3 ms·
[I mean no bad faith in this comment, I'm a fan of yours.] Why answer questions about harmlessness/safety in such a roundabout way? Both OpenAI and Anthropic a
by dpaleka 4y ago
[I mean no bad faith in this comment, I'm a fan of yours.]
Why answer questions about harmlessness/safety in such a roundabout way?
Both OpenAI and Anthropic are clear about what words like "safe" are intended to mean: a stepping stone to "AI does not kill all people when given control".
Avoiding to state this clearly only invites unnecessary culture war disagreements in every discussion about these models.
- goodside 4y agoMaybe you’re right. It’s partially laziness on my part — it takes a while to explain long-term issues, and those who are inclined to care about them are generally aware of who started Anthropic and why.