In 2025, 36% of all European venture capital flowed into deep tech, nearly double the 19% share recorded just four years earlier. That shift is the clearest signal yet that investors, governments, and entire industrial systems are betting on science-driven ventures to define the next economic era. But what is deep tech, exactly, and why does it command so much attention right now?

This guide breaks it down: the definition, the terminology debates, named companies across six verticals, the funding mechanics that set deep tech apart, and the European policy architecture accelerating everything. Whether you're a founder, investor, or corporate strategist, this is the reference point.

What Is Deep Tech? The Definition Explained

Deep tech refers to companies and ventures built on substantial scientific or engineering breakthroughs. These are not businesses that apply existing technology in a new market. They advance the technology itself, pushing the boundary of what is physically, chemically, or computationally possible.

The term was popularised from around 2017 by BCG and the Hello Tomorrow foundation, who needed language to distinguish science-rooted startups from the app-layer software companies dominating venture portfolios at the time. Before "deep tech" entered the lexicon, these ventures were scattered across labels: "science startups," "hard tech," "frontier R&D." The new term stuck because it described something real: a growing class of ventures that did not fit the "move fast and break things" playbook but were nonetheless building enormous commercial value.

What makes deep tech deep is not complexity for its own sake. It is the distance between the scientific insight and a working product. A deep tech founder is not just solving a market problem; they are often solving a physics problem, a chemistry problem, or a computational theory problem first, and then translating that solution into something a customer can buy.

Deep technology ventures share three defining characteristics:

1. Science-based foundations. The core innovation emerges from advances in physics, biology, chemistry, mathematics, or engineering, not from a novel business model or UX layer. Think new semiconductor architectures, not a better food delivery app.

2. Long development cycles. Timelines of 5 to 15 years from lab to market are common. A fusion energy startup does not ship a minimum viable product in six months. Neither does a quantum computing company.

3. Hard-to-replicate intellectual property. The barriers to entry are structural. Reproducing a deep tech breakthrough requires equivalent scientific talent, specialised equipment, and often years of iterative research. This creates durable competitive moats that software startups rarely enjoy.

These three traits form the filter. If a company checks all three, it is deep tech. If it checks one or two, it might be an advanced engineering firm or an applied-AI startup, but it is not deep tech in the strict sense.

Deep Tech vs. Hard Tech vs. Frontier Tech

The terms overlap, and people use them loosely. Here is how they differ:

The key distinction: hard tech is always physical. Deep tech is broader; it includes hard tech but also encompasses software-layer breakthroughs like fundamentally new AI models or cryptographic systems, provided they are rooted in genuine scientific advance. Frontier tech is a timing label. Something is "frontier" until it is proven at scale, at which point it becomes deep tech or hard tech.

One more distinction worth making: not all AI is deep tech. A company fine-tuning a large language model for customer support is applying existing technology. A company developing entirely new neural network architectures or building AI systems that reason about physical processes: that is deep tech.

The line is not about which buzzword you use. It is about whether the core challenge is scientific or commercial.

The Six Pillars of Deep Tech (With Real Examples)

Deep tech is not a single sector. It is a cross-cutting category that spans multiple industries. Below are six verticals where European deep tech ventures are building, each illustrated with named companies and real funding figures.

Defence and Security

Europe's defence posture shifted dramatically after 2022, and deep tech sits at the centre of the recalibration. The European Defence Fund commits EUR 1 billion in 2026 to collaborative R&D, while the EUDIS accelerator offers EUR 120,000 seed vouchers to defence and security startups working on dual-use technologies.

The standout name is Helsing (Munich), which has raised over EUR 450 million to build AI-driven defence systems. Quantum Systems (Munich) builds reconnaissance drones already deployed in contested environments. ARX Robotics (Linz) develops modular unmanned ground vehicles for logistics and reconnaissance.

Space

European space tech has moved well beyond the legacy of Ariane and ESA contracts. A new generation of private ventures is compressing timelines and cutting costs.

Isar Aerospace (Munich) has raised EUR 165 million to build a micro-launcher designed for commercial small-satellite deployment. The Exploration Company (Munich) is developing reusable orbital vehicles, tackling the in-orbit logistics problem. OHB (Bremen), one of Europe's established space system integrators, continues to anchor the continent's satellite manufacturing capacity.

Robotics and Manufacturing

Factory floors are being redesigned. Not with incremental automation, but with robots that learn, adapt, and work alongside humans.

Agile Robots (Munich) builds AI-powered robotic arms capable of delicate manipulation tasks: surgery, electronics assembly, precision agriculture. Wandelbots (Dresden) has developed a no-code platform that lets non-engineers programme industrial robots, attacking the integration bottleneck that slows adoption. Flexion (Zurich) raised EUR 43 million for its robotic joint technology, which enables more dexterous and energy-efficient robotic and manufacturing systems.

Energy

The energy transition is not just about deploying more solar panels. It requires fundamental breakthroughs in how we generate, store, and distribute power.

Proxima Fusion (Munich) is pursuing stellarator-based fusion energy, a design approach that avoids the plasma instabilities plaguing tokamak reactors. Marvel Fusion (Munich) has raised a combined EUR 385 million for its laser-driven inertial confinement approach. Enpal (Berlin) has scaled residential solar installation across Germany, pairing hardware deployment with software-driven energy management.

Advanced Materials

Materials science is the quietest pillar of deep tech, and arguably the most consequential. Every other sector depends on it. Better batteries need better cathode chemistry. Better chips need better substrates.

Better rockets need better alloys. That dependency makes materials the silent constraint on progress across all verticals.

QuantumDiamonds (Munich) is building a EUR 152 million facility for producing synthetic diamonds engineered for quantum sensing and industrial applications. Metafuels (Switzerland) develops synthetic aviation fuels using captured CO2, attacking one of the hardest-to-decarbonise sectors in transport. Both companies operate in the advanced materials space where timelines are long, capital requirements are high, and success reshapes entire supply chains.

Future of Compute

Classical computing is approaching physical limits. Deep tech ventures in this pillar are building what comes next.

IQM (Finland) became Europe's first quantum computing unicorn after raising over EUR 300 million, and now operates superconducting quantum processors used by national research labs across the continent. PASQAL (France) is heading towards a USD 2 billion SPAC listing, bringing neutral-atom quantum computing to commercial markets. The European Chips Act, backed by EUR 4.175 billion from the EU with a target of EUR 43 billion in total private and public investment, provides the policy scaffolding for this entire compute stack.

Deep Tech vs. Software Startups: Why the Funding Model Is Different

A SaaS startup can reach product-market fit with a small team and modest capital. Ship code, measure retention, iterate weekly. A seed round of EUR 2 to 3 million can sustain a software company through to Series A and often to profitability.

Deep tech does not work that way. A seed round for a fusion startup might cover six months of lab access. The differences are structural, and they explain why deep tech requires a fundamentally different investment thesis:

• Capital intensity. Building a fusion reactor prototype costs orders of magnitude more than building a web application. Hardware iterations are expensive. Lab time is expensive. Regulatory approval is expensive.

• The valley of death. Deep tech ventures face a specific funding gap between early-stage grants (which prove the science) and late-stage capital (which scales the product). This middle zone, roughly Technology Readiness Levels 4 through 7, is where promising companies die. The science works in the lab but cannot yet prove commercial viability at scale.

• Timeline mismatch. Traditional VC funds operate on 7 to 10 year cycles. Deep tech ventures often need 10 to 15 years to reach maturity. That mismatch has historically starved deep tech of private capital, pushing founders towards grants and government programmes.

Three mechanisms are closing the gap. Corporate venture capital (CVC) from defence primes, energy majors, and semiconductor firms brings both capital and market access. Government grants, particularly through Horizon Europe's approximately EUR 95.5 billion budget (2021 to 2027) and national programmes like France 2030 and Germany's DeepTech and Climate Fund, de-risk early stages. A new class of specialist deep tech VCs has emerged in Europe, with fund structures designed for longer hold periods and higher capital calls. In March 2026, Cloudberry Ventures launched a EUR 50 million deep tech fund led by a former Google Moonshot X advisor, signalling continued investor appetite.

The result: deep tech is still harder to fund than software. But the infrastructure for doing so is more mature than it has ever been. Capital is learning to read scientific milestones, a successful prototype test, a patent grant, a regulatory pre-approval, as the deep tech equivalent of product-market fit signals.

Why Deep Tech Matters Now: The 2026 Landscape

Three forces are converging in 2026 to make deep tech not just interesting but urgent.

First, the capital is real and growing. European startups raised USD 58 billion in venture funding in 2025, a 9% year-on-year increase. AI was the leading sector for the first time at USD 17.5 billion, while hardware and deep tech claimed USD 10.8 billion. European tech companies raised EUR 72 billion overall, making 2025 the second-strongest year on record (Tech.eu). Meanwhile, the European Innovation Council allocated EUR 1.424 billion for 2026 specifically for deep tech innovators, introducing a simplified EIC Accelerator and ARPA-inspired Advanced Innovation Challenges.

Second, the policy architecture is in place. The European Chips Act, the European Defence Fund, Horizon Europe, the IPCEI (Important Projects of Common European Interest) framework, and national strategies like France 2030 and Germany's DeepTech and Climate Fund form an interlocking system of grants, procurement guarantees, and regulatory support. No other region has this density of coordinated public investment in science-based industry.

Third, the talent pipeline is producing. 76 European spinouts from universities and research institutions have reached unicorn (over USD 1 billion valuation) or centaur (over USD 100 million revenue) status as of 2025, according to the Dealroom European Spinout Report. Institutions like TU Munich, CEA, CNRS, Aalto University, and VTT are no longer just publishing papers; they are generating companies.

The stakes are high. McKinsey estimates that Europe's deep tech engine could generate USD 1 trillion in enterprise value and up to 1 million jobs by 2030. Sweden already illustrates the shift: 65% of its startup funding now goes to deep tech, the highest proportion in Europe.

There is a geopolitical dimension too. Supply chain vulnerabilities exposed during the pandemic, semiconductor shortages, and energy dependency on hostile states have turned deep tech from an investment thesis into a security priority. When a European government funds a domestic chip fabrication plant or a defence AI startup, it is not just backing a company; it is building strategic autonomy.

Europe's Deep Tech Landscape: Key Hubs and Hotspots

Deep tech does not emerge from nowhere. It clusters around research institutions, talent pools, and capital networks. In Europe, several hubs have reached critical mass.

Berlin anchors Germany's deep tech community with TU Berlin, Fraunhofer institutes, and Helmholtz research centres feeding a steady stream of spinouts into the city's startup infrastructure. The city's strength sits in energy tech, AI, and cross-disciplinary research commercialisation.

Munich is arguably Europe's deepest deep tech city. TU Munich and the Max Planck Society provide the research base. Helsing, Isar Aerospace, Proxima Fusion, Marvel Fusion, Agile Robots, Quantum Systems, and QuantumDiamonds all call the region home. The cluster effect is self-reinforcing: talent stays because the companies are there, and companies stay because the talent is there.

Paris brings CEA and CNRS, two of the world's largest public research organisations, plus a growing VC scene anchored by firms like Eurazeo and Partech. PASQAL's quantum ambitions are rooted here.

The Nordics punch above their weight. Helsinki, powered by the Aalto University and VTT Technical Research Centre nexus, has produced outsized deep tech output relative to population. IQM's quantum trajectory started in Finland. Sweden's 65% deep tech funding share speaks for itself.

Eindhoven rounds out the map as the heart of Europe's semiconductor cluster, anchored by ASML's lithography monopoly and a dense network of photonics and chip design firms. ASML's presence created a gravitational pull for chip design talent, which attracted suppliers, which attracted more startups.

What distinguishes Europe's landscape from the US or China is its polycentric structure. There is no single "Silicon Valley" for European deep tech. Specialised hubs complement each other: Munich for aerospace and defence, Helsinki for quantum, Paris for nuclear and AI research, Eindhoven for semiconductors.

This distribution can look like fragmentation from the outside, but it is actually resilience. It creates opportunities for cross-pollination when these communities come together, which is precisely the function of events like DTM26, Deep Tech Momentum (Berlin, 24 to 26 June 2026).

How to Get Involved in Deep Tech

Deep tech is not a spectator sport. The window for establishing positions, as a founder, investor, or strategic partner, is open now while the market is still forming.

Founders building a deep tech venture should apply to the DTM100, Europe's top deep tech startups. The cohort gets direct access to the DTM26 stage, investor meetings, and the broader Deep Tech Momentum network.

Investors seeking structured deal flow across defence, energy, compute, space, robotics, and materials should explore the Guardian Program. Guardians get curated introductions, thematic briefings, and preferred access to the DTM100 pipeline.

Corporates face a simple reality: deep tech will reshape your supply chains, your competitive landscape, and your talent strategy, whether you engage proactively or not. Get your DTM26 ticket and spend two days with the people building Europe's next industrial era.

Capital is flowing. Science is maturing. Policy support is stronger than it has ever been.

So the question is not whether deep tech matters. It is whether you are positioned to benefit from it.

Frequently Asked Questions

What is the difference between deep tech and high tech?

Deep tech involves fundamental scientific or engineering breakthroughs that create entirely new capabilities. High tech refers to the most advanced commercially available technology at a given time. A smartphone is high tech; a quantum processor is deep tech. The key difference is that deep tech pushes the boundary of what is possible, while high tech represents the current state of the art.

Is AI considered deep tech?

It depends on the layer. A company developing novel neural network architectures, new reasoning systems, or AI that models physical processes is deep tech. A company fine-tuning an existing large language model for a specific business application is not. The dividing line is whether the core challenge is scientific (advancing the technology itself) or commercial (applying existing technology to a market).

Why is deep tech important in 2026?

Three converging forces: capital (36% of European VC now flows into deep tech), policy (EUR 1.424 billion from the EIC alone in 2026, plus the Chips Act and Defence Fund), and talent (76 European spinouts have reached unicorn or centaur status). McKinsey projects USD 1 trillion in enterprise value and up to 1 million jobs by 2030.

What are the main deep tech sectors?

The six primary verticals are defence and security, space, robotics and manufacturing, energy (including fusion), advanced materials, and future of compute (quantum, photonics, neuromorphic). Deep tech also spans biotechnology, semiconductors, and climate technology. These verticals are interconnected: materials science underpins energy, space technology serves defence, and AI accelerates discovery across all of them.

How long does it take to build a deep tech company?

Typical timelines run from 5 to 15 years from founding to commercial scale, compared to 2 to 5 years for a software startup. The lengthy development cycle reflects the need to solve scientific problems before commercial ones, navigate regulatory approvals, and build physical infrastructure. This is why deep tech requires patient capital and specialised investors who understand milestone-based progress rather than quarterly revenue metrics.