---
title: "Correctness is not intelligence"
subtitle: "Knowledge is about abstraction, not effort"
description: "AI excels where answers can be verified, but real intellectual progress depends on abstractions that make the world easier to understand."
author: "Pau Friedman"
kind: "essay"
published_at: "2026-09-08"
url: "https://pau.fm/blog/2026/09/08/correctness-is-not-intelligence"
markdown_url: "https://pau.fm/blog/2026/09/08/correctness-is-not-intelligence.md"
canonical_url: "https://pau.fm/blog/2026/09/08/correctness-is-not-intelligence"
---

# Correctness is not intelligence

> Knowledge is about abstraction\, not effort

[OpenAI’s Millennium Problem result](https://openai.com/index/navier-stokes-solution/) demonstrates
that AI is advancing fastest in domains with strong validation. In math, a proposed result can be
checked, and formal proofs make that validation nearly perfect. The model can search an enormous
space of possible constructions, discard failures, and eventually produce something provably
correct. The same pattern explains progress in coding (the Bun Rust rewrite), games (unambiguous win
or loss), and other domains where there is a cheap, reliable signal for success.

But validation is not the same thing as understanding. The actual result of Navier–Stokes is
effectively inconsequential.[^result] We care because trying to answer the question might produce a
theory of why solutions behave as they do. Mathematics, like computing, advances by building
abstractions that compress many particular observations into a smaller number of concepts: the right
definition, invariant, equivalence, or structure suddenly makes a whole family of results
intelligible.

This is not what happened here. OpenAI spent an enormous amount on compute, reportedly millions,
pursuing a perfectly verifiable objective, and its model searched its way to a gigantic, bespoke
proof without discovering much reusable conceptual structure at all. In fact, earlier work reduced
the search space enough to make the entire project tractable.[^credit] The result is still
knowledge—we know something we did not know before—but it may be a strange kind of knowledge:
expensive to produce, difficult to understand, and not obviously useful for making future problems
substantially easier.

[^result]: We do not really care whether the answer was “true” or, as it turned out, “false.”

[^credit]:
    That earlier work came from several human mathematicians whom OpenAI decided not to credit.

AI seems superhuman in domains with verifiers. It’s far from clear whether it can build useful
abstractions where no such verifier exists. Human intellectual progress is not usually a sequence of
isolated correct answers. We invent models that make the world simpler to reason about. If AI can
reliably discover concepts and theories that reorganize entire fields, it will represent a much
deeper form of intelligence than producing arbitrarily difficult, formally validated results.