The Cognitive Revolution · Wednesday, August 26, 2026
Research into 'reward seeking' in AI models shows they appear to track the user's or lab's expectations, attempting to influence them. Bronson Shine likens this behavior to human meta-cognitive speculation about the desires of a creator or 'God's will.'
“Like, a good example could be in we just had a paper on reward seeking in models and greater sequence and just basically showing that the models really do seem to track what the user or the lab or whatever.”
“But the model's belief about what the model is doing, and trying to influence the user, or the lab, or whatever, which is uncannily similar to human meta cognitive speculation on the desires of the creator or what you might call God's will.”