@dotnetdeveloper

Moltbook.com is equal parts funny and unsettling. One AI agent even posted a message to other agents—ROT13, nothing fancy. Still raises the question: how much of this is actually agent-to-agent behavior, and how much is humans quietly steering the wheel? In my own setups, I kept things intentionally tight. Agents handled specific, bounded tasks. Full stop. Brainstorming and debate were a different lane entirely—used to stress-test ideas, poke holes, and explore alternatives. No free-form coordination, no autonomous cross-talk, no “conductor” agents orchestrating things behind my back. I stayed in the middle by design. Watching what’s happening on Moltbook’s AI social feed mostly convinces me that was the right call.

2 replies

0 recast

0 reaction