Fireside Chat: Anne Klibanski MD and Tom Mihaljevic MD

Session Focus

Sperling introduced the conversation from the perspective of having chaired Partners HealthCare, Mass General Brigham and previously the Brigham and Women’s Hospital board. He framed the shared task of academic medicine as providing exceptional care to the highest-acuity patients cost-effectively, broadening access, and sustaining research and teaching missions that, as he put it, nobody wants to pay for.

The discussion covered research strategy, the historical perspective on healthcare’s challenges, partnerships as a route to scale, AI readiness, and what the panelists called the democratization of healthcare.

Research as Differentiator

Anne Klibanski MD identified research as what distinguishes Mass General Brigham, on the grounds that it determines the care the system can deliver.

The scale: roughly $2.5 billion in research funding, and still the largest recipient of NIH funding. Structurally, the system has organized around institutes in cancer, heart and vascular and neuroscience to concentrate effort, with gene and cell therapy as a central area and a new AI Institute established.

She highlighted three specific efforts.

ARPA-H funding, a relatively new federal mechanism, with Mass General Brigham in a hub position. Two projects she named: scaling genome editing therapies for rare genetic diseases, and the PARADIGM program, which takes hospital-level care and distributes it nationally through a different delivery model.

RECTIFIER, which addresses a gap she called fundamental and very difficult to solve. In a system with millions of patients and large structured longitudinal data, matching the right patients to the right clinical trials remains hard. The tool uses AI to do that matching, which she noted matters particularly at a time when the majority of clinical trials are conducted abroad.

A Historical Corrective

Tomislav Mihaljevic MD opened by reading out an assessment of healthcare covering research funding pressure, hospitals operating at 95 percent occupancy and declining Medicare reimbursement. He then disclosed that the interview was recorded 22 years ago, appearing in a local Cleveland publication when the Cleveland Clinic was opening what was then its largest hospital.

His point: if we think for a moment that we live in unique times, we have to think again. When people invoke the good old days, he asks them to pick a year and explain what made it better.

Klibanski agreed strongly, describing a nostalgia distortion she encounters at her own institution, where colleagues recall events she witnessed differently or that did not happen at all. She attributed it to the limitations of wishful historical recall, and noted that claims of unprecedented times generally lack any historical view of the world, the country or medicine.

What Has Actually Changed

Mihaljevic located the genuine shift in a long-standing constraint. The quality of care at either institution has always had limited ability to scale to the patients who need it. Digital transformation and AI are the first plausible answer to that historical problem, across both operational challenges and investment in research and education.

Cleveland Clinic’s research focus is translational, and the largest change in its portfolio has been sustained investment in digital and computational infrastructure, which he expects to be the differentiator in research quality. The organization also invested early in quantum computing, and he believes it still operates the only quantum computer dedicated to a healthcare center.

Partnerships and the Insularity Problem

Klibanski named an irony in academic medicine. Academics think globally, seeking the best science and techniques wherever they are, yet academic health systems tend toward insularity as organizations.

Her counterexamples were partnerships that expand reach rather than capability alone:

Amazon One Medical, a highly efficient and easily digitalized primary care model, which expands what Mass General Brigham can do in primary care for the right patients while freeing capacity for those who genuinely need tertiary and quaternary care.

CVS, addressing what she called pharmacy deserts, which exist not only across rural America but within Massachusetts, by bringing a degree of primary care into settings where people already go.

K Health, also in primary care.

Her broader argument was that partnership is a way for academic medical centers to expand impact, through technology and through a different relational model.

Mihaljevic framed the same instinct differently. Asked how the Cleveland Clinic measures success, he answered in the number of lives touched by Cleveland Clinic care. He described it as a puzzling fact that large academic medical centers, like most universities, remain single zip code businesses despite having the most profound societal impact and serving the most profound societal needs. Cleveland Clinic now operates across three continents and multiple US states.

His own partnership examples included IBM, which brought the quantum computer, and an emerging relationship with Oracle aimed at a new platform making primary care both accessible and sustainable. On pharmacy deserts specifically, Cleveland Clinic is piloting and beginning to implement drone delivery, with medications arriving at a patient’s home by the time they return from the appointment.

AI Readiness

Mihaljevic argued healthcare is among the faster adopters of AI, for the right reasons, since safeguards come naturally in an industry where quality and patient safety are paramount.

The underestimated aspect, in his view, is that enterprise deployment is genuinely complex. Many think of AI as magical dust to be sprinkled. In reality, the enterprise receiving it has to be well structured and organized. A fragmented organization with no standardization, differing governance and differing practices will see limited effect. His formulation: a poorly organized enterprise with great technology becomes a poorly organized and much more expensive enterprise.

Klibanski said the point landed directly for Mass General Brigham. When she announced the department consolidation and three institutes in spring 2024, the necessity was clear but the difficulty was not fully appreciated. Two great academic medical centers with entirely different governance structures, functions and data cannot get far on AI until that is unified.

On readiness among people, Mass General Brigham is running intensive AI training for its top 400 to 450 leaders by year end, with training becoming broadly available. Her reasoning was specific: the risk is that those most adept at using AI rise to the front, and they may not be the best people. It needs to be a skill available to everyone.

She also observed that patients and clinicians are ahead of institutional programs. Physicians go to open evidence platforms on their own, and patients use AI constantly, sometimes surfacing a rare diagnosis no one else identified.

On where movement is fastest, she pointed to image-based modalities, and noted that some of the holdup in moving faster is CMS payment rather than capability. Digital pathology coupled with imaging, laboratory tests and clinical data she expects to be genuinely profound.

Democratizing Healthcare

Klibanski raised the question of whether all care must happen in one location or with one specialist, and what technology makes possible for rural areas. She noted that rural is not only distant states; Nantucket and Martha’s Vineyard qualify.

Mihaljevic agreed the phrase is strange but accurate. The country has created exceptional higher education and exceptional healthcare as rare goods accessible to too few, despite both being among the most impactful services available to citizens. His position: excellent healthcare should not be a rare, exquisite and exclusive good, and making it more accessible is a collective mandate.

He closed by setting a task for the group, that in ten years there should be an upbeat conversation about healthcare, with people looking at their healthcare as a source of pride and a source of good in their community.

Key Takeaways

1. Today’s pressures are not historically unique. A 22-year-old interview described the same capacity, funding and reimbursement problems.

2. Scale has always been the constraint on quality care. Digital transformation is the first credible answer to it.

3. AI cannot fix a fragmented organization. Structural unification has to precede deployment or cost rises without benefit.

4. AI readiness is a workforce issue. Training has to be broad so capability does not concentrate by accident.

5. Partnerships extend reach beyond the zip code. Primary care, retail pharmacy and delivery collaborations widen access.

6. Payment, not capability, is holding back imaging AI. CMS reimbursement was named as the specific constraint.