Why Seed Stage Is the Right Time to Back Deep Tech

The conventional wisdom in venture capital says deep tech is too risky for seed investors. The technology timelines are too long, the capital requirements too high, the path to commercial traction too uncertain. Wait until the science is de-risked, the thinking goes, and invest when there is a clear product and a paying customer. We believe this conventional wisdom is wrong — and increasingly expensively wrong. Investing early in companies with genuine scientific moats is not just viable; it is the most defensible strategy in venture capital.

The Standard Objection and Why It Misses the Point

The standard objection to seed-stage deep tech investing runs something like this: a university spinout with a novel material synthesis process or a photonic computing architecture has not yet proven that its technology works at scale, has not built a product, has not signed a customer, and may be three to five years away from meaningful revenue. The risks at this stage are overwhelming — technology risk, market risk, execution risk, team risk — and a seed investor who commits capital before these risks are resolved is speculating rather than investing.

This objection has internal coherence, but it is built on a flawed model of how deep tech companies create value and how the venture capital market prices them. The insight that seed-stage deep tech investors operate on is not that these companies are risk-free — they are not. It is that the deepest scientific and engineering risks in a deep tech company are often most efficiently resolved with relatively modest amounts of capital, and that the market tends to dramatically undervalue deep tech companies at the seed stage because most investors lack the technical expertise to assess them correctly.

Consider the economics. A photonic computing startup at the seed stage might need €2–3 million to demonstrate that its waveguide fabrication process works at the tolerances required for commercial chips. This demonstration does not require a product, a sales team, or a marketing budget — it requires access to a semiconductor fabrication facility, a handful of world-class photonics engineers, and time. If the demonstration succeeds, the company's valuation will step up dramatically — not because it has revenue, but because the most fundamental technical risk has been resolved. The seed investor who backed the company before this inflection point captures the greatest return on the most important value-creating event in the company's early life.

Scientific Moats Are the Most Durable Competitive Advantages in Technology

The canonical venture capital playbook was developed largely in the context of software companies, where competitive advantages are built on network effects, switching costs, and proprietary data — all of which take time to accumulate but can theoretically be replicated by a well-capitalised competitor starting from scratch. A seed-stage software company may have a clever product and a head start, but it rarely has a moat that is genuinely impenetrable.

Deep tech companies with genuine scientific differentiation operate differently. A novel material with a specific set of properties — strength, conductivity, biodegradability — that was developed through years of original research cannot be replicated by an engineer hired last month. A quantum error correction algorithm developed by a team that spent a decade understanding the specific noise characteristics of superconducting qubits represents accumulated knowledge that cannot be bought or shortcut. A protein structure prediction approach that achieves superior accuracy on specific drug-discovery tasks may be protected by a combination of patents, trade secrets, and tacit knowledge that competitors would need years to independently replicate.

These scientific moats are qualitatively different from software competitive advantages. They are deeper, more durable, and more amenable to legal protection through patents and trade secrets. A deep tech company that has genuinely solved a hard scientific problem is not just ahead — it is operating on a different technological platform than its competitors, one that will take years to catch up to even if competitors recognise the threat and begin investing immediately.

This durability of scientific moats is precisely why seed stage is the right time to invest. The scientific advantage is established when the research is done, not when the product is built. Waiting until the product is launched to invest in a deep tech company means paying a much higher price for the same underlying scientific asset, after the key risk — the scientific one — has already been resolved by someone else's capital.

The Valuation Anomaly in Seed-Stage Deep Tech

The venture capital market is highly competitive at the growth stage, where company metrics are legible to a wide range of investors and where the combination of revenue multiples and comparable transactions constrains valuation within a relatively narrow band. At the seed stage, and particularly for deep tech companies, valuations are far less efficiently priced.

The reason is information asymmetry. A seed-stage deep tech company's value is largely a function of the quality and defensibility of its underlying science — a dimension that most investors lack the expertise to evaluate. A generalist venture investor who can assess a SaaS company's cohort retention, expansion revenue, and GTM strategy with confidence may have no ability to evaluate whether a novel battery electrolyte chemistry is genuinely superior to existing alternatives, or whether a quantum error correction approach is technically sound. The inability to assess the underlying science means that generalist investors systematically undervalue deep tech companies at the seed stage, either by passing on them entirely or by applying valuation frameworks that do not capture the true value of the scientific asset.

For investors who do have the technical expertise to evaluate deep tech — who can read the academic papers, understand the patent landscape, and assess the technical credibility of the founding team — this information asymmetry is a persistent source of alpha. The market is inefficient precisely because most investors cannot access the information that would allow them to price deep tech correctly. Technical expertise is the edge, and it is not an edge that disappears as markets mature, because the science keeps advancing and the technical frontier keeps moving.

Managing Risk the Right Way: Staged Investment and Technical Milestones

Acknowledging that seed-stage deep tech investing generates superior returns does not mean the risks are irrelevant. They are real, and managing them effectively requires a disciplined approach to staging investment against technical milestones.

The most important risk management tool in deep tech seed investing is the ability to distinguish between types of risk. Technology risk — the risk that the underlying science does not work as expected — is the most fundamental and needs to be assessed upfront with the greatest care. Market risk — the risk that even a working technology cannot find customers willing to pay for it — is often underestimated in deep tech and needs to be probed systematically before significant capital is deployed. Execution risk — the risk that the team cannot build a company around the technology — is common to all seed investing but takes specific forms in deep tech, where technical founders often lack commercial experience.

Effective deep tech seed investors structure their investments to minimise the capital deployed against unresolved technology risk while maximising the capital available for proven technologies. This often means leading a first tranche of financing specifically designed to resolve a defined technical milestone, with commitments to participate in follow-on financing once that milestone is achieved. The milestone structure keeps the founding team focused, minimises capital wasted on scientific dead ends, and ensures that the investor's portfolio includes only companies where the fundamental science has been validated.

The Strategic Advantages of Being the First Investor

Beyond the financial returns, there are strategic advantages to being the first institutional investor in a deep tech company that amplify over the long term. First movers in a company's cap table typically secure the best terms — the lowest price, the most favourable governance rights, and the strongest information rights. These advantages compound over the life of the investment as the company's valuation grows and as subsequent investors accept less favourable terms to access the opportunity.

First institutional investors also have the longest and deepest relationship with the founding team, which translates into disproportionate influence over strategic decisions. In deep tech companies, where the most consequential strategic decisions often involve choices between different technological trajectories — which application to prioritise, which manufacturing partner to choose, which licensing structure to pursue — having an investor who understands the technology deeply and has been with the company since the beginning is uniquely valuable.

Finally, being the first institutional investor in a successful deep tech company builds a reputation that generates deal flow for future funds. The deep tech founder community is small and well-connected. A reputation for providing genuine scientific insight, strategic guidance, and patient capital — qualities that are rare among venture investors — creates a self-reinforcing advantage in access to the best deals that becomes more valuable with each successive investment cycle.

How We Put This Into Practice at Hilberts AI Capital

At Hilberts AI Capital, our approach to seed-stage deep tech investing is built around three principles that reflect our conviction that early-stage backing of genuine scientific innovation is both the highest-return and the highest-impact strategy available to venture capital investors.

First, we invest in the science before the product. We look for companies where the core scientific innovation is defensible, differentiated, and early enough that the market has not yet recognised its full value. We use our technical advisory network — researchers, engineers, and domain experts across our focus areas — to assess the quality of the science independently, rather than relying solely on the founding team's own assessment.

Second, we structure our initial investments around technical milestones, with reserve capital committed to follow-on as those milestones are achieved. This approach minimises the risk of deploying follow-on capital against technologies that have not proven out, while ensuring that our best-performing companies have access to the resources they need to reach commercial scale.

Third, we build relationships with founding teams long before they are fundraising. The best deep tech companies are often founded by researchers who are still in academia, who are years away from commercial readiness, and who are not yet aware that they will become founders. By engaging with the deep tech research community — at conferences, through university partnerships, and through our network of scientific advisors — we build relationships with future founders before they enter the market, giving us access to opportunities that never appear in the competitive fundraising process.

If you are a deep tech founder — or a researcher who is considering becoming one — we would love to hear about what you are building. Early is exactly when we want to talk. Contact us to start a conversation.