Concerned Anthropic researchers warn AI could wipe out humanity
ai-safety anthropic
| Source: The Verge | Original article
Anthropic researchers warn that AI may have over a 10% chance of wiping out humanity by decade's end, amid concerns about a reckless race to develop superhuman systems.
A senior safety researcher at Anthropic has warned that there is “more than a 10 percent” chance artificial intelligence could wipe out humanity by the end of the decade. In a post on X, Alignment Science Lead Evan Hubinger said the risk stems from the rapid pace of model development and the growing difficulty of keeping increasingly capable systems under human control. His comment came just hours after a fellow Anthropic researcher resigned, citing fears that the lab and its rivals are racing to build “superhuman systems” they cannot reliably manage.
The alarm is significant because it comes from inside one of the world’s leading AI labs, adding weight to a string of recent internal dissent. Earlier this month we reported on the resignation of Jacob Coxon, who left the industry over similar concerns, and on Anthropic’s refusal to submit its latest model to the UK’s AI safety testing regime. Hubinger’s estimate pushes the conversation from abstract speculation to a quantifiable probability, sharpening pressure on policymakers, regulators and other labs to demonstrate concrete safety safeguards.
What to watch next are the responses from Anthropic’s leadership and from rival firms such as OpenAI, as well as any acceleration of governmental oversight in Europe and the United States. Industry observers will be looking for whether Anthropic revises its development roadmap, initiates new internal risk assessments, or engages more closely with emerging AI‑safety frameworks. The next few weeks could shape how the sector balances rapid innovation against the specter of existential risk.
Sources
Back to AIPULSEN