New research suggests the computational burden of a potential quantum attack against Bitcoin and Ethereum could be substantially lower than previously estimated. A paper shared with CoinDesk reports that researchers, working with AI coding agents, have improved on a key calculation used by Shor’s algorithm, beating a benchmark published by Google in March.
The estimated workload for this part of a future quantum attack has dropped by more than 50% compared with Google Quantum AI’s earlier result.
Researchers from the Ethereum Foundation, Theta Labs, StarkWare and several other organizations developed a quantum circuit requiring 1,151 logical qubits and approximately 1.3 million Toffoli gates.
The two measurements provide a way to describe the scale of the quantum computation. Logical qubits indicate the size of the machine required, while Toffoli gates represent the amount of computational work that machine must perform.
The new design produced a score of roughly 1.5 billion, more than half below the approximately 3 billion figure reported by Google Quantum AI in March. However, the researchers noted that the two results are not directly comparable in every respect because the underlying designs use different interfaces and accounting conventions.
A second circuit designed to more closely reflect the way the calculation would be incorporated into Shor’s algorithm produced a score of approximately 1.96 billion, which was still below Google’s benchmark.
Researchers target a key part of Shor’s algorithm
The team focused its optimization efforts on point addition, a mathematical operation that Shor’s algorithm performs repeatedly.
Shor’s algorithm is significant for cryptocurrency security because a sufficiently powerful quantum computer could potentially use it to derive a private key from an exposed public key. That could allow an attacker to generate valid signatures and authorize transactions as though they were the wallet owner.
Bitcoin and Ethereum both rely on the secp256k1 elliptic-curve cryptographic system targeted by the research.
The work involved more than 100 participants who spent roughly eight weeks tackling the problem through ECDSA.Fail, an open research challenge launched by Eigen Labs. The effort generated more than 400 accepted submissions, with successful improvements providing new starting points for subsequent participants.
AI coding agents played an important role in implementation, repeated testing and incremental optimization. Human researchers generally focused on selecting research strategies and making larger architectural changes. The paper does not attempt to quantify precisely how much of the progress came from people compared with AI systems.
The findings suggest that progress toward a quantum attack does not depend solely on advances in quantum hardware. Reducing the computational requirements could also lower the size and capability of a machine needed to execute the attack.
The research does not mean Bitcoin can be cracked today
The researchers emphasized that their work does not represent a complete quantum attack against Bitcoin or Ethereum.
The circuit addresses one major computational component but does not include physical error correction, the complete Shor algorithm or several hardware-specific expenses that would arise when running the computation on an actual quantum machine.
No quantum computer currently exists that can use the researchers’ circuit to compromise Bitcoin or Ethereum.
The estimates have also continued to improve. The paper’s primary results were based on a July cutoff, but a later design reduced the score further to approximately 1.26 billion.
Another design lowered the required machine size to 813 logical qubits, although it required substantially more computational work.
Jieyi Long, the paper’s lead author and CTO of Theta Labs, told CoinDesk that the developments are not evidence of an imminent attack. Instead, he said the concern comes from the lengthy preparation required to protect cryptocurrency systems.
The research arrives as the U.S. Commerce Department finalized CHIPS Act awards of up to $100 million each for Rigetti, D-Wave and Quantinuum. The government is taking minority stakes in the three companies, with the funding intended to support development of larger fault-tolerant quantum machines.
For Bitcoin and Ethereum, the significance of the latest research is therefore less about an immediate security breach and more about the timeline for preparing defenses. As researchers continue reducing the computational requirements for quantum attacks, the amount of hardware progress needed to make such attacks practical could also decrease.
About The Author
More Stories
Treasury Yields Climb as Bessent Doubles Down on Bond Buybacks
Bitcoin Holds Near $78K as Memecoins and Small Caps Lead Crypto Slide
SGX Opens Bitcoin and Ether Perpetual Futures to U.S. Institutions