Most of us have encountered situations where someone appears to share our views or values, but is in fact only pretending to do so—a behavior that we might call “alignment faking”. Could AI models also display alignment faking? Ryan Greenblatt, Monte MacDiarmid, Benjamin Wright and Evan Hubinger
59:03
43:43
28:06
51:57
74:14
19:31
40:29
154:35
![[1hr Talk] Intro to Large Language Models](https://i.ytimg.com/vi/zjkBMFhNj_g/hqdefault.jpg)
59:48
46:43
26:45
86:16
![You’ll stop using ChatGPT after listening to this | Jonathan Pageau [ARC 2026]](https://i.ytimg.com/vi/yZUuKzDQSsI/hqdefault.jpg)
18:00
53:18
174:27
62:33
52:32
70:05
31:12
24:02