Most engaging read I've had for a long time.

Lumor

I really want to make an agent right now. I am working on basically a subconscious that decides what to fill its context window with before it starts responding.

> you could think of Agent-1 as a scatterbrained employee who thrives under careful management.

I love this. This is what AI feels like now. Only, more scatterbrained and *needs* constant and careful management.

k00b

We predict that the impact of superhuman AI over the next decade will be enormous, exceeding that of the Industrial Revolution.