Skip to content
Submit incident
AII-1000-86211
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

AI drug monitoring system failed to detect nurse's fentanyl theft

An AI-powered medication monitoring system at a Tennessee hospital failed to detect months of fentanyl theft by a nurse. The system, Sentri7, overlooked missing drugs and irregularities that should have triggered alerts, raising concerns about the effectiveness and transparency of AI drug diversion detection software used in hundreds of hospitals.

1 June 2026 Β· Wolters Kluwer Β· Reliability and hallucination
1 June 2026
AII-1000-862112 sources ↓
Quick facts
Category
Reliability and hallucination
Country
United States
Source
CBS News
Date
1 June 2026
Drafted by
Raidu Incident Detection

What happened

At Erlanger Baroness hospital in Chattanooga, Tennessee, a nurse stole fentanyl over a period of months, sometimes daily, by taking leftover drugs from surgical procedures. Anesthesia staff noticed the nurse's impairment on duty, including slurred speech and difficulty staying awake, after which the nurse failed a drug test and was fired. The nurse later admitted to the theft in a Tennessee Board of Nursing consent order.

The hospital used Sentri7, an AI-powered medication monitoring system designed to detect missing drugs faster than human oversight. For months, Sentri7 failed to raise alarms about the missing drugs and other inconsistencies that should have been flagged, according to the nursing board order.

Reported impact

The case raises particular concern because theft of leftover drugs is one of the most common diversion methods, and fentanyl is among the most frequently diverted substances in hospitals, with potency fifty times stronger than heroin.

Index analysis

Healthcare facilities are not required to disclose their implementation of medication monitoring software or report malfunctions to anyone, creating little transparency about how widely AI drug diversion detection programs are used or how often they fail.

David Rastall, a Johns Hopkins Medicine neurologist and AI researcher, noted that because AI technology is heavily proprietary and hospital officials often do not understand how it works, errors are buried rather than fixed, allowing errors to be repeated at other hospitals. Drug diversion prevention experts said they had never seen AI system failures publicly documented like the Erlanger case.

Governance controls

The Drug Enforcement Administration mandates that hospitals confidentially report lost or stolen drugs, and hospitals can also report to state health agencies, but these reports are not required to include details about any AI software involved.

Classification

Category
Reliability and hallucination
Industry
Healthcare, drugs and biotechnology
Country
United States
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Public interest
Harmed
Workers, General public
AI systems
Sentri7

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

CBS News
At a Tennessee hospital, a nurse stole fentanyl and AI missed it, state records say
1 June 2026
Breitbart
AI Drug Monitoring System Failed to Detect Nurse's Fentanyl Theft at Tennessee Hospital
1 June 2026