Google cuts quantum errors 40-fold: Is Bitcoin at risk?

Google has demonstrated a 40-fold reduction in quantum computing errors using its Willow processor, advancing a technology that researchers believe could eventually challenge the cryptography protecting Bitcoin.

Google announced the results on October 15, 2026, following the publication of its research in the scientific journal Science. The experiment used a technique called magic state cultivation to produce quantum states with 99.99% logical fidelity, or accuracy, under the researchers’ testing method. 

The achievement addresses a technical problem that has limited efforts to build reliable quantum computers. Although Google’s experiment did not break Bitcoin’s encryption, its findings concern technology that could eventually help future machines perform the calculations needed to attack existing cryptographic systems.

JUST IN: Google’s Willow chip advances quantum computing, raising fresh questions about Bitcoin’s security

A new technique called “magic state cultivation” cuts quantum errors roughly 40x with 99.99% accuracy, but scaling it up remains a challenge before quantum computers could… pic.twitter.com/9gL46tAPJT

— crypto.news (@cryptodotnews) October 16, 2026

The research follows Google’s March assessment that a sufficiently powerful quantum computer could break the type of elliptic curve cryptography used by Bitcoin with fewer than 500,000 physical qubits under specified hardware assumptions.

Google cuts quantum computing errors by 40 times

Google’s latest experiment focused on reducing errors in special quantum states used to perform certain advanced calculations. Known as magic states, they provide resources needed for operations that ordinary quantum error-correction methods cannot perform on their own.

Creating these states has been one of the expensive parts of quantum computing. According to Google’s researchers, some estimates suggest that producing sufficiently reliable magic states could consume as much as 95% of a quantum computer’s capacity.

The new method begins with a quantum state that contains errors and subjects it to repeated checks. Attempts that fail the tests are discarded, leaving the successful ones with substantially lower error rates.

Using the Willow processor, Google demonstrated that the cultivation process reduced errors by approximately 40 times compared with states created without the method. Its researchers reported a record logical fidelity of 99.99% for the resulting T-states, which are used in important quantum computing operations.

Research scientists Emma Rosenfeld and Craig Gidney described the result as the first experimental demonstration of an efficient method for producing these resources on real quantum hardware.

The team developed the approach from theoretical work conducted in 2024 by Gidney, Noah Shutty, and Cody Jones. Turning the proposal into a working experiment required changes to both the hardware controls and the techniques used to measure errors.

During testing, researchers discovered that electrical drift and temperature variations were interfering with their measurements. They introduced new controls and developed a benchmarking method that allowed the cultivation process to help verify its own accuracy.

Google subsequently demonstrated that purified states could transfer between two quantum error-correction methods, known as color codes and surface codes. The experiment successfully moved a state from a distance-3 color code into a distance-5 surface code.

The company cautioned that the technique still needs testing at larger scales before it can support practical quantum computers.

What Google’s quantum advance means for Bitcoin security

Bitcoin uses public-key cryptography to authorize transactions and prove that funds are being spent by their rightful owners. Private keys remain secret, while public keys can be used to verify digital signatures.

A sufficiently powerful quantum computer running Shor’s algorithm could theoretically calculate a private key from an exposed public key. Such an attack would threaten funds protected by existing elliptic curve signature systems.

In March 2026, Google researchers estimated that future machines could break 256-bit elliptic curve cryptography using fewer than 500,000 physical qubits. The research involved hardware assumptions and calculations that have not been demonstrated on a machine capable of attacking Bitcoin. 

The estimate represented approximately a 20-fold reduction from earlier resource calculations. Google’s researchers developed optimized quantum circuits intended to perform the mathematical operations required to recover private keys from public keys.

However, the work still requires millions of quantum operations and reliable error correction. Magic states are needed to support some of those operations, making Google’s latest experiment relevant to the technical requirements identified in its earlier research.

The Willow processor contains 105 physical qubits, far below the resources described in Google’s Bitcoin security study. Its latest achievement involved preparing and verifying quantum states, not extracting cryptocurrency private keys.

As previously examined in Google’s accelerated quantum security timeline, the company has urged organizations to prepare for post-quantum cryptography before capable machines become available.

Google has not announced when its cultivation method will work at the scale needed for practical cryptographic attacks. The company acknowledged that technical challenges remain in expanding the experiment to larger systems.

Glassnode identifies 6.04 million BTC with exposed public keys

The potential threat to Bitcoin depends partly on whether an attacker’s target has a publicly visible cryptographic key.

Blockchain analytics firm Glassnode estimated in May that approximately 6.04 million BTC, equivalent to 30.2% of the mined supply at the time, had public-key exposure while sitting in unspent transaction outputs.

The research separated exposed holdings into two categories. Approximately 1.92 million BTC were classified as structurally exposed because their address formats revealed public keys by design.

Another 4.12 million BTC were considered operationally exposed through practices such as address reuse and partial spending. In certain transaction structures, spending Bitcoin reveals a public key that was previously hidden behind a cryptographic hash.

The figures measure exposure under Glassnode’s methodology, not coins that quantum computers can currently steal. They do not establish that all 6.04 million BTC would be accessible to an attacker under identical conditions.

The findings were covered in an earlier report on quantum computing risks to blockchain security, which examined European regulatory concerns and the concentration of Bitcoin in potentially vulnerable addresses.

Another challenge involves coins whose owners may no longer possess the private keys needed to move them. Researchers have estimated that millions of BTC could remain in old wallets because their owners lost access, abandoned the accounts, or have not moved the holdings for years.

A June study by Quantus estimated that approximately 2.3 million BTC could be difficult or impossible to migrate using ordinary wallet procedures.

The research, previously examined in warnings about lost Bitcoin and future quantum attacks, described potential risks involving dormant wallets and early Bitcoin holdings.

Those figures remain estimates. Public blockchain data cannot conclusively determine whether an inactive wallet’s owner has lost access or simply chosen not to spend the coins.

Coinbase proposes steps to protect Bitcoin from quantum attacks

As researchers continue improving quantum hardware, cryptocurrency companies are examining how Bitcoin could transition to signature systems designed to resist quantum attacks.

On October 13, Coinbase published recommendations addressing Bitcoin’s long-term security, including the treatment of exposed coins, new wallet formats, and the practical challenges of replacing existing cryptographic signatures.

The exchange argued that protecting Bitcoin requires more than selecting a quantum-resistant algorithm. A successful transition would need to consider transaction fees, limited block space, wallet compatibility, custody systems, and the ability of independent nodes to verify transactions.

Coinbase described several possible approaches to coins that cannot move before a future security upgrade. Options included restrictions on spending from vulnerable addresses and recovery mechanisms that could allow legitimate owners to reclaim affected holdings.

The exchange did not present those possibilities as agreed Bitcoin policy. Decisions involving dormant coins and potential restrictions would require discussion and acceptance within Bitcoin’s decentralized development community.

For active wallets, Coinbase supported allowing users to adopt quantum-resistant options through a staged transition. The proposal would give developers and custodians time to test new signature systems before existing protections become vulnerable.

Some Bitcoin developers have already proposed changes through Bitcoin Improvement Proposals, including BIP-360, which introduces a proposed output type intended to reduce public-key exposure in certain circumstances.

Researchers have warned that migrating large numbers of Bitcoin transactions could require several years, depending on the selected signature system, network capacity, and the number of users moving their funds.

A separate June study projected that a migration beginning in 2026 might finish between 2029 and 2031 under its assumptions. The calculation remains conditional, and Bitcoin has not approved a network-wide deadline for switching to post-quantum signatures.

Google’s latest research provides a new experimental result for quantum hardware development, while Coinbase’s October recommendations remain open for discussion among Bitcoin developers, researchers, custodians, and other network participants.