Scaling Requires Governance: Navigating the Complex Terrain of AI in Healthcare
Joshua Tamayo-Sarver, MD, PhD, FACEP, FAMIA
As a technologist at heart, I’m naturally drawn to speed, efficiency, and the thrill of building something new. However, my experience in healthcare has taught me a valuable lesson: true innovation in this field requires more than just technical brilliance; it demands collaboration and a deep understanding of the healthcare ecosystem and inner workings.
It’s not enough to simply create tools that function well — those tools must also seamlessly integrate into clinical workflows and enhance, rather than complicate, the work of healthcare providers. To do this, we need the input of clinicians, administrators, data scientists, and even patients themselves. By combining technological advancements with the insights of those who work on the front lines of healthcare, we can create solutions that are both groundbreaking and truly impactful. Innovation, after all, isn’t just about what’s possible — it’s about what makes a real difference in the lives of patients, clinicians and caregivers.
A Tale as Old as Time, or at least healthcare technology innovation
I have seen countless companies and technologies that have promising healthcare technologies and an amazing proof point from a solid pilot. Afterward, many of them pivot to start scaling, boasting a huge sales funnel where everyone seems eager to purchase their product every day of the week and twice on Sunday. And then, well, nothing. Failure. Exit stage left. The behind-the-scenes story is usually pretty similar and repetitious:
- The company had a great idea and a great team that understood the problem space.
- They followed best practices to find a definitional customer, get an alpha version and rapidly iterate until they handed a great MVP at that site.
- They nail their product-market fit and hit full throttle on the sales motions.
- The health system clients like the value proposition, find the pilot compelling, and want the product TODAY.
- But wait. They followed the tried-and-true consumer tech playbook of “move fast and break things” to develop a great working product in record time at minimal cost. What they didn’t do, however, was create a robust solution with healthcare enterprise-level security, governance, and compliance standards. While their definitional customer may have made exceptions and implemented alternative safeguards, no one else will. No matter how impressive their product is, no other healthcare CIO or CISO will approve technology that jeopardizes the ecosystem upon integration. As a result, they can’t scale.
- The company must go back to the drawing board and start over with appropriate governance, security, and compliance and rebuild the product. Game over.
- Why me?
The healthcare industry is unique in its complexity and the stakes involved. When we talk about implementing AI in healthcare, we’re not just discussing a new app or a cool feature — we’re talking about tools that can potentially impact human lives. This reality demands a level of responsibility and accountability far beyond what’s typically required in other tech sectors. When we “break things” in healthcare, those “things” are also called “people”, and people do not enter the healthcare system to be broken.
To truly harness the power of AI in healthcare, we must bridge this gap between theory and practice. This means getting our hands dirty, diving deep into the trenches of healthcare delivery, and understanding the nuances of clinical workflows, regulatory requirements, and patient needs. It’s not enough to have a brilliant algorithm; we need to ensure that it can be seamlessly integrated into existing systems, comply with stringent security standards, and actually improve patient outcomes. This isn’t just reasonable — without it, no one will purchase, implement, or use the technology. And without that, there’s no scaling or successful exit.
What do I do?
This is where governance becomes crucial. As we scale AI solutions in healthcare, we need robust frameworks to ensure these technologies are developed and deployed responsibly. This isn’t about stifling innovation — it’s about fostering innovation that is durable, scalable, and ultimately profitable.
I’m fortunate to work within an incubator that’s part of a larger healthcare organization. This allows us to start with an established foundation of people, processes, governance, compliance, and technology infrastructure. These elements ensure we can meet even the most stringent information security requirements without issues. In fact, my AI engineering team operates within the security team, further streamlining our ability to address these concerns from the outset.
Implementing these governance structures may seem daunting, especially for startups and innovators eager to bring their solutions to market quickly. However, I’ve learned that this approach actually accelerates innovation in the long run. By addressing these critical issues upfront, we can help build AI solutions that are not only technically sound but also trusted by healthcare providers, embraced by patients, and more importantly capable of scaling across diverse healthcare environments.
This approach aligns perfectly with the needs of healthcare systems and payers. These stakeholders are looking for long-term partnerships with technology providers who understand the complexities of healthcare and are committed to delivering sustainable, impactful solutions. There are certainly plenty of venture-backed tech companies that have a short time horizon for an exit rather than building a sustainable product or business. However, that approach conflicts with my goals as a physician and is antithetical to the values of healthcare providers. I want my solutions to be durable, meaningful, and scalable.
As we continue to push the boundaries of what’s possible with AI in healthcare, let’s remember that our ultimate goal is to improve patient care and outcomes. This requires more than just cutting-edge technology — it demands a holistic approach that combines technical expertise with deep healthcare knowledge, ethical considerations, and robust governance frameworks.
Ultimately, scaling successfully in healthcare seems strikingly similar to the old saying “If you want to go fast, go alone. If you want to go far, go with a group.” I would assert that if you want to have a shiny demo and a great sales funnel quickly, invest first in your UX and sales motions. But if you want to have a durable, scalable, and profitable health technology business, invest first in your governance, security, and compliance infrastructure.
In healthcare, success isn’t just measured by speed to market — it’s about trust, reliability, and long-term impact. By prioritizing the foundations of governance and security, you not only scale farther but also ensure you’re improving patient care in a meaningful, lasting way.
