Hospital CEO Says He Would Replace Radiologists with AI Right Now If He Could

The CEO of America's largest public hospital system says he is ready to start replacing radiologists with artificial intelligence, raising urgent questions about the future of clinical roles in healthcare, including nursing.

Mitchell H. Katz, MD, president and CEO of NYC Health + Hospitals, made the remarks during a March 25 panel discussion hosted by Crain's New York Business. The 11-hospital system serves over a million New Yorkers annually, and Katz said AI could deliver "major savings" by handling first reads of mammograms and X-rays.

"We could replace a great deal of radiologists with AI at this moment, if we are ready to do the regulatory challenge," Katz told the audience.

Katz joined a discussion with One Brooklyn Health CEO Dr. Sandra Scott and WMC Health Network CEO Dr. David Lubarsky and all three CEOs discussed that they are open to AI replacing Radiologists and some other clinician roles.

Hospital CEOs Push for AI-Only Image Reads

Katz, who has led NYC Health + Hospitals since 2018, outlined a vision where AI handles initial breast cancer screenings and radiologists step in only to double-check abnormal findings. He asked fellow hospital CEOs on the panel whether they should push for changes to New York state regulations that would allow AI to read images "without a radiologist."

• Fellow panelist David Lubarsky, MD, MBA, president and CEO of the Westchester Medical Center Health Network, said his system is already deploying AI for breast cancer screening with strong results. "For women who aren't considered high risk, if the test comes back negative, it's wrong only about 3 times out of 10,000," Lubarsky told the audience, calling the technology "actually better than human beings."
• Sandra Scott, MD, CEO of One Brooklyn Health, agreed. "I mean, I'm in charge of a safety-net institution. It would be a game-changer," Scott said about using AI to replace radiologists at her hospital system.

 >>Listen to The Latest Nurse News Podcast

Radiologists and Researchers Push Back Hard

The response from the radiology community has been swift and fierce. Mohammed Suhail, MD, a San Diego-based radiologist with North Coast Imaging, called Katz's comments "undeniable proof that confidently uninformed hospital administrators are a danger to patients." He said administrators are "easily duped by AI companies that are nowhere near capable of providing patient care."

"Any attempt to implement AI-only reads would immediately result in patient harm and death, and only someone with zero understanding of radiology would say something so naive," Suhail added.

The controversy follows similar remarks from Anthropic CEO Dario Amodei, who recently claimed AI has surpassed radiologists at reading scans. Radiologists roundly criticized those statements as well.

Adding weight to the skeptics' case, Stanford University researchers recently discovered that leading AI models used for medical imaging can "hallucinate" findings on X-rays they never actually analyzed. The researchers called it an "AI mirage," where models construct detailed, plausible-sounding diagnoses without any real visual input. In the most extreme case, one AI model achieved top marks on a chest X-ray benchmark without ever viewing an image.

Meanwhile, a February 2026 CNN investigation found that radiology has actually become a "case study for why AI won't replace human workers." Demand for radiologists continues to climb, with American diagnostic radiology residency programs offering a record 1,208 positions in 2025 and the specialty ranking as the second-highest-paid in medicine at an average income of $520,000.

What Nurses Need to Know

While this debate centers on radiologists, nurses should be paying close attention. When hospital administrators openly discuss replacing one clinical role with AI to cut costs, it signals a broader willingness to automate healthcare positions wherever possible. Nursing roles that involve data interpretation, triage decisions, or pattern recognition could face similar scrutiny down the line.

Patient safety is the most pressing concern. Nurses are often the last line of defense in catching errors, and introducing AI-only diagnostic reads without adequate human oversight could increase the burden on bedside nurses to identify missed or incorrect findings. If AI misses a cancer on a mammogram, it may fall to nurses to recognize symptoms that should have been caught during screening.

For nurses working in radiology departments, imaging centers, or oncology settings, this conversation directly impacts workflow and job security. And for all nurses, it reinforces the importance of advocacy, both at the bedside and in the policy discussions shaping the future of healthcare delivery.

🤔 What do you think, nurses? Should hospital CEOs be pushing to replace clinical roles with AI, or does patient safety demand human oversight? Share your thoughts in the comments.