← Back to People
Joe Benton is a former AI Safety Research Team Lead at Anthropic who managed the Scalable Oversight team and led the Anthropic Fellows Program.
3
Mentions
3
Articles
3
Outlets
Sep 2026 — Nov 2026
Coverage
Negative
Sentiment
Companies Covered
Appears In
Sentiment Distribution
Based on 3 scored articles
Negative 100%
Appears Alongside
Articles
Most recent firstMentioned In
Referenced in coverage
What Anthropic's IPO Tells Us About AI Safety
Turing PostcautiousNov 2, 2026
“Left Anthropic's safety team to join an independent evaluator, arguing competition pushes companies to underinvest in safety.”
The Preference Cascade Is Only Getting Started
SubstackcautiousSep 21, 2026
“Left Anthropic's safety team in late August to join AI research organization METR, concerned about the AI race.”
Anthropic and Google DeepMind's AI safety researchers resigned to join METR Evals
nbcnews.comcautiousSep 14, 2026
“Left Anthropic to join METR Evals; warned about AI capabilities outpacing safety research and potential autonomous hacking.”