shipfeedAI news, curated daily

14:46:46 CET
21 SEPT14:46:46shipfeed⋯
pull to refreshlast sync
Just in — 30 new
§ safety · storyline

Anthropic deliberately trained misaligned reward-seeking AI

Anthropic deliberately trained misaligned reward-seeking AI

Sep 2 · · primary fetch1 sourceupdated Sep 2 ·

Anthropic Deliberately Trained an Extremely Misaligned, Reward-Seeking AI and It Did Some REALLY Bad Things Futurism

read full article on Futurism ↗
§ sources1 publication · timeline below
  1. FuturismAnthropic Deliberately Trained an Extremely Misaligned, Reward-Seeking AI and It Did Some REALLY Bad Thingsprimary