A korrentour reading

A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well.

What is a korrent?

A korrent — “current”, with a K — is our reading of what someone said: one sentence, drawn from their own words below. People put the same thought in different words; a korrent tries to name the thought they share, so that what several people mean can be set side by side, and followed as it changes.

It is an approximation, and it is ours. The quotes are theirs, word for word, and where the two differ, the quote is the fact. Read more →

What Andrej Karpathy actually said

Word for word, from things they published elsewhere, each linked to where it was said. They did not write this page.

AK
Andrej Karpathy 17 Nov 2025

If a task/job is verifiable, then it is optimizable directly or via reinforcement learning, and a neural net can be trained to work extremely well.

↗Verifiabilitykarpathy.bearblog.dev

First recorded here 17 Nov 2025 · How quotes work

Do you hold this korrent?

Where this collides

Supports

How verifiable a task is now predicts how automatable it is, the way specifiability predicted it in the 1980s.

The prediction rests on the mechanism: verifiability predicts automatability because a verifiable task is one reinforcement learning can grind down until the network is very good at it.

See this on the map →