Quantum is still a press release

Every few weeks, a new quantum computing headline lands with the force of revelation. A chip that solves a problem in minutes that would take a classical supercomputer ten septillion years. A startup valued at a billion dollars before it has shipped a single commercial product. A government pouring billions into national quantum strategies. The press releases are relentless, breathless, and almost entirely disconnected from what quantum computers can actually do today. The gap between quantum announcements and quantum utility is arguably the widest in all of technology. And in April 2026, it's worth asking a blunt question: what has quantum computing actually shipped?

The announcement treadmill

The pattern is unmistakable. Google unveils Willow, a 105-qubit superconducting chip that achieves a new benchmark in quantum error correction. IBM publishes an updated roadmap promising fault-tolerant quantum computing by 2029. Microsoft announces a breakthrough in topological qubits. Startups raise hundreds of millions on the promise of quantum advantage. Each announcement is real science. None of it is fake. But there is a crucial distinction between a research milestone and a production-ready technology, and the press releases almost never make that distinction clear. Google's Willow chip, for example, demonstrated that increasing the number of qubits in an error-correcting code actually reduces the error rate, a result that had eluded researchers for nearly 30 years. That is genuinely important. But Willow has 105 physical qubits. Running a single useful fault-tolerant computation, like simulating a complex molecule for drug discovery, would likely require millions of physical qubits. The chip is a proof of concept for error correction, not a computer that can run your code. IBM's roadmap tells a similar story. The company aims to demonstrate "the first examples of scientific quantum advantage" by the end of 2026, using quantum processors working alongside classical high-performance computing. The first large-scale fault-tolerant quantum computer, IBM Quantum Starling, is targeted for 2029, with 200 logical qubits and 100 million quantum gates. That is an engineering plan, not a product you can use today.

Fifteen years of "almost here"

If you have been following technology for any length of time, you have heard this refrain before: "practical quantum computing is just five years away." The problem is that people have been saying this for roughly 15 years, and the goalpost keeps moving. Contrast this with AI. Large language models went from research curiosity to production deployment in roughly three years. GPT-3 arrived in 2020 as a research demo. By 2023, AI was embedded in products used by hundreds of millions of people. The trajectory from lab to production was steep, fast, and undeniable. Quantum computing has followed a very different arc. D-Wave shipped its first commercial system in 2011. Fifteen years later, there is still no quantum computer running production workloads that outperform classical alternatives for any real-world business problem. The most credible near-term applications, quantum simulation for chemistry and materials science, remain in the research phase. Optimization and machine learning, two areas frequently cited as quantum use cases, have not yet demonstrated consistent, scalable advantages over advanced classical methods. This does not mean quantum computing is a scam. It means the engineering challenges are genuinely harder than the hype suggests.

Why the money keeps flowing

If quantum computers are not doing useful work yet, why did the sector attract over $4.6 billion in funding between Q1 2025 and Q1 2026? Why is Nvidia investing in quantum startups, and why are governments worldwide committing tens of billions to national quantum strategies? A few forces are at play. Narrative economics. Quantum computing has an irresistible story: a technology that could break encryption, simulate nature at the atomic level, and solve optimization problems that are intractable for classical machines. Investors fund narratives, and this one is compelling. FOMO. No major tech company or government wants to be caught flat-footed if quantum computing does reach a tipping point. The cost of being wrong by investing too early is measured in billions. The cost of being wrong by investing too late could be measured in national security. National security framing. Quantum computing has become intertwined with geopolitical competition. China's patent filings in quantum technology grew 31% year over year, and total public funding globally reached $56.7 billion by 2025. When a technology is framed as a matter of sovereignty, funding flows regardless of commercial viability. The long-game bet. Some investors genuinely believe that quantum computing will eventually deliver transformative value, and that the winners will be decided by who builds the best hardware and software stack now. This is not irrational. It is just a very long bet.

The security angle is real, but distant

The most frequently cited near-term threat from quantum computing is to encryption. A sufficiently powerful quantum computer running Shor's algorithm could, in theory, break the RSA and elliptic curve cryptography that protects most internet traffic today. Recent research has added urgency to this concern. In April 2026, Nature published findings from two groups suggesting that quantum computers could crack common security keys before the decade is over, requiring fewer qubits and less time than previously estimated. Headlines called it a "real shock." But context matters. "Before the decade is over" still means years away, and the estimates depend on aggressive assumptions about hardware performance. No quantum computer today is anywhere close to breaking production encryption. More importantly, the defense is already shipping. NIST finalized its first three post-quantum cryptography standards in August 2024: ML-KEM (FIPS 203) for key encapsulation, ML-DSA (FIPS 204) for digital signatures, and SLH-DSA (FIPS 205) as a hash-based backup. These algorithms are designed to resist attacks from both classical and quantum computers. NIST's transition roadmap targets full migration of federal systems to post-quantum cryptography by 2035. Major organizations are already moving. Meta published its post-quantum cryptography migration framework in April 2026, sharing lessons from deploying PQC across its infrastructure. Google, Apple, and Signal have integrated post-quantum key exchange into their products. The "harvest now, decrypt later" threat, where adversaries collect encrypted data today to decrypt with future quantum computers, is being taken seriously, and the countermeasures are real. This is the part of the quantum story that actually matters for builders right now. Not the quantum computer itself, but the post-quantum security migration that is happening in response to its theoretical future capabilities.

What builders should actually care about

If you are a developer, an engineer, or a technical leader in 2026, quantum computing should be on your radar but not on your roadmap. Here is what is actually worth paying attention to: Post-quantum cryptography migration. If your organization handles sensitive data with long-term confidentiality requirements, the time to start planning your PQC migration is now, not when a quantum computer can actually break your encryption. The standards exist. The tooling is maturing. The migration itself is complex and will take years, similar to the SHA-1 to SHA-2 transition that took over a decade across industries. Hybrid quantum-classical experiments. If you work in pharmaceuticals, materials science, or financial modeling, it may be worth exploring hybrid quantum-classical workflows through cloud platforms offered by IBM, Google, or Amazon Braket. But approach these as research experiments, not production solutions. Ignore the noise. For 99.9% of software development, quantum computing is irrelevant today. No quantum algorithm will speed up your web application, your data pipeline, or your machine learning model. The problems where quantum computers might eventually help, simulating quantum systems, certain optimization problems, some cryptographic tasks, are a narrow slice of computing.

The honest take

Quantum computing is real science solving real problems in physics and engineering. The error correction breakthroughs are genuine milestones. The people working on this technology are among the most talented researchers in the world. But the timeline from milestone to utility is measured in years, probably decades. The gap between what quantum computers can do in a lab and what they can do for your business is vast. And the press releases, with their breathless language about "game-changing" chips and "revolutionary" breakthroughs, consistently obscure this gap. The most useful thing quantum computing has done for the average developer in 2026 is motivate the rollout of better encryption standards. That is not nothing. But it is a long way from the quantum revolution we have been promised. Quantum will matter eventually. The science is too compelling and the investment too massive for it not to. But if someone tells you quantum computing is about to change everything, ask them a simple question: what is it running today? The answer, for now, is still mostly press releases.

References