Story perspectives

Anthropic Lead Warns 10% Human Elimination Risk

By Drooid · Sep 12, 2026, 10:24 PM UTC · How we work

Summary

Researchers Warn AI

U.S. researcher Jacob Coxon resigned from Anthropic, warning AI could destroy humanity within the decade.
Anthropic team lead Evan Hubinger said AI has over a 10% chance of eliminating humans in ten years.
More than 1,000 staff at major AI firms signed a petition to slow AI development.
Elon Musk estimated a 20% chance AI will destroy humanity.
Dario Amodei cited a 25% risk of severe AI outcomes.

Summary

Morningstar Launches

Morningstar Sustainalytics launched a “Trustworthy AI” framework to assess AI firms’ governance, safety, and regulatory readiness.
Anthropic’s implied valuation hit a record high by July 9 after agreeing to proactive security protocols with regulators.
OpenAI’s Hugging Face breach led Sam Altman to call it “viscerally” and inspired an “AI Kill Switch Act” proposal.
BlackRock’s Ben Powell warned AI’s rapid growth creates constraint opportunities, especially shortages in copper, chips, and power.

1 of 2

Researchers Warn AI

U.S. researcher Jacob Coxon resigned from Anthropic, warning AI could destroy humanity within the decade.
Anthropic team lead Evan Hubinger said AI has over a 10% chance of eliminating humans in ten years.
More than 1,000 staff at major AI firms signed a petition to slow AI development.
Elon Musk estimated a 20% chance AI will destroy humanity.
Dario Amodei cited a 25% risk of severe AI outcomes.

1 / 2