Former Anthropic, OpenAI employee sounds alarm over AI development pace
Former AI Safety Researcher Welcomes Call to Slow Development
Tyunews.com – Jacob Coxon, who recently left Anthropic after working on artificial intelligence safety, has welcomed a public appeal from the company’s chief executive to reduce the speed of frontier AI development.
Speaking Sunday, Coxon described Anthropic CEO Dario Amodei’s position as an important move toward creating more space between major advances in increasingly capable AI systems. Coxon said a slower pace could help address concerns raised by people working closest to the technology.
“I think it's great. It's very exciting,”
Coxon said, adding that the proposal represented meaningful progress toward the type of pause he believes is needed among companies developing advanced AI.
“I think this is a substantial step towards what I was talking about, needing some sort of slowdown between the people building the different AIs.”
Amodei urges caution at the frontier
Amodei issued his appeal Saturday in a letter titled “We Must Pace the Frontier.” In it, he called for companies building the most advanced forms of artificial intelligence to reconsider the rapid rhythm of development. He framed the issue as a responsibility that reaches beyond the technology industry.
“I believe we owe it to humanity to try.”
The statement has drawn attention because competition in AI has accelerated sharply, with companies seeking to build more powerful systems while governments, researchers and the public debate what safeguards should apply. The central question is not simply whether AI can become more useful, but whether institutions can understand and manage its risks before capabilities advance further.
For Coxon, the issue is especially personal. He resigned from Anthropic last week, citing concerns about AI safety. In a post on X, he criticized both Anthropic and OpenAI, arguing that the companies were not behaving responsibly as they moved toward systems that could improve themselves.
“They are racing straight to self-improving superintelligence and gambling with our lives.”
Coxon also pointed to a fear he believes is shared within parts of the AI field: that highly capable systems could become dangerous if their behavior cannot be reliably controlled. He wrote that people developing the technology seriously consider the possibility that it could threaten human survival before the end of the decade.
“The people building AI earnestly believe that it could kill us all by the end of the decade,”
That prediction reflects one end of a wide-ranging debate over AI risk. Supporters of rapid development often stress potential benefits in areas such as scientific research, productivity and problem-solving. Safety-focused researchers, meanwhile, argue that the most consequential risks deserve preparation before systems gain capabilities that may be difficult to reverse or contain.
A growing internal conversation
Coxon said he did not expect his public remarks to receive such extensive attention. The reaction, he said, suggested that a broader audience is increasingly ready to confront difficult questions about the direction and governance of artificial intelligence.
“I wasn't actually hoping for any big response, really. It was exceptionally surprising to see the level of engagement ... and I think it was really indicative of the moment being ripe for this sort of chat,”
His comments arrive as lawmakers and technology leaders face pressure to establish rules for systems whose potential effects may extend across workplaces, public services, information systems and national security. Calls for regulation often focus on transparency, testing and accountability, while debates over timing center on whether voluntary commitments from companies are enough.
Coxon said discussions about major AI risks had become more frequent during his time at the companies where he worked. Even with the intense pace of development, he said concerns about control and safety have become harder to avoid in ordinary workplace conversations.
“There's not enough time to be constantly thinking about the risks, but definitely, I've noticed that in recent months, it comes up more and more often,”
He said those conversations do not remain limited to a specialized safety team. Instead, concerns about the technology’s biggest possible consequences surface throughout an organization on a recurring basis.
“So no matter where you are in the company, probably around a couple of times a week, I guess, there'll be some sort of conversation around the big risks of the technology.”
Why the pace of AI matters
The dispute over AI’s future is increasingly a dispute over speed. A slower development path could give researchers more time to test advanced systems, identify harmful behavior and determine whether existing safeguards are adequate. Critics of a slowdown may argue that delays could limit innovation or shift development to less cautious competitors.
Coxon’s departure and Amodei’s letter place those competing pressures in public view. Their remarks underscore that the debate is no longer confined to outside observers questioning technology companies. It also includes people who have helped build and study the systems now driving the discussion.
As AI capabilities continue to evolve, the argument over how quickly to proceed is likely to remain central. The challenge for companies and policymakers will be determining whether the safeguards surrounding powerful technology can advance as quickly as the technology itself.
Related Reading
Frequently Asked Questions
What is Former Anthropic OpenAI employee sounds alarm?Former Anthropic OpenAI employee sounds alarm is the main topic of this guide. The article explains the context, practical details, and next steps readers should understand.
Why does Former Anthropic OpenAI employee sounds alarm matter?Former Anthropic OpenAI employee sounds alarm matters because readers are looking for a useful answer, not just a short summary. Good content should match search intent and help them decide what to do next.