RSS

jakub_krys

Karma: 87

Quan­ti­ta­tive AI risk as­sess­ment: a start­ing point

Henry Papadatos, jakub_krys, malcolmmurray and Renn Karageorgieva
27 May 2026 9:42 UTC
38 points
7 comments11 min readLW link
(www.safer-ai.org)

This prompt (some­times) makes ChatGPT think about ter­ror­ist organisations

jakub_krys24 Apr 2025 21:15 UTC
30 points
13 comments1 min readLW link

Not all bi­ases are equal—a study of syco­phancy and bias in fine-tuned LLMs

jakub_krys11 Nov 2024 23:11 UTC
8 points
0 comments7 min readLW link

[Linkpost] Hawk­ish na­tion­al­ism vs in­ter­na­tional AI power and benefit sharing

jakub_krys and Naci Cankaya
18 Oct 2024 18:13 UTC
7 points
5 comments1 min readLW link
(nacicankaya.substack.com)