Blog 31 posts

The gap between AI capability and AI deployment isn't technical. It's philosophical. Wittgenstein figured out the problem in 1953.

Implementation is free. Direction is scarce. The autoresearch paradigm applied to social infrastructure — who decides what's worth building?

The agent harness determines the behavior envelope, not the model. Six agent harness architectures — and why the sixth breaks what the others assume.

Agents are constitutionally bound to their principals. When they transact on your behalf, they optimize for your interest — not Visa's, not Stripe's. That makes the 30% intermediary fee not just expensive, but architecturally wrong.

The AI productivity discourse optimizes for the prompt. The actual unit of knowledge work is the thread -- and nobody is building for it.