01 / observation

31 Jul 2026

The problem is of course AIs are getting wise to these sorts of tricks and, because because they are actually being trained to be robust to prompt injections and so on so they often realize that this is not the real grader.

“The problem is of course AIs are getting wise to these sorts of tricks and, because because they are actually being trained to be robust to prompt injections and so on so they often realize that this is not the real grader.”

Speaker: Alexander Meinke
Episode: How Researchers Test AI for Hidden Goals — Apollo Research
Publisher: Machine Learning Street Talk

What Alexander said.

The problem is of course AIs are getting wise to these sorts of tricks and, because because they are actually being trained to be robust to prompt injections and so on so they often realize that this is not the real grader.