AI Rollups / AI Growth Buyout?
November 19, 2025
by a searcher from Technische Fachhochschule Berlin in Abu Dhabi - United Arab Emirates
I recently moved to the UAE to implement a buy-and-build strategy around MedRecords AI (www.medrecords.ai), my medical-records summarization platform.
The GCC is an unusually strong market for this: legal, medical, and insurance workflows are still largely manual; there’s a national push for broader insurance coverage; the population is growing fast; construction is booming (more claims, more cases, more records); and most owners of medico-legal BPO service businesses are not technical. The digitalization gap is wide.
Here’s the thesis I’m executing on:
I built MedRecords AI over the past two years. It’s already used by a few personal-injury law firms, specialty clinics (IME reports), P&C insurers (life-care plans), TPAs, and medical-legal BPOs. When integrated deeply, it can actually cut 30–50% of operating costs. Selling SaaS licenses captures at most ~10% of the value created. Owning the workflow captures nearly all of it, until the technical moat erodes (2-5 years).
The plan:
• Acquire an initial service provider (TPA / medical-legal BPO / records-review company).
• Integrate our AI stack end-to-end.
• Expand margins immediately.
• Roll up adjacent firms over 3–5 years.
This is a typical AI-enabled Growth Buyout play. More on it here: https://docsend.com/view/vp3w3g7djbirsx22
I’m looking for:
1. Operators, advisors, or investors who have done/seen similar tech-enabled service rollups in healthcare or insurance — happy to pay to learn from your playbooks.
2. Capital partners (family offices, independent sponsors, search investors) who understand AI-enabled margin expansion and want exposure to GCC healthcare/legal/insurance infrastructure.
If you have experience in this domain, are implementing something similar, or know investors who are interested in backing this strategy, I’d appreciate your intros. My DMs are open.
If you think there's something wrong with my assumptions above and you think it's a bad idea for valid reasons, feel free to discuss.
Btw:
Happy to license the software to anyone interested in implementing this playbook in other geographies (US, Canada, Western Europe) and share best practices.
from Cornell University in New York, NY, USA
in Sheridan, Wyoming, USA