2026's most anticipated AI advances—and how docs are navigating the promise and pitfalls
Industry Buzz
Many AI tools are currently in pilot stages that have the potential to significantly improve quality of care, expand access, and improve clinician well-being.
—Thomas Kingsley, MD
I hope that AI will allow me to stop having to interface with the hospital’s electronic medical record systems for ordering medications, tests, and arranging follow-up for patients without me having to turn my back on them to type into a computer.
—Jorge Nieva, MD
AI has entered the healthcare space, with many physicians embracing new technology. But experts say rapid advancements in AI present both opportunities and challenges.
“AI is a double-edged sword. It’s much more efficient for doing literature searches than searching with a web browser, so for people who already have expertise, it’s a tool that increases efficiency and reduces noise when asking a question," Jorge Nieva, MD, a medical oncologist at USC Norris Cancer Hospital, tells MDLinx..
"I worry, though, about trainees using it, because reading all that extra noise in manuscripts and textbooks is how one builds context and better understands how the body works, how drugs work, and how diseases affect us," Dr. Nieva says. “Being told the answer to a question is efficient, but you learn a lot more and retain the information better if you filter through a lot of writing and do a lot of thinking to get that answer."
AI can improve nearly every part of the workday
AI has already changed some workflows across the healthcare space. Thomas Kingsley, MD, director of applied artificial intelligence at UCLA Health, says AI’s potential in healthcare is far-reaching.
I share the broad enthusiasm about AI’s potential to address some of healthcare’s most persistent challenges. That potential spans nearly every clinical and operational workflow.
—Thomas Kingsley, MD, director of applied artificial intelligence at UCLA Health
"AI-powered clinical decision support can help clinicians deliver more standardized, evidence-based care; tools that summarize complex patient charts can reduce time [spent] chart reviewing significantly; ambient AI scribes can meaningfully reduce documentation burden; and AI can help triage patients, improve navigation of the health system, and streamline scheduling, ultimately improving access to care," Dr. Kingsley says.
Related: AI is here—5 ways it’s already changing medicineAI hallucinates studies
But AI in the health space is not without flaws. Last year, the FDA’s AI tool, “Elsa,” was unveiled. [] The tool was meant to speed up drug approvals. But FDA employees raised the alarm that the tool was making up studies, a phenomenon known as AI hallucination.
“I am concerned. Over the past 3 years, we’ve seen an explosion of AI startups and healthcare products, many of which are untested, and some of which have real potential to cause harm. Hallucinations—where an AI system confidently generates false or fabricated information—are one clear example of how these systems can fail in high-stakes settings like medicine,” Dr. Kingsley says.
AI could transform clinical practice
The experts who spoke with MDLinx are hopeful that AI will help streamline clinical practice.
“I hope that AI will allow me to stop having to interface with the hospital’s electronic medical record systems for ordering medications, tests, and arranging follow-up for patients without me having to turn my back on them to type into a computer,” Dr. Nieva says.
Dr. Kingsley argues that physicians will benefit most from AI if they understand that all AI systems have positives and negatives.
“Physicians should approach AI tools the same way we approach any other tool used in clinical practice: by understanding both their strengths and their limitations. No AI system is perfect, and each one has its own failure modes—just as human clinicians do," he says.
He continues: “This is really just the beginning. At UCLA, many AI tools are currently in pilot stages that have the potential to significantly improve quality of care, expand access, and improve clinician well-being. Providers have been overwhelmed for years by growing administrative and documentation demands—many [of which are] driven by a complex payer-provider dynamic in the US. AI won’t solve every problem, but when applied thoughtfully, it has the potential to meaningfully reduce that burden and allow clinicians to focus more on patient care.”