Johannes Gasteiger, né Klicpera
@gasteigerjo
🔸
Safe & beneficial AI.
Favorite papers at Opinions my own.
参加 February 2009
351 フォロー中 3K ファン
My AI Safety Paper Highlights of April 2026:
- *Research sabotage propensity*
- 2 sabotage benchmarks
- Alignment research automation
- Misaligned AI organizations
- Exploration hacking
- Conditional emergent misalignment
More at
もっと見る
