Joe Collman

Karma: 1,696

Existing Safety Frameworks Imply Unreasonable Confidence

Joe Rogero, yams and Joe Collman

10 Apr 2025 16:31 UTC

46 points

3 comments15 min readLW link

(intelligence.org)

Truthfulness, standards and credibility

Joe Collman7 Apr 2022 10:31 UTC

12 points

2 comments32 min readLW link

Review of “Learning Normativity: A Research Agenda”

Gyrodiot, adamShimi and Joe Collman

6 Jun 2021 13:33 UTC

37 points

0 comments6 min readLW link

Review of “Fun with +12 OOMs of Compute”

adamShimi, Joe Collman and Gyrodiot

28 Mar 2021 14:55 UTC

65 points

21 comments8 min readLW link 1 review

A Critique of Non-Obstruction

Joe Collman3 Feb 2021 8:45 UTC

13 points

9 comments4 min readLW link

Optimal play in human-judged Debate usually won’t answer your question

Joe Collman27 Jan 2021 7:34 UTC

33 points

12 comments12 min readLW link

Literature Review on Goal-Directedness

adamShimi, Michele Campolo and Joe Collman

18 Jan 2021 11:15 UTC

80 points

21 comments31 min readLW link