Intro
A small $10 million fund would not usually move the needle in U.S. healthcare. But Treehub and its AI Health Fund, backed by Esther and Anne Wojcicki and a Stanford-heavy team, are quietly experimenting with something far more radical: redesigning how AI-first healthcare companies are born. If the model works, it could reshape who gets to build the next generation of healthtech — and how fast research escapes the lab and reaches real patients. This isn’t just another accelerator; it’s a testbed for a new kind of company-creation pipeline.
The news in brief
According to TechCrunch, investor and former Google product manager Mary Minno has launched Treehub, a six‑month residency program for early-stage founders at the intersection of AI and healthcare, along with a connected investment vehicle, the AI Health Fund.
Treehub’s residency splits into two 12‑week phases: the first focused on finding product–market fit, the second on setting the company’s direction — whether that’s raising a seed round, entering another accelerator, or piloting with hospital systems. Minno positions the program as “pre‑company”: in more than half of cases, Treehub even introduces founders to the lawyers who incorporate their startups.
The AI Health Fund plans to raise $10 million and write $50,000–$150,000 checks, primarily into academic spin‑outs. It has already completed a first close of $1.5 million, including $500,000 from family and friends and a $1 million commitment from investor Tim Draper. The fund aims to back around 60 companies, and has already invested in 12, including women’s hormone tracker Clair Health and a pediatric autism startup.
Anne Wojcicki joins as operating partner, while educator Esther Wojcicki serves as founding advisor. Faculty from Stanford’s biomedical data science department are also involved.
Why this matters
Treehub is less interesting as a fund and more interesting as a process experiment. The capital is small; the ambition is not. Minno and the Wojcickis are effectively asking: What if we treated the creation of healthcare startups as a structured, hands‑on residency — like medicine itself?
The primary beneficiaries are likely to be academic founders who sit on years of data and publications but lack the storytelling, regulatory and go‑to‑market skills venture investors expect. Traditional accelerators like Y Combinator or Techstars are optimized for software generalists and fast iteration, not for a dermatology researcher who needs to navigate ethics boards, clinical validation and insurer incentives.
By pairing academics with “operators” and dropping the performative demo day in favour of bespoke timelines, Treehub is acknowledging an uncomfortable reality: healthcare doesn’t move at SaaS speed. For AI in particular — which often needs access to sensitive health data, integration with hospital IT and clinician trust — trying to cram everything into a three‑month sprint can do more harm than good.
Who loses? Potentially, generic accelerators that claim to do “healthcare + everything else”. If Treehub shows that domain‑specific residencies produce stronger companies with fewer blow‑ups, capital will follow, and sector tourists will look increasingly out of touch.
There is also a subtle power shift. This model gives early, small‑check investors disproportionate influence by acting as quasi‑co‑founders and gatekeepers at the “idea to incorporation” stage. That can be extremely helpful — or constraining — depending on governance and diversity of voices in the room. Health systems and patients will feel those choices years later.
The bigger picture
Treehub fits into several converging trends:
Hyper‑specialised accelerators. Over the past five years, we’ve seen fintech‑only, climate‑only, and biotech‑only accelerators proliferate. Healthcare AI is ripe for the same treatment: it’s capital intensive, regulated, and deeply political. A founder who treats an algorithm for triaging cancer patients like another mobile app will not survive long.
The rise of “pre‑seed studios” and venture builders. Instead of passively waiting for founders, investors increasingly create companies from scratch, assembling teams around a thesis. Treehub doesn’t fully own the startups like a classic studio might, but its behaviour — introducing co‑founders, lawyers, and early strategy — blurs the line between accelerator and company‑builder.
AI as infrastructure, not a feature. Many of the most promising healthcare ideas are not “AI apps” in a consumer sense; they are embedded into clinical workflows, diagnostics, or hospital ops. That requires tight collaboration with institutions and regulators. A residency model that includes operators and academics is more aligned with this complexity than slide‑deck‑driven, batch‑based programs.
Historically, attempts to accelerate health innovation have oscillated between two extremes: corporate innovation theatres attached to hospitals that never ship real products, and move‑fast‑and‑break‑things startups that run afoul of regulators or ethics boards. Treehub is trying to carve out a middle path: move quickly, but embed clinical reality and regulatory awareness from day one.
Compared to large U.S. players like a16z’s Bio + Health efforts or Rock Health, Treehub is more of a precision instrument than a growth fund. It bets that deeply hands‑on help at the nano‑stage of a company is worth more than a larger check later. If it’s right, we should expect copycats — including in Europe.
The European / regional angle
From a European perspective, Treehub is a reminder of both an opportunity and a gap.
On one hand, Europe is strong where Treehub wants to play: academic excellence in medicine and life sciences; massive hospital networks; and, increasingly, rich health data infrastructure, from Nordic registries to the planned European Health Data Space (EHDS). Many of the world’s top AI‑in‑medicine papers still come out of European universities.
On the other hand, Europe struggles with exactly what Treehub is trying to fix: getting that research out of the lab and into production. Complex procurement processes, fragmented national systems, and a high regulatory burden (GDPR, the Medical Device Regulation, and soon the EU AI Act) raise the barrier for first‑time founders.
Treehub’s current focus is U.S.‑centric, but its model is easily transplantable — and arguably better suited — to Europe. Imagine a residency that combines:
- local hospital partners familiar with GDPR‑compliant data use,
- training on CE‑marking and MDR, not just FDA,
- and alignment with the AI Act’s requirements for high‑risk medical AI systems.
European healthtech founders today often ping‑pong between generic accelerators that ignore regulation and national grant schemes that move too slowly. A Treehub‑style program in Berlin, Barcelona, Ljubljana or Zagreb could bridge that gap, provided it’s built with regulators and clinicians at the table from day one.
For now, European founders might eye Treehub as a gateway into the U.S. market — especially those already collaborating with American universities. But the more strategic move for Europe is clear: build our own specialised residencies, tuned to our legal and cultural realities, before U.S. programs simply vacuum up the best talent.
Looking ahead
Treehub and the AI Health Fund are still in the experimental stage, and Minno is open about that. Several questions will determine whether this is a curiosity or a template for the next decade of healthtech company‑building.
First, can such a high‑touch model scale without diluting quality? Working “almost like a co‑founder” for 60 companies is labour‑intensive. Either Treehub stays relatively small and influential, or it develops a playbook others can adopt — perhaps via regional partners.
Second, we’ll see whether the check sizes ($50k–$150k) are enough to meaningfully de‑risk the earliest phase in a capital‑hungry sector. My bet: Treehub will eventually need follow‑on capital or close relationships with larger funds to avoid losing its winners too early.
Third, clinical and regulatory outcomes will matter more than vanity metrics. Over the next 3–5 years, the real indicators of success will be:
- How many portfolio companies achieve regulatory clearance (FDA, CE) or embed into major hospital systems?
- Do any reach Series B+ without catastrophic ethical or safety issues?
- Are patient outcomes measurably improved — not just workflows optimized?
For readers — especially founders and clinicians — the key signal to watch is whether other universities and health systems start copying the residency playbook. When you see European university hospitals announcing AI‑health residencies with embedded funds, you’ll know Treehub’s experiment worked.
The bottom line
Treehub and the AI Health Fund are small in dollars but big in intent: they treat healthcare AI startups less like lucky accidents and more like products of a deliberate, mentored residency. If they succeed, generic accelerators and arm’s‑length investors will find themselves outclassed by domain‑specific, pre‑company builders. The open question for European readers is simple: will this model be imported, adapted, or out‑innovated on our own turf — and who will control the pipelines that turn medical research into real‑world AI products?
