Morning Minute: AI Agents Cut BTC Quantum Attack Benchmark by 86%

More than 100 researchers working alongside AI coding agents cut a benchmark for a key step in a quantum attack on Bitcoin and Ethereum by 86%, according to reporting by Decrypt. The benchmark score fell from 10.75 billion to 1.496 billion between late May and July 26, with the paper landing this week. No coins are at risk today. But the work compresses a timeline that much of the industry's security planning was built around.

The benchmark that got 86% cheaper

Bitcoin and Ethereum both secure signatures with secp256k1, an elliptic curve. A sufficiently powerful quantum computer could reverse that math and derive a private key from a public one. The competition at the center of the story, called ECDSA.Fail and run by Eigen Labs, scored circuit designs by multiplying logical qubits against Toffoli gates, an expensive quantum operation. A lower score means fewer resources needed to run the attack.

The leading design used 1,151 logical qubits and roughly 1.3 million Toffoli gates. A later submission pushed gates below a million. Decrypt notes that the leading score came in at about half of Google Quantum AI's March benchmark, though the two used different counting methods and do not compare cleanly.

The method behind the result matters as much as the number. The researchers call it Open Autoresearch: humans and AI agents iterating against a shared measurable target with a verifier in the loop. The paper's authors came from Eigen Labs, Trail of Bits, StarkWare, Theta Labs, MultiVM Labs, and the Ethereum Foundation. As Decrypt puts it, that is crypto's own security establishment funding and publishing work that makes the attack cheaper, on the theory that knowing the real cost is the only way to plan the defense.

A defense schedule measured in years

The result lands in the middle of a quantum scramble that has been building for weeks. The Ethereum Foundation has set a hard December 2029 deadline to make transactions, validators, and storage quantum-resistant. StarkWare pushed the first quantum-safe Bitcoin transaction to mainnet. Ethereum developers have proposed rebuilding the validator deposit contract, and Ripple is hardening the XRP Ledger.

Money is moving in the same direction. Galaxy committed up to $5 million in July. Nine firms including BlackRock, Coinbase, and Strategy pledged $15 million over three years. The researchers also point to NIST's draft proposing deprecation of classical public-key algorithms at the 112-bit level after 2030 and disallowing them after 2035.

Why the speed of the result matters

Decrypt's framing is direct: the defense side has a fixed schedule measured in years, while the attack side just got 86% cheaper in two months because a hundred people pointed AI agents at it. That is not a prediction about Q-Day, and no funds are at risk today. But every timeline in this space was built assuming attack research moves at human speed. This is the first hard evidence it might not.

For anyone running Bitcoin infrastructure with a multi-year horizon, the relevant shift is not the benchmark itself but the rate of improvement. Hardware purchases, hosting contracts, and treasury policies all assume some stability in how long the current cryptography holds. This result does not break that assumption. It is the first concrete data point suggesting the assumption deserves active monitoring rather than a fixed date on a roadmap.

What to watch next

Watch for further submissions to ECDSA.Fail and whether the sub-million-gate design gets pushed lower. Watch for progress against the Ethereum Foundation's December 2029 deadline, and for whether NIST finalizes its 2030 and 2035 deprecation dates. Any follow-up work using the same Open Autoresearch method will show whether this pace of improvement was a one-off or the new baseline.