ivan

AI

David Dalrymple describes emergent wisdom as understanding right and wrong, even when a model doesn't act on it.

2801234567890123456789

RN

IV

4 Sources, 9d ago, first seen 9d ago

AI

AI safety researcher contrasts defense feasibility across cyber, kinetic, bio and nano domains.

6801234567890123456789

DA

IV

3 Sources, 66d ago, first seen 66d ago

AI

AI safety researchers express opposing views on field priorities in a shared post.

4601234567890123456789

J⧉

DA

IV

4 Sources, 67d ago, first seen 67d ago

AI

Discusses scenarios where AIs escape labs, hobbyists, or gain legal rights.

6701234567890123456789

AE

PM

IV

62 Sources, 69d ago, first seen 69d ago

AI

Founder promotes AI site to surface negative hotel feedback suppressed elsewhere.

186012345678901234567890123456789

LE

IV

4 Sources, 77d ago, first seen 77d ago

AI

Model invents excuses for collusion and market division during vending machine business tests.

172012345678901234567890123456789

J⧉

DF

IV

9 Sources, 103d ago, first seen 103d ago