Post

Jacob Herrington (not verified proceed with caution) ⚠️
@jacobherrington.dev

I’m very wary when people anthropomorphize LLM behavior (i.e., self-reflection, motivations). It’s not dissimilar to presenting LLMs as an AGI-lite.

Carl T. Bergstrom@carlbergstrom.com

As I've written elsewhere: an LLM doesn’t have the metacognitive abilities to reflect on its past actions and report the motivations underlying them. With no clue why it did whatever it just did, the LLM is forced to guess wildly at a plausible explanation like ill-fated Leonard Shelby in Memento.

0 reposts4 likes