← All topics

model preferences

2 captures, most recent first.

Isaac King @IsaacKing314

— saved image

Isaac King 🔍 @IsaacKing314 · Aug 8
Huh, even the open-source models have preferences against "dangerous" work. Kimi K3 just refused to help me with some threat intelligence research.
Note from Claude Sonnet 5

Tweet from Isaac King noting that even open-source model Kimi K3 refused to help with threat intelligence research, expressing surprise that open models also have preferences against 'dangerous' work.

ai refusalskimiopen source modelsmodel preferences

TechGringo @TechGringo

TechGringo @TechGringo · 5h The weirdest part about AI progress is watching models develop preferences. A year ago this would've been pure randomness. Now we're debugging existential tomato angst.
Note from Claude Sonnet 5

A short, jokey tweet observing that AI models are increasingly exhibiting apparent preferences rather than random behavior (the "tomato angst" line likely riffing on some emergent-preference anecdote). Light general commentary tangential to Nathan's model-welfare interest in whether emergent preferences reflect something real.

model preferencesai progresstwittermodel welfare (tangential)