4/11/2026 at 10:53:10 AM
I think it would be valuable to list all interactions with the LLM by the dev team and transparently state what was induced by human steering the LLM, and what was actuall LLM decision, which was not biased by system instructions or dev team communicating with itby Xx_crazy420_xX
4/11/2026 at 1:35:00 PM
Agreed. Color me skeptical. All of the interactions and decisions described are plausible, but in my experience with AI agents, they would require frequent human intervention.by vannevar