Health Progress Articles

Artificial intelligence is already changing the way that many healthcare providers do their jobs. In many doctors' offices and hospitals, AI is working in the background. Computer algorithms scan X-rays and MRIs and flag any urgent issues for radiologists before they view them.¹ In the intensive care unit, advanced AI-enabled software monitors patients in real time to alert nurses if there is a problem. For physicians, it can translate to no longer having to meticulously take notes while listening to their patients' concerns, as voice tools automatically do the work for them.

These changes, among others, raise a host of questions about how AI will transform the healthcare workforce as we know it. What are the consequences for the people in these positions and the medical students hoping to get there? Will AI steal their jobs, as skeptics predict? Or can slow and steady adoption free doctors, nurses and others in the field to get back to doing what they do best: helping patients? But will all of this translate into faster but less personal care?

THOUGHTFUL INTRODUCTION AND UPDATES
AI needs to be incorporated methodically and with intention, according to Josh Clark, vice president of quality and safety operating systems for the Institute for Healthcare Improvement (IHI). "In the best-case scenario, the physician and nursing staff spend a lot less time documenting and looking for information, and more time at the bedside taking care of patients," he said. "But that has to happen by design; it won't just happen on its own. As we think about giving more time back to clinicians, we need to design it so that it results in actual patient care, and not just more and more patients seen faster because now we're more efficient with AI."

Clark's concern about AI's efficiency causing patient overload is especially worrisome to healthcare providers. The U.S. is projected to have a shortage of roughly 500,000 healthcare workers, including physicians, RNs and LPNs, by the year 2037, according to a December 2025 report by the Health Resources & Services Administration.² So, in addition to concerns about losing their jobs to AI, they are also concerned about the potential increase in their patient loads.

"If we're just using AI to create efficiency so we can see more patients and make more money, that's not going to benefit the greater good," said Clark.

Michael Kramb, SSM Health's deputy chief people officer, believes that for healthcare organizations to be successful, "they need to treat AI as a workforce and change management strategy" and not just a tool. "AI is really pushing education and workforce deeper toward adaptability and continuous learning," he said.

Done right, AI will retrain people rather than replace them, said Kramb. "AI gives us that opportunity to pivot and to redeploy versus displacement. Instead of eliminating those individuals, we can actually leverage AI to skill them up: either to a different level within their current position or teach them to go into something different. We now have the education and the tools and the venue to provide them with that skill set so that they can be successful in other roles within the organization."

The hope is that AI will reduce the burden related to tasks like documentation, allowing those working in healthcare more time to focus on their patients. Dr. Andrea Rock, a pediatric cardiologist with SSM Health in Wisconsin, uses ambient AI to record her patient appointments. "The best part is that it largely eliminates the computer between me and the patient and allows me to have more natural conversation than I have since EMR charting began, while capturing more details than paper charting or typical EMR use," she said.

"The best part is that it largely eliminates the computer between me and the patient and allows me to have more natural conversation than I have since EMR charting began, while capturing more details than paper charting or typical EMR use." — Dr. Andrea Rock

In Rock's case, the use of AI saves time and gives her that chance to interact more directly with her patients — but it has its shortcomings. "The reports that are being generated are still not as clear and natural as a physician's written thoughts," she said, "so in some cases the clarity of communication to other doctors is not as good if I don't supplement the AI-generated report." Still, the AI component is trainable around which details should be included. "At first, it would eliminate details about family, school, travel or hobbies as 'small talk,' but the more I work with the system, the better it does at capturing these."

One way that SSM Health is supporting its clinicians — specifically helping to prevent burnout — is through an AI-powered clinician retention and well-being dashboard being rolled out across the organization in partnership with AI technology company Atalan, according to Dr. Heather Schmidt, system medical director of employee well-being. "The clinician retention and well-being dashboard organizes already available data within a framework that focuses on seven core measurement domains that together capture the full spectrum of workload, environment and individual risk factors influencing well-being and retention across our clinical workforce," she said. "The goal is not simply to surface information. The goal is to help leaders at every level use earlier insight to drive more informed decisions, better conversations and stronger retention outcomes." SSM Health and Atalan are also building a similar tool with the organization's nursing staff. 

Schmidt says it is still early in the journey. "But this opportunity to think differently about how data can tell the story of workload burden and how it contributes to clinician well-being has the potential to be transformational."

ADAPTING EDUCATION TO AI
Medical education can be slow to adapt, and this will be difficult given the hyperspeed at which AI is being deployed and evolving, according to Clark.

"Graduate medical education changes usually take a decade. The nursing curriculum is still anchored in things that were being taught several decades ago. It also often takes six months to a year to get a new publication out," he said. "For us to be able to identify how AI is changing healthcare delivery and react in a way that mitigates risk and maximizes benefit, we're going to have to become more agile as an industry in education, in science and ultimately in application."

Creighton University's Lindsay Iverson, DNP, associate professor of nursing and director of Clinical and Simulated Environment for the Center of Interprofessional Practice, Education and Research, understands the challenge of moving education forward quickly while maintaining all necessary safeguards. She is constantly looking for new ways to challenge and support her students. These days, that is often with the help of AI.

By incorporating AI into her teaching, she has accomplished two important tasks: taking advantage of its time-saving features and introducing it to her students, as AI will become a larger part of nursing and other healthcare careers.

"I think we're going to start to see AI in clinical settings more and more through prediction models. Is a patient at risk for falls, for bed sores?" she said. "The AI can interpret all their medications and give you a summary. And students are also using AI to help write their clinical notes in a safe, HIPAA-compliant environment to improve their note-taking and management of that patient."

Iverson and her colleagues also carefully consider how they integrate AI into their nursing clinical education, particularly early in students' training when they are still developing their clinical and diagnostic skills. "One concern is that if our students develop an over-reliance on AI for tasks like clinical note writing, it could potentially limit opportunities for students to fully develop their own clinical reasoning skills." AI is not used as a replacement for thinking, but as a tool to enhance learning, she explained. "For example, students may first develop their own clinical notes and go through the diagnostic reasoning process, and then use AI to compare, refine and receive feedback on the note. This allows them to identify gaps, strengthen their clinical thinking, and improve efficiency over time," she said.

The school has also introduced RX24, a chatbot designed to help students study and master some of nursing's notoriously difficult subjects, such as antimicrobial pharmacology. "Students click on a link and have access to an avatar: basically, a one-on-one tutor based on the specific content. I uploaded the guidelines, past exams, past quizzes and all the PowerPoint lectures on that specific content, so students can ask any question they don't understand," explained Iverson. "Instead of turning to Google or messaging me at 2 in the morning when I'm sleeping, they can access this tutor 24 hours a day, seven days a week. It will also give them multiple-choice questions based on those materials, and they can keep quizzing themselves prior to the exam." This approach allows students to expand their skill sets and knowledge.

The school is also using chatbots to help students practice speaking with patients and their families. "Both our undergraduate and our graduate students said that they do not get enough practice navigating difficult conversations in healthcare. When they got to the clinical setting and had to deal with a patient who needed to receive some difficult news, they felt very unprepared," said Iverson. "So, we thought, what could we do at Creighton University to prepare them, to build that competency, to build that confidence, so they could really handle those difficult conversations, whether it's delivering a terminal diagnosis or telling a family member that their patient was in a car accident?"

"A lot of universities are telling students, 'Don't use AI.' We're saying, 'Use it, see what works, see what doesn't work, and then try to improve upon that.' The whole point of using AI ethically is to give our students those safe learning environments so they are set up for success to use AI appropriately." — Lindsay Iverson

All of this is helping prepare students for environments that are rapidly evolving to include more AI. "We know that our students at Creighton University are going to use AI on most things that they do. We want to prepare them to have AI literacy: to use AI safely and ethically," said Iverson. "A lot of universities are telling students, 'Don't use AI.' We're saying, 'Use it, see what works, see what doesn't work, and then try to improve upon that.' The whole point of using AI ethically is to give our students those safe learning environments so they are set up for success to use AI appropriately."

Anna E. Price, PhD, a professor of public health in the College of Health Professions at Sacred Heart University, sees faculty responses ranging from full adoption to strict limits. "But most see AI as a tool with clear benefits and drawbacks. It can assist with learning, writing and test prep, but it also raises concerns, including environmental impacts and broader societal effects," she said. "A major challenge is assessment," she said, explaining that educators are rethinking how to evaluate genuine student understanding, not just a student's ability to leverage AI. "For some, this is driving greater use of in-class exams, oral assessments and hands-on simulations."

Even with the myriad challenges, AI is driving important innovation, said Price. "It encourages educators to reevaluate long-standing practices and try new teaching methods," she said. "To do this effectively, institutions need to invest in professional development and foster open, inclusive discussions about how AI is, or is not, used in the classroom."

PATIENTS SEEK HUMAN CONNECTION
Those same discussions are taking place outside the classroom as well, especially among patients and what they do or do not want from their providers and AI.

AI will reshape at least some of the healthcare workforce by changing roles, creating new opportunities, and potentially automating certain tasks, said Price. But one point she has found in her research is clear: "Patients do not want AI to replace their providers," she said.

Price and her research partner, Foluke Omosun, assistant professor in Sacred Heart's School of Communication, Media and the Arts, recently conducted a national poll with GreatBlue Research on healthcare and AI.³ Poll results showed that AI use in seeking health information is already widespread. "But the public expects transparency and choice. A large majority wants to know when AI is used, understand its role, and have the option to opt out," she said.

According to the poll, more than one-third of U.S. adults already use AI tools to research medical topics, and 61% rely on search engines that now integrate AI responses. Americans demand transparency and choice: Nearly 88% want disclosure whenever AI is used in care, 83% believe patients should have the right to opt out of AI-driven care, and approximately 86% want AI's role explained in plain language. The poll also found that people in the U.S. have concerns about AI and data privacy, but they are optimistic, with more than half believing that AI can improve access to healthcare and information.

Price keeps the poll results in mind as she chooses how to integrate AI into her teaching and prepares her students for careers that are evolving because of it. She also tries to emphasize the importance of deploying AI equitably.

"Preparing students involves equipping them to collaborate effectively with AI while maintaining strong clinical judgment and ethical decision-making," said Price. "It also requires addressing issues of equity. Variations in access to AI tools and training could exacerbate existing disparities in health and digital literacy. As AI becomes part of healthcare systems, deliberate investments in AI literacy across all communities are essential."

Like many other educators, Price believes that AI should be understood as a tool to support care, not replace human connection. "Evidence shows that patients value and expect personal relationships with their providers, and that must remain central," she said. "Importantly, the commitment to human dignity and whole-person care means that technology should enhance, not undermine, the clinician-patient relationship. Preserving empathy, trust and personal connection is essential."

EMBEDDING ETHICS AND OVERSIGHT IN DECISION-MAKING
As part of SSM Health's commitment to carefully integrate AI into the healthcare workplace, the organization established an AI Ethics and Oversight Committee in 2023. The committee is composed of people with clinical backgrounds and leaders in IT and other areas, like human resources, said Kramb.

"It's not just about governing what type of AI technology we're using but looking at what does work look like in the future," he said. "We need to start asking the questions and planning for the future, because if we can leverage AI as a workforce strategy, that's how we can be well positioned to react to the changes that we know will be coming."

IHI's Clark agrees; even just a couple of years in AI is an "eternity," he said. "We're going to have some fundamental changes in how things are being done, and we just need to be really intentional and strategic."

ELIZABETH GARONE is a freelance writer who has covered health, business and human-interest topics. Her writing has appeared in The Wall Street Journal, The Washington Post, BusinessWeek and The Mercury News, among other publications.

NOTES

1. Hanae Armitage, "Artificial Intelligence Rivals Radiologists in Screening X-rays for Certain Diseases," Stanford Medicine, November 20, 2018, https://med.stanford.edu/news/all-news/2018/11/ai-outperformed-radiologists-in-screening-x-rays-for-certain-diseases.html.
2. "Health Workforce Projections," HRSA, December 2025, https://bhw.hrsa.gov/data-research/projecting-health-workforce-supply-demand.
3. "Healthcare and AI Poll," Sacred Heart University, November 25, 2025, https://www.sacredheart.edu/media/shu-media/press-room-2026/SHU-2025-Health-Care-AI-Poll-ADA.pdf.