Ask PE operating partners what is stopping AI from scaling across their portfolios and the most common answer is people. In FTI Consulting's 2026 Private Equity AI Radar, 35% of respondents named AI talent and skills shortage as the top barrier, ahead of data readiness (33%) and time-to-value (29%). The most popular fix, chosen by 68%, is to hire more AI talent.
For a healthcare roll-up, that fix is usually the wrong one.
The Barriers Behind the Barrier
FTI's list of scaling barriers reads like a description of a typical MSO or DSO:
- Data readiness and accessibility (33%): patient and scheduling data spread across several EHRs and PMSs.
- Time-to-value and deployment speed (29%): a hold period that does not wait for a two-year build.
- Integration with legacy systems (28%): every acquired practice brings its own stack.
- Organizational change (25%) and siloed ownership (25%): front-desk teams at dozens of sites, each with their own way of working.
FTI concludes that scaling AI is "as much an operating model and leadership issue as a technical one." Hiring a data scientist does not fix siloed ownership or integrate a legacy PMS.
Why Hiring Doesn't Fit a Healthcare Platform
- The talent is scarce and expensive. Healthcare platforms compete with tech companies and funds for the same engineers.
- One hire is not a team. Putting AI into production takes integration engineers, workflow designers, compliance expertise and ongoing operations. A small team at a mid-market MSO becomes a single point of failure.
- It is slow. Recruiting, onboarding and building takes quarters. FTI's respondents mostly expect AI to pay back within 7 to 24 months.
- The work is operational. Much of the value in a roll-up comes from running front-office workflows well every day, and a research team is not built to do that.
The Other Options in FTI's Data
Hiring was not the only answer respondents gave. 59% pointed to strategic acquisitions, 59% to budget allocation and 51% to partnerships. For a healthcare platform, partnering is the fastest way to get AI capability, as long as the partner does more than sell software.
An AI-native execution partner provides the platform and also deploys it, runs it and is accountable for results. Samara works this way:
- The Rollup Engine (AIP) handles integration and data unification across acquired practices, the data readiness and legacy integration barriers.
- The Operations Engine (AIT) runs scheduling, intake, recall, reviews and patient communication at every location, so front-desk teams work with AI instead of learning to build it.
- Deployment takes under 30 days, which addresses time-to-value.
- The data, playbooks, configurations and workflows stay with the client on its Enterprise Sovereign AI OS. The platform gets the capability without depending on a vendor for it.
Where Internal Talent Still Matters
FTI describes three capability gaps that foundation models are opening: moving staff to higher-value work, embedding AI into daily workflows, and building orchestration and governance skills. A partner model does not remove the need for internal leadership. It changes the job. The platform needs an operator who owns outcomes and governance. It does not need an AI engineering department.
Next Step
Read about Samara for private equity, or book a demo to see how a platform closes its AI talent gap without a hiring plan.
Frequently Asked Questions
What is the biggest barrier to scaling AI in PE portfolios?
AI talent and skills shortage, cited by 35% of respondents in FTI Consulting's 2026 Private Equity AI Radar, followed by data readiness (33%) and time-to-value (29%).
Should an MSO or DSO build an in-house AI team?
For most mid-market healthcare platforms, an execution partner is faster and lower risk than hiring. Internal leaders should own outcomes and governance while the partner deploys and operates the AI.
Who owns the AI when a partner runs it?
With Samara, the client. Proprietary data and playbooks, configurations, custom-built IP and workflows always belong to the client enterprise.