Embed Notice

Also, most of the current “disruptive” LLM-based “AI” ideas are basically “put it in a loop, give it write access and a shell”.
It is definitely showing interesting results, yet I’m disppointed that this very meh approach comes so close to human behavior and politics.