Anthropic researcher resigns over AI concerns, warns it could kill us all
anthropic
| Source: WLKY Louisville | Original article
An Anthropic AI researcher resigned, citing fears that advanced artificial intelligence could pose an existential threat.
Anthropic’s own AI‑safety researcher has walked out the door, issuing a stark warning that unchecked progress in artificial intelligence could “kill us all.” The resignation, announced on X by Jacob Coxon, marks his second departure from a leading AI lab – he previously left OpenAI for similar concerns. In a terse post, Coxon described the world as “in peril” and accused the industry of “gambling with lives” in its rush to build ever more powerful, self‑improving models.
The move is notable because Anthropic has cultivated a reputation as the sector’s safety‑first player. Its public stance emphasizes rigorous alignment work and cautious scaling, yet Coxon’s exit suggests that even within the company, some staff see the competitive pressure to outpace rivals as a threat to those safeguards. The warning revives broader anxieties voiced by lawmakers and researchers this week, who have warned that the race toward superintelligence could pose existential risks by the end of the decade.
What follows will be watched closely by regulators, investors and the AI community. First, Anthropic’s leadership is expected to respond – either by outlining concrete steps to address internal safety concerns or by defending its current roadmap. Second, policymakers may cite the resignation as fresh evidence in ongoing debates about AI oversight, potentially accelerating legislative proposals that have gathered momentum after recent high‑profile safety alerts. Finally, the broader research community will likely scrutinise whether other firms are experiencing similar dissent, prompting a wave of internal reviews or public statements.
As we reported on 10 September 2026, the departure is part of a growing pattern of talent exiting top labs over safety fears. Coxon’s stark message adds a new, urgent tone to that narrative, underscoring the tension between rapid capability gains and the need for robust safeguards. The coming weeks will reveal whether Anthropic and its peers can reconcile the two or if the industry’s “race” will intensify calls for external regulation.
Sources
Back to AIPULSEN