If the COVID-19 pandemic taught us anything, it is this: primary care physicians are the backbone of a functioning health system. Yet, in the United States, the supply of those physicians is dwindling at precisely the moment demand is peaking. The Association of American Medical Colleges (AAMC) for example, projects a shortage of up to 124,000 physicians by 2034, including at least 17,800 in primary care. For patients, that shortage translates into longer wait times, rushed visits, and missed opportunities for early intervention, while for physicians these shortages manifest in “burn out”. Indeed, according to a study by Stanford Medicine, 62.8% of doctors in 2021 reported experiencing at least one symptom of burnout, up from 43.9% in 2017.
This is by no means a recent issue. For decades, policymakers and educators have attempted to plug the gap. Medical schools have expanded enrollment. Nurse practitioners and physician assistants have been given greater authority in many states. Health systems have reorganized clinics to streamline operations. Yet progress is slow. Residency slots remain capped by outdated Medicare funding limits, while the burden of medical school debt continues to deter future doctors, especially those from underrepresented communities.
In other words, the pipeline is constricted, and the solutions of the past won’t be enough for the future.
For entrepreneurs and health leaders, it is a call to rethink the model itself.
AI as a Physician Extender
If training more clinicians can’t solve the shortage, the next best option is to help existing clinicians practice at the top of their license and reclaim time that’s currently wasted. This is where artificial intelligence (AI) enters the conversation. Unlike workforce expansion, which requires decades of investment, AI can begin improving efficiency today by lightening the physician’s load.
Consider the following of how AI can help primary care physicians:
- Automated medical scribes: Early rollouts in the UK indicated that 4 out of 5 general practitioners (GPs) believe that automated medical scribe tools free up doctors’ attention, with 80% of GPs reporting they saved time and improved patient rapport.
- Chatbots: AI-assisted triage tools, like Cedars-Sinai’s CS Connect AI chatbot platform have been used to support more than 42,000 patients, reducing administrative burden and speeding up care navigation.
- Decision support: Studies have found that medical AI decision support tools can improve “optimal” recommendations largely due to their capabilities to capture and critically analyze information from intake or EMR data allowing physicians to adapt their recommendations accordingly.
Yet even efficiency gains inside the exam room only go so far. The bigger opportunity, and the real promise of AI, lies in reimagining how primary care is delivered in the first place. Instead of waiting for patients to show up when something is wrong, AI can shift the entire relationship from sporadic encounters to a continuous, data-driven partnership.
Traditionally, primary care has been episodic: patients schedule visits, doctors respond reactively. AI is shifting that paradigm toward a continuous experience. Remote patient monitoring powered by AI (e.g., using wearables, sensors, and predictive algorithms) can track vital signs in real time, detect anomalies early, and prompt timely interventions.
Instead of sporadic encounters, patients receive ongoing oversight, while clinicians gain better insights into their patients’ health trajectories. This continuity reduces the risk of crises and redistributes physician attention to where it’s needed most.
These developments aim to turn primary care into an efficient continuous data-driven process rather than periodic visits.
Building AI That Works in the Real World
But efficiency and continuity alone are not enough.
As AI takes on more clinical tasks, a new concern emerges: will technology distance patients from their doctors? The very tools designed to relieve burnout and extend capacity risk sparking fears of a colder, less personal form of care.
This fear of AI is not abstract. It’s deeply human. But when deployed wisely, AI can do the opposite. By stripping away administrative drudgery, AI gives clinicians more time to connect. Listening, explaining, and engaging are the moments that build trust, and can flourish when doctors aren’t buried in paperwork. AI in healthcare is not created with the purpose of replacing doctors, but rather to let them stay true to their job by taking on the heavy, repetitive tasks.
The promise of AI will only materialize if the tools are designed with human connection and safety at their core. AI must integrate seamlessly into existing workflows; no clinician has time for clunky software that adds friction instead of removing it. Privacy and transparency are non-negotiable. And developers must collaborate with clinicians and patients throughout the design process to ensure tools address real-world challenges rather than introducing new ones.
Looking Ahead
AI is not a silver bullet. Solving the physician shortage will still require broader investments in training and retention. But ignoring the efficiency gains of AI is no longer an option. The pandemic revealed how fragile our healthcare workforce is. The next decade will reveal whether leaders are willing to adapt.
The choice is stark: either continue down the current path where thinly stretched clinicians struggle to meet rising demand or embrace AI as an extender of human capacity. Done right, AI will not replace the art of medicine. It will preserve it.
About Thomas Kluz
Thomas Kluz is a distinguished venture capitalist with over a decade of experience. He’s the Managing Director of Niterra Ventures, where his investments focus on energy, mobility, and healthcare. With deep expertise in healthcare-focused venture capital, he has a proven track record of success with various organizations, such as Qualcomm Ventures and Providence Ventures.
