AI Failed to Catch a Nurse Anesthetist Stealing Drugs for Months, Coworkers Stepped In

An AI-powered drug-diversion system used at hundreds of U.S. hospitals failed to catch a months-long string of fentanyl thefts by a Tennessee nurse anesthetist, according to a state disciplinary document reviewed by CBS News. The case is raising fresh questions about how much hospitals can rely on automated monitoring to protect controlled substances and patients.

The nurse, John Stevenson, worked at Erlanger Baroness in Chattanooga, Tennessee. State records say he began diverting “unused fentanyl that would otherwise have been wasted after surgical procedures” in March 2025, using it once or twice a week at first before escalating to daily use by June.

Throughout that stretch, the hospital’s Sentri7 software stayed silent. It was human eyes, not the algorithm, that finally caught him.

On June 30, 2025, coworkers reported that Stevenson appeared impaired while on duty. According to the nursing board order, he “had slurred speech, appeared extremely tired, was seen standing with his eyes closed and swaying, exhibited head nodding while standing upright and appeared to have difficulty keeping his eyes open.”

He subsequently failed a drug test and was fired. When the hospital audited his dispensing records over the four-month period, it found approximately five instances when Sentri7 did not flag missing drugs, plus additional inconsistencies between drug dispensing and waste documentation that, the board said, should have been flagged by the automated monitoring system.

Stevenson was not charged with any crime in connection with the Erlanger case. He settled with the Tennessee Board of Nursing by signing a consent order in November 2025, and the state released the disciplinary order in December. His nursing license was placed on probation and he was required to undergo drug counseling.

>>Listen to The Latest Nurse News Podcast

Sentri7 is made by Wolters Kluwer, a Dutch technology company, and is used by roughly 700 hospitals, according to the company, while a competitor product called ControlCheck is used by more than 1,500. The systems are designed to spot the kind of dispensing and waste irregularities that often signal diversion.

The Tennessee board said Sentri7 was in its “initial learning phase” at Erlanger but offered no further details. Experts told CBS News they were struck that the technology was named so directly. “I’ve never myself seen these technologies be called out in that specific way,” said Jacob Smith, a pharmacist at Johns Hopkins Medicine. David Rastall, a Johns Hopkins neurologist and AI researcher, noted that “when an AI is found to be making some type of error, that becomes very transparent.”

Not everyone is convinced the software is to blame. Terri Vidals, founder of Rxpert Solutions, questioned whether the lapse stemmed from user error rather than malfunction, saying, “This is the most basics of basics for this software.” Erlanger and Wolters Kluwer declined to comment substantively, CBS News reported.

Drug diversion is not a rare problem on the floor. The Healthcare Diversion Network estimates that as many as 15% of all healthcare workers divert drugs at least once, which means most nurses will work alongside diversion at some point in their careers, often without realizing it.

This case is a reminder that automated surveillance is a backstop, not a substitute, for human vigilance. The technology missed months of diversion, but attentive coworkers who recognized the signs of impairment did not. Knowing the behavioral red flags, slurred speech, excessive fatigue, swaying, and nodding off, and feeling empowered to report them, remains one of the most reliable safeguards on any unit.

🤔 Have you ever relied on automated medication-monitoring software at your facility, and do you trust it to catch diversion, or do you think human oversight will always have to come first? Tell us in the comments.

If you have a nursing news story that deserves to be heard, we want to amplify it to our massive community of millions of nurses! Get your story in front of Nurse.org Editors now – click here to fill out our quick submission form today!

Crime