A bitter union dispute in New York City has sparked an urgent debate over whether hospitals are quietly trading seasoned nurses for automated algorithms.
We have all heard the jokes about robots taking over our day jobs. But nobody really expected an algorithm to clock in for scrubs and a stethoscope.
That scenario just became reality for healthcare workers in New York City. A group of veteran nurses says their hospital handed their duties straight to software, and they are sounding a loud alarm.
This is no longer a distant theoretical concern. It is happening in hospital hallways today, with direct consequences for patient care.
The Modern Healthcare Crunch
Skyrocketing administrative costs and persistent staffing shortages are driving hospitals straight toward automated systems.
Healthcare facilities across the country are facing serious operational strains. Between demanding shifts and administrative overhead, clinicians are experiencing unprecedented burnout rates.
Federal workforce projections from the Health Resources and Services Administration show a nationwide deficit of nearly 109,000 registered nurses. To manage these gaps, community hospitals now lean on machine learning systems to forecast patient volume and handle routine operational tasks.
Hospital administrators want to cut overhead, and tech vendors are offering digital tools to streamline operations. But friction quickly emerges when budget cuts replace human clinical oversight with computer code.
The Showdown in the Bronx
View this post on Instagram
A dozen veteran nurses in the Bronx received layoff notices after being instructed to process patient files through a private software platform.
The conflict surfaced at Montefiore Medical Center in the Bronx. Just months after nurses concluded a 41-day citywide strike to secure stronger contract safeguards, twelve utilization review nurses received 45-day layoff notices. These were not entry-level personnel. Marilyn Shuler, for instance, had dedicated 39 years of service to Montefiore patients.
Utilization review nurses examine complex medical charts daily. Their clinical evaluations demonstrate to insurers that extended inpatient care, specific medications, or surgeries are medically necessary.
Following the strike, the nurses were directed to run charts through Datavant, a healthcare platform partnered with Palantir. Shuler noted that staff initially assumed the software was only intended to clear a strike-related documentation backlog.
Save this article
Enter your email address and we'll send it straight to your inbox.
Management then issued the termination letters. “We just can’t believe that a machine can replace all the decades of clinical judgment and knowledge that we have,” Shuler stated. The New York State Nurses Association filed a class-action grievance, alleging the move violated collective bargaining protections against unilateral technological displacement. NYSNA President Nancy Hagans declared that artificial intelligence should never replace real human caring from a nurse.
The union also cited compliance and privacy concerns, pointing to Datavant’s $900,000 settlement regarding a 2024 data breach that affected more than 50,000 records. Montefiore administration disputed the union’s characterization. Senior Vice President Joe Solmonese called the union’s claims inaccurate and misleading, saying the hospital deployed a nonclinical program to manage paperwork and improve operational efficiency.
Related: 10 Common Myths About Artificial Intelligence
The Fight Over Clinical Intuition
Survey data and clinical device studies show that frontline practitioners remain skeptical of algorithmic assessments in medical decision-making.
Automated systems often struggle to evaluate complex, subtle symptoms such as pain tolerance, anxiety, or gradual patient deterioration. A survey of over 2,300 registered nurses conducted by National Nurses United highlights deep clinician concern. Sixty percent of respondents reported that they do not trust hospital leadership to prioritize patient safety when deploying automated systems.
Frontline clinicians report frequent discrepancies between algorithmic outputs and bedside reality. Safety analyses reflect these concerns. A study published in JAMA Health Forum found that 43 percent of recalls for AI-enabled medical devices occurred within one year of regulatory authorization.
Meanwhile, healthcare industry organizations are pursuing balanced integration strategies. Rather than advocating for workforce displacement, the Johnson & Johnson Foundation has taken the position that technology should support nurses rather than replace them.
The Johnson & Johnson Foundation provided $2.5 million alongside Google.org to support a $5 million American Nurses Enterprise educational initiative. The grant focuses on training frontline clinicians on AI literacy and ethical oversight, ensuring practitioners maintain active governance over clinical care.
Through educational partnerships with institutions like Duke University School of Nursing, Johnson & Johnson maintains that digital tools should alleviate clerical burdens while keeping licensed nurses at the center of clinical judgment.
The Real Stakes for Patients and Providers
Automating utilization review risks undermining the primary line of clinical defense between patient care and insurance payment denials.
Utilization review nurses serve as critical patient advocates within complex reimbursement systems. When software reviews medical records automatically, it can easily miss subtle clinical documentation that justifies essential hospital stays. While software flags structured data points, licensed nurses evaluate holistic clinical needs that may not fit neat algorithmic criteria.
Bedside practitioners also worry about operational precedent. If back-office nursing roles are fully automated, health systems may eventually rely on algorithms to assign floor tasks to lower-cost, unlicensed personnel. Such transitions risk treating medical care like an automated supply chain rather than an individualized clinical service.
A Reality Check for Hospital Automation
Digital platforms can process documentation quickly, but genuine patient safety relies on licensed clinical discernment.
Healthcare requires more than repetitive data entry. While software tools can organize electronic charts, they lack the diagnostic intuition gained through decades of clinical practice. The labor dispute in New York demonstrates that rolling out automated systems without clinician collaboration invites significant workplace friction and regulatory scrutiny.
As automated tools continue to expand across hospitals, what boundaries should be established to ensure technological efficiency does not override professional clinical judgment?
