4 ms·
Ex-OpenAI safety researcher says he's "terrified" by AI development's pace
- JohnnyLarue 2y agoI'll be terrified when it can actually answer a question without gaslighting me.
- noman-land 2y agoMaybe the fact that it can't but we still shove it everywhere is scarier.
- hindsightbias 2y agoNow I wonder if Drs. Chandra and Forbin went cheap on the training budget.