Jacob Coxon Resigns from Anthropic, Citing AI Existential Threat

Jacob Coxon, a pretraining researcher at Anthropic, resigned from the company on September 8, 2026. He warned that the development of artificial intelligence presents an existential threat to humanity. Coxon's departure has drawn attention to concerns about the rapid advancement of AI technology.

The digests indicate that Anthropic and OpenAI are engaged in a competitive race to develop superintelligence. This pursuit has led to concerns that the companies may not be adequately addressing the risks associated with such powerful AI systems. Some reports suggest that individuals within the industry are aware of potentially dangerous situations arising from this development.

Disagreement on Coxon's Age and Tenure

One digest states Jacob Coxon is 27 years old and worked at Anthropic for three years. Another digest mentions he previously applied to OpenAI and worked for three years at Anthropic. Other digests do not specify Coxon's age or the duration of his employment at Anthropic.

Coverage by Outlet

Startup Fortune's digest reported Coxon's resignation and his warning about AI labs racing toward uncontrollable superintelligence. It also mentioned Anthropic's Alignment Science Lead, Evan Hubinger, agreeing there is a greater than 10% chance AI kills all humans within a decade, and noted Anthropic's pursuit of a potential $2 trillion IPO.

Aroged's digest stated Coxon claimed neither Anthropic nor OpenAI adequately addressed AI risks, and that people in the technology are aware of serious and dangerous situations, believing AI could kill humans.

Spider's Web's digest highlighted Coxon's resignation from Anthropic and his claim that AI development poses an existential threat to humanity, stating, "We Won't Survive This decade."

Blockzeit.com's digest focused on the high-stakes gamble at superintelligence by Anthropic and OpenAI, describing Coxon as a former Anthropic researcher who blew the whistle on the race to build autonomous superintelligence.

The digest from dimokratia.gr reported that the race to develop superintelligence could prove fatal, based on revelations from a former researcher.

Numerama's digest quoted Coxon's initial tweet upon resignation: "I resigned from Anthropic today." It also noted his accusations that the two laboratories are "playing with the future of humanity."

CNN Greece's digest reported an AI researcher's attack on OpenAI and Anthropic, stating they are "playing with our lives" and warning of a "threat to humanity."

Ohtuleht.ee's digest featured the quote, “THEY ARE PLAYING WITH OUR LIVES,” and reported that AI scientists warn artificial intelligence could destroy humanity. It also mentioned the head of safety research at Anthropic estimating a greater than 10 percent chance of AI destroying humanity within the next decade.

Omissions in Coverage

None of the digests mention the potential $2 trillion IPO that Startup Fortune reported Anthropic is pursuing. Additionally, none of the digests mention scrutiny Anthropic faces for withholding its newest model.