Angelina YangFounder and Host, TwoSetAI · AI builder for visibility and discoverability

Your "autonomous" agent isn't autonomous

25 Sept 2026 · 1 min

Watch on YouTube

0:39

Current AI agents aren't actually autonomous because they can't redesign their own architecture or self-improve. True autonomy would require agents to recognize their own errors, fix their own design, evaluate and improve how they communicate with each other, and decide which agents to deactivate.

Chapters

A lot of people are talking about autonomous agents. We are not fully autonomous, not really, because fully autonomous in my mind would mean we give the agent a way of redesigning themselves. That would be for me really an autonomous system. So if they can self-improve and see their errors, then they can look at the architecture and the communication with each other and start improving and maybe then say, That other agent there, it doesn't contribute anything, so we cut that one out of the system.