x

LESSWRONG

LW

William Harrison — LessWrong

William Harrison

William Harrison

Message

Aspiring AI safety researcher

Organizer at Stanford AI Governance and Alignment

7

1

4d

William Harrison

Aspiring AI safety researcher

Organizer at Stanford AI Governance and Alignment

OpenAI's Astra alignment claims are dubious and there is good evidence it is misaligned

OpenAI touts Astra as their "most aligned model." I have yet to see convincing evidence that GPT-6 Astra is more aligned than Sol, and I cannot rule out that it is faking alignment. And even if it wasn't faking alignment, it could still pass many of the safety tests OpenAI...