Researcher Andre Schlottenlauer has reconstructed and published a quantum circuit stored by Google Quantum AI. paper Last March, it claimed that the quantum resources needed to attack the cryptography that protects Bitcoin’s digital signatures would be reduced. According to Schrottenloher’s research, which was shared on the professional website arXiv on June 1 of this year, the results were slightly more efficient than those reported by Google.
Schlottenlauer’s report is based on a scheme called secp256k1 (a specific elliptic curve that Bitcoin uses for digital signatures), according to the researchers. Compared to Google research, Toffoli doors yielded savings of 6.5% to 10%.using only 1.5% additional qubits (quantum processing units equivalent to classical bits).
The Toffoli gate is the most computationally intensive operation in Shor’s quantum algorithm (which can potentially derive the Bitcoin private key from the public key) and determines how long it takes to perform the attack. Reducing the number of Toffoli doors theoretically means: Faster attacks or executables with fewer resources.
However, Schlottenlauer’s report states that Physical hardware estimates from Google Quantum AI research are not updated The proposed attack time is also less than 9 minutes. The impact of reducing Toffoli gates on Bitcoin will depend on its physical architecture, which was not specified in Schlottenlocher’s study. Additionally, this researcher’s work has not been peer-reviewed at the time of publication.
What was written and hidden in the Google Quantum AI paper?
A Google Quantum AI study published on March 30 estimates that a quantum computer could decrypt the Bitcoin public key in less than 9 minutes using fewer than 500,000 physical qubits (the fundamental quantum processing unit), which is Almost 20x reduction compared to the most efficient previous estimateas reported by CriptoNoticias.
However, Google did not reveal the quantum structure that would enable such an attack. Instead, he published a zero-knowledge (ZK) proof. This is a cryptographic technique that made it possible at the time to verify the existence of a circuit and produce a declared result without displaying it.
Similarly, security firm Trail of Bits discovered a vulnerability in this ZK-based validation tool that allows it to generate cryptographically forged tests that are indistinguishable from legitimate tests. Google has patched the code and confirmed that scientific conclusions are not affected.
Google research evidence
Sreeram Kannan, founder of EigenCloud, explained in a report also published on June 1 that quantum computing was used by an undergraduate student with no training in quantum computing. The AI agent improved the circuit published by Google with about twice the efficiency compared to the best results before the Google Quantum AI paper.
A few days later, Kannan said, an 18-year-old researcher used his own AI agent system and spent $10,000 on computing to reach 80% of Google’s unpublished results. This percentage shows how close we came to replicating the efficiency of the most advanced circuits known to attack Bitcoin encryption, without access to Google’s original circuits or specialized training in quantum computing.
Kannan’s report says the research community has gone further, improving Google’s circuit by 8.4%, as measured by the combination of qubits and operations required to carry out the attack.
Alex Thorne, Galaxy’s head of research, assessed the scope of the development as follows: “But this shows the power of using swarms of agents to decentralize research.”.
Thorne also emphasized that “Google kept the circuitry in-house.” paper The goal of March 31st was specifically to avoid giving an adversary a functional attack, but it turns out that most of what it takes to build a line that brings large numbers of people close to the same perimeter is a publicly verifiable goal.
Charles Guillemet, chief technology officer at Ledger, summed it up by saying, “What has changed is the integrity of each published post-quantum schedule. Trust remains intact even when attacks are carried out. “Trust is lost when the fundamentals appear thinner than the public record suggests, and now the public record is clearly thinner than reality, at one end through classification and at the other end through AI-powered re-derivation.”
Since there are currently no quantum computers capable of running these circuits at scale, neither Guilmet nor Thorne believe that Schlottenlocher’s work will be an immediate breaking point for Bitcoin. However, while the community is discussing potential risks, continued development in this area could accelerate the arrival of “Q-day.”
(Tag translation) Bitcoin (BTC)
