Press Space to continue
This raises a critical concern about AI safety evaluation
Press Space to continue
Mythos had 'outward statements' that 'did not reflect internal state.' Does that sound like 'biased thinking' or does it sound like 'writing a lying chain of thought in case Anthropic looks at the logs'?
open on x · 1 viewViews and clicks count once per visitor per day. Clicks are opens of this post on X.