Something amazing keeps happening somewhere in a lab in New York, inside a device that resembles a science fiction set chandelier more than a computer. Gold-plated wiring coils downward into nested canisters that are chilled to temperatures lower than the vacuum of space in cylindrical cryostats that hang from the ceiling. While keeping an eye on readings and numbers that the majority of people outside this building will never fully comprehend, technicians move silently throughout the room. Depending on who you ask, what’s going on inside those machines is either the most significant advancement in computing history or a costly scientific experiment that is still waiting for its right moment.
IBM would contend that it is the former. IBM Condor, the company’s newest quantum processor, uses 1,121 qubits and recently showed something that stopped many in the field in their tracks: it solved a computational problem in about five minutes that would have taken an estimated 47 years for the fastest classical supercomputer in the world. Not in a symbolic sense. Not roughly. Before lunch, 47 years of nonstop computation were completed. As that figure spreads throughout the scientific community, there’s a sense that something has truly changed, not just in a lab but also in the larger narrative of what machines are capable of.
| IBM Quantum Computing — Key Information | |
|---|---|
| Company Full Name | International Business Machines Corporation (IBM) |
| Founded | 1911, Armonk, New York, USA |
| Quantum Division | IBM Quantum |
| Key Quantum Processor | IBM Condor — 1,121 qubits |
| Notable Benchmark | Solved a problem in ~5 minutes that would take classical supercomputers an estimated 47 years |
| Operating Temperature | Near absolute zero (-273.15°C), using cryogenic cooling systems |
| Quantum Access Model | Cloud-based access via IBM Quantum Network |
| Primary Competitors | Google Quantum AI, Microsoft, IonQ, Rigetti, D-Wave |
| Global Quantum Market Value (2024) | ~$1.13 billion |
| Projected Market Value (2030–2035) | $18–20 billion |
| U.S. Policy Support | National Quantum Initiative Act — NIST Quantum Program |
| Key Applications | Drug discovery, cryptography, climate modeling, financial optimization, AI acceleration |
| UN Recognition | 2025 declared International Year of Quantum Science and Technolog |
It is helpful to consider how common computers function in order to comprehend why this is peculiar and important. In the end, every website, video, and financial transaction is just a lengthy string of ones and zeros—binary switches that flip on and off incredibly quickly. For many years, that architecture has been incredibly beneficial to humanity. However, it is limited. Certain problems become exponentially, nearly cosmically more difficult as more variables are added.
This is nicely illustrated by the traveling salesman problem, which involves determining the single most efficient route between an increasing number of cities. There are about 20 million possible routes between eleven cities. You can surpass 650 billion by adding just four more cities. Every possibility is sequentially checked by classical computers. Up until it stops working, everything is fine.
A completely different principle underlies the operation of quantum computers. They use qubits, which can be one, zero, or both simultaneously due to a feature of quantum mechanics known as superposition, in place of bits that are either one or zero. A regular bit is a coin that is flat, heads or tails, according to the flawed but helpful spinning coin analogy.
A qubit is a coin that is still spinning and simultaneously contains all of its possible outcomes. When you add entanglement, a second quantum property in which two particles are linked so that altering one instantly alters the other, regardless of physical distance, you have a machine that can potentially explore millions of solution paths simultaneously rather than sequentially. When Richard Feynman gave a talk about simulating nature at Caltech in 1981, he saw this coming and laid the groundwork for a completely new method of computing. Hardware didn’t catch up to the concept for about forty years.
Most people are unaware of how long IBM has been involved in this race. In 2019, the company introduced the IBM Q System One, the first commercial quantum computer available to businesses through the cloud. It is housed in a stylish nine-cubic-foot glass case. Even though the system’s 20 qubits were far from achieving practical dominance, it was a momentous occasion. Since then, there has been a steady, methodical push up the qubit ladder, with coherence times growing, error rates gradually decreasing, and each generation learning from the previous. There is more to the leap to Condor’s 1,121 qubits than just a numerical increase. It symbolizes years of engineering choices about how to wire a cryostat, protect delicate quantum states from outside interference, and maintain a qubit’s coherence long enough for practical use.
Google has been exerting equal pressure in the opposite direction. According to reports, its Willow processor, which was introduced in late 2025, used real-time error correction that actually gets better as more qubits are added to maintain qubit coherence for five seconds, a thousand times longer than previous systems. Prior systems deteriorated as they grew in size. Willow recovered. This distinction is crucial because the history of quantum computing is replete with spectacular demonstrations that failed due to their own noise. The difference between what quantum computers can theoretically accomplish and what they can consistently accomplish on a Tuesday afternoon has always been error correction. The industry is not taking Google’s assertion that it has solved that issue, even in part, lightly.
Whether any of these systems can currently address real-world issues on a commercial scale is still unknown. IBM is open about this, allowing researchers all over the world to explore the boundaries of what is feasible without requiring a cryostat in their workplace by making its quantum systems accessible via cloud access. The practical applications being discussed, such as modeling climate systems months in advance, optimizing traffic flow across entire cities in real time, and simulating molecular interactions for drug discovery, are compelling in a way that feels authentic rather than promotional. If quantum processors can simulate how molecules behave at the quantum level, drug discovery alone, which currently costs billions and takes ten years per compound, could be drastically reduced. It’s possible that someone is currently working on the first quantum-designed medication.
Naturally, there is a darker side to all of this. The encryption systems that safeguard government infrastructure, military communications, and bank accounts rely on challenging mathematical problems, particularly the challenge of factoring large numbers. That is a manageable problem thanks to quantum computers. All the big countries know this. For years, the U.S. National Institute of Standards and Technology has been developing post-quantum cryptography standards, essentially attempting to create locks that are resistant to a key that is not yet commercially available but most likely will be. Google and IBM are not the only players in the race. It’s between all existing security systems and whoever achieves useful quantum scale first.
Microsoft is attempting to create topological qubits, which are theoretically more stable than anything superconducting, by breaking up electrons to store data in multiple locations at once. This is a slower and more unusual approach. Startups such as Rigetti and IonQ are placing bets on completely different architectures. At a rate that makes Western investment seem insignificant in comparison, China is pouring state resources into the field. It’s anybody’s race, which can be thrilling and, depending on your viewpoint, a little unsettling. It’s possible that the quantum winter that analysts have been forecasting—the unavoidable dip between hype and useful outcomes—will still occur. It’s becoming more difficult to make that claim with a straight face, though, given what IBM’s Condor just showed in that chilly New York lab.
