@nprofile1q... this originally started as “what would LLM-aided decision-making modeled after coop-y structures look like” and then spiraled very quickly and badly from there