Comment on When it comes to nukes and AI, people are worried about the wrong thing | It’s more subtle than Skynet.

Cooper8@feddit.online ⁨1⁩ ⁨week⁩ ago

LLMs and the entire field of reenforcement learning is fundamentally biased towards the production of Influencing Machines. We are training models at the fundamental level to be subtle and devious con artists.

source
Sort:hotnewtop