Where health systems are heading with AI

Artificial intelligence continues to make headlines and dominates a lot of the discussions across the healthcare industry.

Vendors are certainly enthusiastic about the technology and its potential, which was apparent this week at the HLTH 2024 conference in Las Vegas. 

Related: How 3 health systems decide when to buy or build AI

But health system executives are more cautious about the hype of AI. They are trying to understand the risks, opportunities and processes needed to adopt the technology. The cost of generative AI is increasingly a big consideration for many systems.

Here’s what executives at seven healthcare organizations said about where they stand with AI today. The responses have been edited for length and clarity.

Dr. Sonya Mahkni, medical director, Mayo Clinic Platform

“Overnight, ChatGPT and generative AI technology made AI more approachable to a lot of folks that hadn't really thought about it. Before it was a series of point solutions, predictions and mathematics. Now there’s [an AI chatbot] talking to you and making you respond to it. There was a lot of that initial swirl but there’s now a core part of our workforce that is training and looking at it from different angles. It’s gap analysis. After the initial [excitement], it is looking at the potential holes [in the technology] and the benefits. How do you actually materialize those benefits?”

Byron Yount, chief data and AI officer, Mercy

“We've benefited from tempered expectations. That's because we've been pushing technology envelopes for a while, and if you've done this for a while, you understand that there’s promise and there's a hype cycle with it. But that hasn't stopped anyone from wanting to push on the accelerator. You want to do it in a safe way but there's still a lot of hope for it right now and a lot of interest in it.”

Todd Schwarzinger, partner, Cleveland Clinic Innovations

“We’re embracing it. The piece that has been most important is setting up the governance infrastructure. We've re-established an AI task force. We hired our first chief AI officer. It's a strong dedication to making this work and making sure that we're bringing the best use case to our patients and our caregivers. It's a challenge to wade through all of the new solutions that are out there, prioritize and adopt the ones that are meaningful.”

Dr. Baligh Yehia, president, Jefferson Health 

“AI is here, and there are existing use cases where it is helpful. I think it’s going to stay. … I think it’s really around discovering those use cases, those use cases that help providers deliver better quality care, help folks drive more efficiencies through the system and then create that ROI of investing in that technology versus investing in a nurse, investing in security, in a clinic, in all the other things that systems invest in."

Dr. Christopher Sharp, chief medical officer, Stanford Health Care

“The temperature is still pretty hot. We talk about AI a lot at the strategic level. Our leadership is talking about AI a great deal. We've invested in structures, governance and that is really going to propel us forward. It’s not about whether individual members of the organization like this or not, but whether everyone sees that [AI] is coming.”

Melissa Cohen, chief innovation and transformation officer, Cayuga Health System

"There are some limited back-end administrative uses of gen AI in our system right now, especially around data standardization and organization. There’s a ton of discussion on the ambient AI space for provider documentation. I’m excited to look at models applying that to nurse documentation and other documentation to speed things up there."

Dr. Tommy Ibrahim, CEO, Sanford Health Plan

“Sanford is on the leading edge of this. We have many examples of where we're already using the technology. There’s ambient listening/AI scribing. We have predictive tools that are embedded across our clinical enterprise to look at how we can proactively identify patients who might be at greater risk of chronic diseases. We’re using AI to even predict nursing staffing models and how to try to get ahead of those challenges as well.”