Almost everything that matters in the modern world is kept in a kind of vault. bank documents. medical backgrounds. private correspondence. intelligence from the country. the passwords you use too frequently. That vault has been protected for decades by mathematics so intricate that it would require ten thousand years of nonstop operation of all Earth’s computing power to crack it. It seemed to last forever. secure. Resolved. That’s not how it feels anymore.
Google’s Quantum AI team published research last month that subtly changed the foundation of the whole digital security sector. The study revealed that a form of encryption based on mathematical structures known as elliptic curves, which is utilized by Bitcoin, Ethereum, and numerous other secure communications systems, may be far more susceptible to quantum attack than previously thought. Not in a theoretical, far-off sense.
| Category | Detail |
|---|---|
| Technology | Quantum Computing |
| Core Threat | Breaking widely used public-key encryption (RSA, elliptic-curve) |
| Key Term | “Q Day” — the projected date quantum computers can crack modern cryptography |
| Major Players | IBM, Google Quantum AI, PsiQuantum, Caltech–Berkeley–Oratomic |
| IBM Milestone | 120-qubit chip unveiled late 2024; fault-tolerant system targeted by 2029 |
| Google Finding (March 2026) | Fewer than 500,000 qubits may crack elliptic-curve encryption — ~10x less than earlier estimates |
| Caltech–Berkeley–Oratomic (March 2026) | ~26,000 atomic qubits could break Bitcoin encryption in days |
| NIST Deadline (USA) | Transition to post-quantum cryptography largely completed by 2035 |
| Australia Deadline | Australian Signals Directorate urges transition by 2030 |
| Defenses Available | NIST-standardized post-quantum algorithms; Google Chrome and Cloudflare already deploying hybrid protections |
| Cryptocurrencies at Risk | Bitcoin, Ethereum (both use elliptic-curve cryptography) |
Practically speaking, count the qubits, build the machine. According to the study, a quantum computer with fewer than 500,000 physical qubits could break that encryption in a matter of minutes. That seems big until you consider that previous estimates put the figure at about five million. The bar simply fell ten times.
It’s difficult to ignore how silently this occurred. No press conference in an emergency. No alert about breaking news. Just a carefully examined research paper that slipped into the public record while most people were preoccupied with stock markets, artificial intelligence, and the daily cacophony.
There are two lanes in the race, and both are speeding. In terms of hardware, IBM unveiled a 120-qubit chip late last year with the goal of achieving a true quantum advantage over classical computers in specific tasks. A more potent fault-tolerant system is scheduled for 2029.
Given that Google is both one of the organizations creating the threat and one of the first attempting to counter it, it is noteworthy that Google has been pushing alongside them and announcing plans to accelerate its adoption of post-quantum cryptography in its own systems.
There are also more recent players joining. PsiQuantum is utilizing traditional chip manufacturing infrastructure to pursue a light-based qubit approach. A few years ago, it would have seemed unthinkable that experimental neutral-atom platforms could control thousands of qubits in a lab setting.
Then there is the algorithmic aspect, which may be equally important but receives less attention. According to a preprint published in March 2026 by a partnership between Caltech, Berkeley, and a business called Oratomic, Shor’s algorithm—named for the mathematician Peter Shor who developed it in 1994—may be able to crack RSA encryption using as few as 10,000 to 20,000 atomic qubits.
According to the paper, one suggested design that uses about 26,000 qubits could crack Bitcoin’s encryption in a matter of days. The direction of the discovery is what counts, even though RSA with a 2048-bit key would take longer and require more. Even before the necessary hardware is available, codebreakers are becoming more effective.
Q Day is a term that has become popular in policy and security circles. It describes the point at which a quantum computer is able to crack the encryption systems that safeguard the majority of the global digital infrastructure. For many years, it was viewed as a far-off horizon that needed to be watched but not immediately addressed. That mindset is changing.
The National Institute of Standards and Technology in the United States has suggested that the shift away from quantum-vulnerable cryptography be finished by 2035. By 2030, Australia’s Signals Directorate hopes to have the same completed. These timelines are not set in stone. They are tangible, and they are already tight for big companies with intricate legacy systems.
Observing this from the outside, it’s odd that the urgency usually appears in technical documents before it becomes apparent to the general public. Elliptic-curve cryptography is unfamiliar to the majority of people.
The majority of people are unaware that their phone’s encrypted messages and cryptocurrency wallet security probably rely on the same mathematical framework that Google’s researchers recently showed may be far more brittle than previously thought. In the past, there has been a costly lag between what researchers know and what institutions prepare for.
The optimists point out that defenses already exist, and there are good reasons to be optimistic. Several post-quantum cryptographic algorithms that are thought to be impervious to quantum attacks have been standardized by NIST. In certain protocols, Google Chrome and Cloudflare have already implemented hybrid protections. The instruments are accessible.
Whether businesses move fast enough to employ them is the question. Because cryptocurrencies are decentralized and require coordination to upgrade their underlying cryptographic schemes, they pose a unique challenge. This was specifically noted in Google’s March paper, which urged blockchain systems to start migrating right away.
It’s possible that Q Day won’t happen in the disastrous, overnight manner that some people worry about. It’s also possible that even the specialists directly involved in these systems are still taken aback by the rate of advancement in both algorithm efficiency and hardware. One of the most eminent figures in contemporary computing, Geoffrey Hinton, recently stated that artificial intelligence was developing “very, very fast, much faster than I expected.”
Quantum computing appears to be in a similar situation. It appears that those who are creating it, researching it, and attempting to defend against it are all adjusting and, for the most part, advancing their timelines.
The vault is still intact. However, the developers of the tools needed to open it are improving their skills more quickly than almost anyone could have imagined. Even though the math behind it is challenging and the headlines haven’t quite caught up, that is something worth paying attention to.
