Quantum computing is suddenly attracting serious capital again. More than $4 billion of venture capital has flowed into quantum companies this year, almost as much as during all of last year. The change in sentiment is significant: quantum is moving from a technology that was perpetually “five years away” toward something investors increasingly believe could become commercially relevant.

There are good reasons for that optimism. Quantum computers operate according to the same physical rules that govern atoms and molecules, which makes them particularly interesting for problems that are extremely difficult to simulate on classical computers. Drug discovery, chemistry, batteries and new materials are among the most promising areas. Recent experiments also suggest that quantum machines can already outperform classical systems for certain highly specific tasks. At the same time, companies such as IBM and Google continue to invest heavily in the field, although even they are exploring different technical architectures. That alone shows how early the market still is: investors are not only betting on the right company, but to some extent also on the right physics.

This makes quantum very different from a traditional software investment. In software, investors can often observe relatively early whether customers use a product, whether they are willing to pay for it and whether retention improves over time. In quantum, many of the largest risks come before those metrics become meaningful. Investors have to understand whether the underlying technology works, whether it can scale to thousands or potentially tens of thousands of qubits, whether error rates can be controlled and how expensive the machines will ultimately be to build and operate. A compelling scientific demonstration is therefore not enough. What matters is whether every technological milestone brings the company closer to a commercially viable system.

That also changes how I would think about financing these businesses. Quantum companies can require enormous amounts of capital long before they generate meaningful revenue. A $100 million financing round should therefore not simply fund another few years of research. Ideally, it removes a specific uncertainty: error correction, scalability, manufacturing, system cost or the first economically relevant application. In capital-intensive deep tech, the critical question is not just how much money a company raises, but what risk that money eliminates. A company can become a technological success and still turn out to be a poor venture investment if reaching that success requires too much capital.

Even if the technology works, however, an even more interesting question remains: who captures the value? Imagine that a quantum computer helps a pharmaceutical company discover a drug worth several billion dollars. The quantum system may have created enormous economic value, but that does not automatically mean the quantum company captures much of it. If it simply sells computing time, most of the upside could remain with the pharmaceutical company. This is the distinction between value creation and value capture, and it may become one of the defining business-model questions in quantum computing.

That is why investors are already discussing models in which quantum companies participate more directly in the economic success they enable, for example through licensing, partnerships or profit-sharing. Whether customers will accept such structures is still unclear, but the logic is compelling. If your technology is essential to discovering a billion-dollar molecule, selling a relatively small amount of compute may not be the optimal way to monetize it. The strongest companies may therefore need to move beyond being infrastructure providers and find ways to own part of the application or outcome.

Technology history suggests that the biggest winners may not necessarily be the companies building the hardware. In previous computing waves, large amounts of value were often created higher up the stack. PCs created Microsoft, the internet created platforms such as Google and Amazon, and cloud infrastructure enabled an entire generation of software businesses. Quantum could follow a similar pattern. A small number of hardware platforms may emerge, while some of the most valuable companies build highly specialized applications for pharmaceuticals, chemistry, materials or other industries.

Ultimately, customers do not want quantum computers. A pharmaceutical company wants a better drug, a battery manufacturer wants better materials and a chemical company wants a more efficient process. The winning companies may therefore not be those with the largest number of qubits, but those that are best at translating qubits into economic value.

For investors, this means that a quantum investment today is really three bets at once: which technology will work, whether it can be built and operated at attractive economics, and whether the company can capture a meaningful share of the value it creates. The billions flowing into the sector suggest that investors are becoming increasingly confident about the first question. The second and third may ultimately matter much more for returns.

That is why the most important investment question in quantum may no longer be “Will quantum computing work?” It may be “If it works, who makes the money?”

• Financial Times: FT article
• TechCrunch: TechCrunch article