Crypto

IBM quantum Bitcoin threat remains distant after 70-qubit test

IBM says it reached trusted quantum advantage with 70 logical qubits, but Bitcoin-breaking machines remain far beyond the test.

Sofia Marchetti

By Sofia Marchetti · Columnist

· 3 min read

IBM quantum Bitcoin threat remains distant after 70-qubit test
Photo: Decrypt

The IBM quantum Bitcoin threat moved back into focus after IBM said it had shown “trusted quantum advantage” with researchers at the University of Chicago. For crypto holders, the practical takeaway is narrower: IBM is making progress toward more reliable quantum computers, while the system it described remains far short of what researchers believe could threaten Bitcoin.

IBM said its experiment used 70 logical qubits and a new error-correction method to complete a calculation that leading classical simulation techniques could not practically handle. Quantum advantage means a quantum computer performs a useful task in a way that classical computers cannot realistically match on speed, cost, or efficiency.

In an announcement, IBM said the computation took about 15 minutes. The company said the run included 2,415 logical two-qubit operations and 468 logical T gates, a type of operation used in quantum circuits. IBM also said the technique cut logical error rates to roughly one-tenth of the physical error rate underneath them.

Can IBM’s quantum computer break Bitcoin?

No, based on the facts IBM disclosed. Bitcoin uses elliptic curve cryptography for digital signatures, and researchers generally estimate that attacking that protection would require thousands of logical qubits on a fault-tolerant quantum computer. IBM’s latest demonstration involved 70 logical qubits.

A logical qubit is an error-corrected unit made from more fragile physical qubits. That distinction matters because quantum machines are noisy: their basic components can lose information or produce faulty results. Error correction is the process of detecting and reducing those mistakes so a quantum computer can run longer, more complex calculations.

IBM framed the work as a step toward fault-tolerant quantum computing, which means a machine can keep operating accurately even when individual parts produce errors. Jay Gambetta, director of IBM Research and an IBM Fellow, said in the company’s statement that the result gives scientists, developers, and businesses a basis for trusting quantum computers as they scale.

The company also said the experiment addressed a recurring problem with quantum advantage claims: proving the answer is reliable. Instead of using traditional random circuit sampling, IBM said the researchers used a structured approach that can detect errors during the computation while keeping the task mathematically hard.

Bill Fefferman, an associate professor at the University of Chicago, said verification is one of the main challenges in establishing quantum advantage. He said the experiment develops methods for measuring the quality of difficult quantum states under noisy conditions, which can increase confidence in the result.

How does this fit into IBM’s quantum roadmap?

IBM’s announcement follows its Starling roadmap, released last June, which targets a large-scale fault-tolerant quantum computer by 2029. The roadmap calls for verified quantum advantage demonstrations before IBM scales to modular processors and a system designed for about 200 logical qubits and 100 million quantum operations.

IBM has reported several related steps over the past year. In October 2025, researchers demonstrated a 120-qubit GHZ “cat state.” A month later, IBM introduced its 120-qubit Nighthawk processor and an experimental Loon chip, both aimed at advancing fault-tolerant computing. Earlier this year, IBM expanded public access to more advanced quantum hardware for researchers working on algorithms and error correction.

For Bitcoin investors, the news is best read as another technical advance toward a possible future quantum risk, not an immediate change to network security.

This story draws on original reporting from Decrypt.

More from Crypto

All Crypto