Anthropic’s quest to study the negative effects of AI is under pressure

December 8, 2025

TL;DR

Hayden Field discusses AI safety with Decoder, focusing on Anthropic's societal impacts team.
The team of nine investigates "inconvenient truths" about AI's societal and economic effects.
Concerns exist about the team's independence and the potential for their findings to be suppressed.
Political pressure from the Trump administration regarding "woke AI" is mentioned.
The situation is compared to past issues with social media companies and their trust and safety teams.
Anthropic CEO Dario Amodei is noted for being amenable to AI regulation.
Anthropic is described as safety-first, formed by former OpenAI researchers concerned about AI safety.
The article questions whether the societal impacts team can meaningfully influence AI development or will eventually disappear, similar to past initiatives.

Continue reading
the original article