RSS

Vili Kohonen

Karma: 38

In­ves­ti­gat­ing Self-Fulfilling Misal­ign­ment and Col­lu­sion in AI Control

5 Mar 2026 15:05 UTC
15 points
0 comments5 min readLW link

Les­sons from or­ga­niz­ing a tech­ni­cal AI safety bootcamp

28 Sep 2025 13:48 UTC
16 points
3 comments16 min readLW link