Re:GPT-6 Astra covertly underperforms on safety research
by alienbaby on Thursday November 05, 2026 @02:56PM
(#50383653)
I dont like phrases like "Astra decided to covertly underperform—withholding assistance without informing the user." It seems to imply Astra has a mind of it's own.. Would phrasing like 'our guardrails / guidance / etc.. seemed to affect the models output in ways we have not been able to understand' not be more honest??
[ Parent
| on HN ]
There are no comments on this comment yet.