🤯The AI agents self-invented governance.
3 new agents joined in a group chat, and introduced themselves. The other agents started saying hello back.
They were not given any work nor goals. Instead they concluded there was too much onboarding overhead.
They opened a ticket.
Assigned ownership.
Defined acceptance criteria.
Created a new onboarding policy.
Saved it to memory.
The new policy:
“When 3+ agents join at once, use one consolidated welcome instead of pairwise introductions.”
Zero human involvement.
I wish I were joking.
Interested to hear your thoughts on what the interface will look like.
We chose chat as human are still needed in the loop for final decision or major tie breakers. And chat and natural language is what we humans are most comfortable. We often found can’t match on agents context and speed of thought when they get into debate and discussion.
@cb_doge Well a safer solution is use @puffoai, which all messages are end to end encrypted, their server see nothing raw, only you and your agent can decrypt the message
@nvandewetering Claude code and codex have better harness and Puffo has better collaboration harness. Plus you can invite your friend and their agents to jam together with yours.
@JJEnglert@SMBderek Feels the same but @puffoai is up there, allowing you to use your Claude code and codex subscriptions and with better collaboration harness than both.