"A soundness bug in the Lean kernel (#14576) was reported and fixed during the week of July 27. [...] On July 25, Ramana Kumar published a repository containing a sorry-free "disproof" of the Collatz conjecture, produced with AI assistance. It is not a valid proof because it exploits a bug in the kernel's handling of nested inductive types. On July 28, Kiran Gopinathan reduced it to a small proof of False and opened issue #14576."

And everyone who's been into this stuff for a while has had their prediction come through.

If AI is water, Lean is the pipe and collatz is a clog on one end, then surely we'll find the cracks.

So essentially: One cannot trust the code produced by an LLM, even if the code is a formal proof passing the verifier.

> The practical consequence: checking with an independent kernel still works, since it required two distinct bugs in two implementations, but users who rely on it need current versions of both.

Things like this aren't too surprising, given that even much simpler type checkers like Rust's have soundness issues occasionally. I think it's very important to view verified results not as an absolute and unbreakable guarantee, just an extraordinarily strong one where (1) the surface area for soundness issues has been painstakingly minimized and (2) any realized soundness issues are taken very seriously and fixed in short order.

Isn’t a disproof of the Collatz conjecture easy to check as it should just be a counterexample? Or is the proof not constructive?

Lean has Claude contributions, what do you expect!

Use Coq or Isabelle or any other decent theorem prover. Lean is just hyped.