Biolevate raises €30M Series A for life sciences AI

Paris-based Biolevate has raised a €30M Series A co-led by RAISE France and Orange Ventures, joined by MSD Global Health Innovation Fund and Station F, with returning backer EQT Ventures, to scale its evidence-grade AI platform for life sciences and open a Boston office.

Every drug approval still ends the same way: thousands of pages, assembled by hand, checked line by line. The reality is that life sciences runs on documents nobody enjoys writing – and that is exactly where the money is now going. Below, I lay out the round, the team behind it, and why regulated paperwork is quietly becoming one of Europe’s most investable AI markets.

Two Paris co-leads and a pharma CVC

Biolevate announced the round on 22 September 2026: a €30M Series A co-led by RAISE France and Orange Ventures. MSD Global Health Innovation Fund and Station F come in as new investors, and EQT Ventures returns – the fund that led Biolevate‘s €6M seed in November 2024. Total raised stands at €36M, less than two years after incorporation.

The board changes with the money: Aymar Hénin takes the chair, and Orange Ventures’ Jérôme Berger joins him around the table. The capital itself goes three ways – international expansion starting with a Boston office, the company’s first outside France, deeper product capabilities, and scaled deployments with life sciences organisations.

“Customers are increasingly embracing AI automation at scale, grounded in evidence and under rigorous human oversight,” said co-founder and CEO Joël Belafa in the announcement.

Dataiku alumni running 100,000 agents

Biolevate was founded in Paris in 2024 by former Dataiku engineers and AI researchers, and it shows in the product’s posture: enterprise-grade, auditable, allergic to black boxes. The platform automates the document-heavy middle of life sciences – regulatory dossiers, systematic literature reviews, health technology assessments, clinical trial workflows and drug development support. It can run up to 100,000 agents concurrently, with outputs grounded in cited evidence and kept under human review. That “evidence-grade” framing is the whole pitch in a sector where one hallucinated reference can sink a submission.

The traction figures are company-reported, and they are striking: more than a dozen enterprise customers in production, annual recurring revenue up 20x over the past twelve months, systematic literature reviews accelerated sixteenfold, and an 80% efficiency gain in regulatory intelligence work. Sanofi, NVIDIA and Microsoft feature among its partners. The team stands at roughly 50 people and is expected to double over the coming year, with seven patent applications drafted and two already filed in Europe.

A $1.31bn niche compounding at 18.6%

The segment Biolevate sells into – AI for regulatory affairs – was worth $1.31bn globally in 2024 and is forecast to reach $6.65bn by 2033, an 18.6% CAGR (Grand View Research, October 2025). Zoom out one level and the picture widens considerably: AI in life sciences overall stands at $21.58bn in 2026 and is projected to hit $69.34bn by 2031, a 26.3% CAGR (MarketsandMarkets, May 2026). Remember: pharma’s spend follows its risk, and nothing in pharma is riskier than a rejected filing.

For scale against our own records: Penelope Health drew a $100M commitment last week for real-time payer policy intelligence, and Spott closed a $21M Series A for AI-native recruitment software the same morning Biolevate announced. A €30M European Series A in applied AI is no longer an outlier; it is the going rate for a category leader with proof inside a regulated industry.

French applied AI is going vertical

So how do we read this round? Three signals stand out. First, the French AI teams raising serious Series A money right now are vertical ones – companies that picked one regulated industry and went deep, rather than shipping another horizontal copilot. Second, look at the cap table: two corporate investors in Orange Ventures and MSD Global Health Innovation Fund, plus Station F, alongside classic venture money from RAISE France and EQT Ventures. Corporates do not chase demos; they buy domain moats. Third, the expansion city is Boston, not San Francisco – for life sciences software, the customers and the FDA consultants sit on the US East Coast, and a Paris-Boston axis is the shorter bridge.

We keep telling ourselves Europe is behind on AI. The good news is that the race that matters here is not the model race: it is the workflow race, and the workflows of pharma – EMA procedures, HTA submissions, pharmacovigilance – are ones Europe understands better than anyone. The Dataiku diaspora building Biolevate is a reminder that our first generation of AI scale-ups is now seeding the second.

What to watch: whether the 20x revenue year repeats outside France, and whether the next quarter of European rounds keeps rewarding vertical AI over horizontal tools. The opportunity is clear – let’s build on it.