Three people who spent their careers inside the patent system have decided to take it apart. Dominic Davies spent twenty years as a patent attorney. Ola Wassvik, as chief technology officer of the touchscreen company FlatFrog, managed more than 300 patent families on roughly $100 million of venture funding. Markus Andreasson wrote software for 25 years at Sony and the retail-tech company Voyado and holds more than 50 patents of his own. In 2023 the three founded Lightbringer in Malmo to do to patent firms what software has already done to most professional services, and this week the company raised €8.6 million to carry that idea into the United States.

The Series A, $10 million in dollar terms, was co-led by London’s 6 Degrees Capital and Amsterdam’s Newion, with Thomas Olszewski of 6 Degrees and Dorus Olgers of Newion taking board seats. It builds on a €4.2 million seed led by Luminar Ventures and Alliance VC in late 2024. The capital is pointed almost entirely at one move: product development and a US launch, with Davies himself relocating to run American operations.

What the founders are actually selling

For the past few years, legal AI has mostly meant tools that make lawyers faster. Software that drafts the first version of a contract, surfaces the relevant case, trims the billable hours at the edges. Lightbringer’s pitch starts by rejecting that entire framing. The interesting product, the founders argue, is not the one that helps a patent firm protect its margins but the one that removes the firm from the transaction. They call the model service as software, a hybrid in which the platform does the work that used to require a building full of attorneys, and the few attorneys that remain sit inside Lightbringer rather than billing from outside it.

In practice the platform owns the full lifecycle of a patent: invention capture, drafting, filing, portfolio strategy, portfolio management, competitor intelligence, and white-space analysis, run through a single interface. The company says it compresses a filing that normally takes about two months into a few days and cuts the cost roughly in half through a fixed-price subscription rather than an hourly bill. Since launching in 2024 it reports serving more than 200 deep-tech companies across 17 countries, including the defence-drone firm Arctic Ravn, the quantum-sensing startup DIASENSE, the critical-infrastructure hardware specialist TERASi, and the clean-air company Cler, and it claims 300% year-on-year revenue growth in the second quarter of 2026.

“Most legal AI helps law firms become more efficient and protect their margins,” Davies said. “We built Lightbringer to do the opposite: take on Big Law and return that value to entrepreneurs.” It is a clean line, and it is also where the difficulty starts.

Replacing the firm is a much larger promise than helping it

The reason most legal AI companies position themselves as assistants is that assisting is safer. A patent is an adversarial document. It is examined, contested, and sometimes litigated for years, and a single claim drafted too narrowly can hand a competitor the right to build the exact thing the inventor thought they had protected. A tool that helps a lawyer carries the lawyer’s judgment and the lawyer’s liability. A platform that sets out to replace the firm is volunteering to absorb both, which is why Lightbringer keeps human patent attorneys in the loop rather than promising pure automation. That hedge is sensible, and it also quietly undercuts the cleanest version of the software-margin story, because attorneys on staff are the expensive part the model was supposed to engineer away.

The competitive backdrop is loud. Harvey, the San Francisco legal-AI company, raised $200 million in March at an $11 billion valuation; Luminance, founded in Cambridge, raised a $75 million Series C for contract analysis. Both are far larger and both, for now, mostly make lawyers faster rather than redundant. Lightbringer’s wager is that the narrower, harder lane, owning the whole patent process end to end, is the more defensible one precisely because nobody with that much capital is trying to do it. The market it is reaching for is real: the global patent law services segment was valued at around $20 billion in 2024 and is projected to roughly double over the next decade as research spending and filings climb.

There is also the matter of where the company is going. The United States is the largest market for intellectual property and also the most litigious, the one environment where a badly drafted claim is most likely to be tested in court. Choosing it as the first expansion target is either confidence or exposure, and the founders are betting it is the former. Their argument, in the end, is an insider’s argument: they helped run the system, they have seen exactly where its seams are, and they think the value trapped inside law-firm billing belongs to the inventors instead.

Whether that conviction survives contact with an American courtroom is the open part. What the founders are selling is not a faster version of the old process but a claim that the process itself was never built for the people paying for it. The €8.6 million buys them the chance to prove it in the one market where being wrong is most expensive.

Filed by eu-tailwind. Independent reporting on the rise of European tech.