Kaiser nurses say AI that rates their empathy and tracks their call time is forcing them to cut short conversations with suicidal and terminally ill patients

A nurse at Kaiser Permanente cut short a phone call with an elderly woman who had just received a terminal cancer diagnosis. Not because the patient was stable. Not because another call was waiting. She cut it short because she was afraid that staying on the line would damage her monthly performance score and trigger a reprimand from her manager 1.

This is the AI-in-healthcare story that has nothing to do with diagnosis. Seven current and former Kaiser nurses told CalMatters that the health system uses AI to rate their empathy and tone of voice, productivity-prediction software to assess their daily output, and call-length metrics that penalize anyone who spends more than 15 minutes on a single call 1. The technology is not trying to replace the clinician. It is managing the clinician. And the metrics it enforces are pulling in the opposite direction from the care it claims to support.

Raquel Alvarez Sanchez, a Kaiser Permanente advice nurse in Vallejo since 2010, described staying on a call for more than an hour with a suicidal patient, waiting for police to arrive. She knew the call length would throw off her average handle time for weeks and could trigger questions from management. She stayed anyway 1.

"I think at some point all of the nurses have been talked to about their average handle time," Sanchez said. "The only thing I can think of is they're doing it for profit" 1.

The nurse who cut short the call with the terminally ill cancer patient asked to remain anonymous, fearing retribution. She told CalMatters she initially thought the woman was suicidal, then realized she was in shock and needed someone to talk to. The woman cares for her daughter. The nurse wanted to stay and offer comfort. Instead, she caught herself: "I had to ask myself: Am I going to get disciplined for going off script or saying more than what is necessary?" 1

That question is the whole story. When you algorithmically score empathy, you create an incentive structure that penalizes the conversations empathy actually requires. The hour-long call with a suicidal patient. The gentle tone with a dying woman. The instinct to stay on the line a little longer, coded as a performance failure.

Kaiser Permanente denies using call-duration data for evaluations. Spokesperson Vincent Staupe said Kaiser does not use average handle time to assess agent performance or enforce call time metrics, and that any contact-center tools support quality assurance efforts with human review and oversight 1. Staupe declined to share details about internal systems, citing security and operational reasons 1.

The gap between that statement and what seven nurses describe on the record and anonymously is where the tension lives. Nurses say they are routinely called into performance evaluation meetings for calls exceeding 15 minutes. Sanchez, a union steward, said she has accompanied colleagues to those meetings where they were found to have done everything right on a call except staying on the line too long 1.

Kaiser Permanente is the largest private employer in California, providing healthcare to more than 9 million people in the state and 3 million more Americans elsewhere 1. The California Nurses Association is bargaining for 25,000 nurses this month, including roughly 1,000 in call centers. Kaiser nurses already went on a one-day strike against AI in March and picketed last fall 1.

California lawmakers are simultaneously debating several bills on AI in the workplace, including one that would protect doctors and nurses from retaliation when they override automated care recommendations 1.

The patient impact remains difficult to measure. A 2024 public records request by CalMatters to the California Department of Managed Health Care found no patient complaints against Kaiser related to call times 1. Call center nurses lose contact with patients after hanging up, so downstream effects are invisible to them. Michele Ramos, a patient advocate with Consumer Watchdog, told CalMatters that many Kaiser patients begin their care journey on the advice line and she suspects that is where problems originate 1.

Sanchez has not seen a nurse fired specifically for a long call. But she fears the sustained pressure will push nurses to quit or retire early 1.

The standard AI-in-healthcare question is whether machines can match human judgment. This story poses a harder one: what happens when machines manage the humans, and the management metric penalizes the compassion that makes the care worth receiving? An empathy score that makes nurses afraid to be empathetic is not measuring empathy. It is measuring compliance. And the patients on the other end of the line have no way to know the difference.