As artificial intelligence becomes increasingly integrated into healthcare, a critical question arises: can only well-resourced health systems afford this technology?
In August, Daniel Yang, MD, vice president of artificial intelligence and emerging technologies at Kaiser Permanente, voiced concerns to The Wall Street Journal about the widening gap in AI adoption.
"The AI 'haves' will be large, well-resourced systems like Kaiser Permanente that invest in testing, evaluating, and responsibly deploying AI technologies for the benefit of our members," Dr. Yang said. "The AI 'have-nots' will be health systems like county hospitals, federally qualified health centers, and rural hospitals that lack the infrastructure or expertise to deploy these technologies effectively, or that do so without fully understanding their capabilities and limitations."
Rebecca Mishuris, MD, chief medical information officer at Mass General Brigham, echoed these concerns to Becker's.
"They're currently very expensive and are probably pricing out some health systems and providers as a result. If this has an opportunity to truly be an inflection point in provider burnout, we don't want to leave health systems out because of the cost," she said. "And so I'm hoping that we see the cost come down dramatically."
However, some industry leaders argue that the cost of AI tools has already begun to decrease.
"One of the exciting things about the field is how rapidly the pricing structures are evolving along with the tools themselves," Sara Murray, MD, chief health AI officer of San Francisco-based UCSF Health told Becker's. "Tools like AI scribes which were unaffordable for many health systems just over a year ago now cost a fraction of the price. As these tools are continually improved upon, I expect value to health systems will continually improve."
Similarly, Sarah Hatchett, senior vice president and CIO of Cleveland Clinic, pointed out that while investing heavily in AI offers a competitive edge, broader access to AI is becoming more feasible.
"Cloud-based and open-sourced solutions require technical expertise but can significantly reduce the cost of AI development," she told Becker's. "Most health systems will more likely pursue strategies around vendor partnerships and embedded solutions in enterprise platforms to leverage the power of AI rather than an internal build approach due to the cost barrier."
As the landscape of AI in healthcare continues to evolve, the challenge will be ensuring that its benefits are accessible to a diverse range of health systems, not just the most resourced ones.