The Centers for Medicare & Medicaid Services has quietly built a payment model designed for AI agents, and according to TechCrunch AI, most of the tech world hasn’t noticed. The program is called ACCESS (Advancing Chronic Care with Effective, Scalable Solutions), a 10-year CMS experiment that picked 150 participants to test what AI-driven medical care looks like at federal scale. It goes live July 5.
TechCrunch AI reports that one of those participants, Pair Team, was accepted on April 30. Founder Neil Batlivala put the shift bluntly: “The government is creating swim lanes for AI innovation in traditionally regulated industries. The best solution wins, which, in regulated industries like healthcare, that’s not been the case.”
What ACCESS Actually Changes
Traditional Medicare pays for clinician time. Minutes with a doctor, check-ins on a schedule, activities you can bill for. There’s no line item for an AI that calls a patient at 9pm to confirm they took their meds, or coordinates a housing referral, or just talks to someone for an hour.
ACCESS creates that line item for the first time.
The model rewards outcomes, not activities. Participating organizations get predictable payments for managing qualifying conditions, and they earn the full amount only when patients hit measurable goals like lower blood pressure or reduced pain. Covered conditions include:
- Diabetes
- Hypertension
- Chronic kidney disease
- Obesity
- Depression
- Anxiety
This is the real news. As Batlivala told TechCrunch AI, “It’s a payment model transformation. You just couldn’t do this before.”
Who’s in the First Cohort
The 150 participants range from AI doctor startups to virtual nutrition therapy providers, connected device companies, and wearable makers like Whoop. Batlivala is skeptical of some of the fit. “I’m a big fan of wearables, but for a senior who’s struggling with food insecurity, I don’t know how much Whoop is going to be able to do.”
Pair Team’s own bet is voice AI. About nine months ago, the company deployed an agent called Flora as its primary patient-facing interface. It runs 24/7, handles intake, coordinates referrals, and keeps patients engaged between visits. The first call that shifted Batlivala’s thinking was with a 67-year-old woman living out of her car, managing PTSD and congestive heart failure. She spoke with Flora for over an hour. Hour-long conversations are now routine. “That’s the companionship piece. And it turns out that is truly an intervention.”
Why This Matters for AI Builders
What stands out here is who designed ACCESS. The program was built by Abe Sutton, Director of the CMS Innovation Center and a former healthcare VC at Rubicon Founders, and Jacob Shiff, Chief AI and Technology Officer at the Innovation Center and a former healthcare founder. Their fingerprints are all over the design: outcome-based payments, direct-to-consumer enrollment, and deliberate competition between vendors.
For AI companies, that’s a federal reimbursement pathway that didn’t exist a year ago. The catch is that CMS is paying less per patient per month than many participants expected. Batlivala calls that a feature: “If you want to build a model that truly incentivizes the use of AI, the reimbursement rates have to be low. The economics only work if you’re running a lean, AI-first op.”
Translation: human-heavy care teams can’t make the math work. AI-first ones can.
The Risks Worth Flagging
Two concerns sit on top of this rollout:
- Data exposure. Participants are feeding extraordinarily sensitive patient data, intimate conversations about housing, disease, and mental illness, into federal infrastructure with a documented history of breaches, including exposed Social Security numbers.
- Mixed financial track record. A 2023 Congressional Budget Office analysis found that the CMS Innovation Center increased federal spending by $5.4 billion in its first decade rather than producing projected savings.
Still, ACCESS is the clearest signal yet that federal payers are willing to write checks for AI agents, not just clinicians. July 5 is the live date. The companies that figure out how to deliver outcomes inside that envelope will set the template for the next decade of regulated AI healthcare. More details at the original source.