OpenAI claims cryptography breakthrough, scrutiny remains open

The announcement on October 26 saw researchers begin an independent review of the claim. It is important to note that the disclosed result does not establish a practical attack against Bitcoin.

The core of the claim involves 3-SAT, where the hypothesis suggests solving these puzzles requires exponential time in the worst-case scenario based on the number of variables.

OpenAI states its AI system refuted this ETH, and the finding was verified within Lean, a proof assistant. However, the scope note details a weaker time bound than the manuscript's sharpest assertion.

The manuscript, dated October 17 and revised on October 25, proposes a deterministic algorithm that counts satisfying assignments for fixed-width Boolean formulas in subexponential time and polynomial space.

Independent reviewers have explicitly stated that their participation should not be interpreted as certification of the claim, calling for careful scrutiny.

A theoretical speed-up in SAT counting does not equate to a functional advantage in Bitcoin mining or spending. Mining involves searching for a block-header hash, while signatures authorize spending—these are distinct processes.

A subexponential SAT bound alone does not demonstrate how to defeat concrete mechanisms like deriving a secp256k1 private key from its public counterpart; a practical attack requires a concrete encoding and cost comparison.

Why it matters

While the claim is significant in theoretical computer science, readers should note that the disclosed subexponential SAT bound does not equate to a practical threat to Bitcoin's security. Further concrete steps and independent cryptographic audits are required before any market impact can be considered.