The Disruptive Impact of AI in Health Care

It can be helpful to recognize the nature of artificial intelligence (AI) as a disruptive technology within health care, explained Kenneth A. Richman, PhD, professor of philosophy and health care ethics at Massachusetts College of Pharmacy and Health Sciences University in Boston, during a presentation at the American Society of Health-System Pharmacists (ASHP) Pharmacy Futures 2024 in Portland, Oregon. When brought in as a new technology for the health care workforce to utilize, the disruptive nature of AI will naturally lead to varying reactions from the workforce when asked to bring applications of the technology into their practice.

“Disruptive technology is a term that people use sometimes, and let's think about what it means [to define] AI as disruptive,” Richman said. “Whether good or bad, AI shakes things up. It forces change, and it's disruptive in the sense that we were going one way, and we may have been doing okay, or thinking that we're doing just great, but then all of a sudden we're forced to change, and we may not be clear on whether that's a good thing or not a good thing.”

For example, when pharmacies moved from manual written ledgers to keeping patient records and pharmacy records in a computer, some pharmacists were upset and did not want to change how they did their jobs. According to Richman, these individuals may have felt that their system worked fine, and they did not see a need for change. These individuals may, as a result, have resisted the change of using computers; for them, paper documentation was a stable form of documentation that they did not need to move away from.

“But then, when enough people, enough systems, and enough pharmacies start using computers, [these pharmacists] couldn't continue to do it the old way, and they were forced to change to using the new technology,” Richman said.

A similar issue occurred with use of X-ray technology, according to Richman. The X-ray, which was invented at the end of the 19th century, allowed providers to observe information that was not possible to observe before the invention of this technology.

Examining Mauser Bullet by X-ray, 1900. Image Credit: © Wellcome Images

“Disruptive technology can allow us to literally see things that we could not have seen before,” Richman said. “In this image of some surgeons looking at a bullet in a soldier, it wouldn't have been possible to find this without the X-ray.”

However, Richman noted that disruptive technologies can distract us from what we would have seen without the use of the technology, in certain cases. In the example of the painting Examining Mauser Bullet by X-ray, 1900, because the X-ray highlights things to see that are new, this can draw attention away from or hide other things, according to Richman.

“One of the most remarkable things that I see in this image is that [these] surgeons are not looking at the patient, they're looking at a screen,” Richman said. “The patient's right there. He's a whole person, and they're just fascinated by the screen. No one's looking at his face, right? It's going to give a lot of information. No one's looking at his skin, or the wound on the outside.”

In this way, the surgeons examining the Mauser bullet in the painting are fixated on a screen in a similar fashion as modern humans have become fixated on screens with the technology we have available to us today, which will shape our perception in a different way than without the presence of a screen, according to Richman.

In a similar manner, machine learning allows us to see things we could not have seen without its presence, as machine learning can help us make use of data in quantities that human beings, even groups of human beings, could never capture in such large amounts, Richman explained. Similarly, the use of machine learning may shape our focus towards the capabilities of these tools, guiding us to fit how we process our work within the scope of what is possible for these new technologies to process.

Richman explained further that when thinking of how we want AI to work for us within health care, the value of governability will remain important. Image Credit: © Werckmeister - stock.adobe.com

Richman explained further that when thinking of how we want AI to work for us within health care, the value of governability will remain important. Specifically, when thinking of the future of AI within health care, the governability of AI as a tool has implications for its ethical use within health care.

“Governability is an interesting value. We don't want a situation like in The Terminator movies, The Matrix, or Robocop, where we have what would be a really good system, except that we can't stop it once it gets going,” Richman said. “So, governability is important. Even if it's very good and does what we want it to do, we need to be able to control it and monitor it, and it shouldn't be too independent.”

Notably, Richman explained that sustainability, it turns out, is an even greater issue than Hollywood films have led us to believe regarding the problems that will occur with the development of AI.

“I've learned since I put these slides together that the computing power that's required to do even the kinds of things that we did to generate these slides is immense. An AI-driven image search or image generator is much, much more resource intensive than a regular old-fashioned search [from] even a couple of years ago,” Richman said. “There are vast resources, with a significant carbon impact with use of these of these new technologies. So, sustainability is an issue.”

Reference

Richman KA. (Joseph A. Oddis Ethics Colloquium) Ethical Dimensions of AI in Pharmacy Practice, Part I: AI Behind the Counter. American Society of Health-System Pharmacists Pharmacy Futures 2024; June 8-12; Portland, Oregon.