Jacob Coxon quit, fearing extinction. Security experts see a familiar fight”
Jacob Coxon’s Extended Interview - YouTube Video
Fortune - “The ex-Anthropic researcher’s warning that AI could kill us all is missing something important”