TL;DR
Turn quiet afternoons into listening time
- Thousands of audiobooks, podcasts and originals
- Listen on your phone, tablet or Echo — also offline
- Cancel anytime
At the Association of Health Care Journalists Fall Summit in Los Angeles, UC Berkeley researcher Mishal Khan said AI can harm healthcare workers even if it does not eliminate jobs. She cited risks including deskilling, workplace monitoring, reduced autonomy and a shift toward algorithmically managed, gig-style work; the scale of these effects remains unclear.
Mishal Khan, a senior researcher at the UC Berkeley Labor Center, warned that artificial intelligence could worsen working conditions and weaken healthcare workers’ skills, even if it does not eliminate their jobs. Speaking at the Association of Health Care Journalists Fall Summit in Los Angeles, she described risks ranging from workplace surveillance to automated task assignment and a possible shift toward gig-style healthcare work.
Khan grouped workplace AI into three categories: electronic monitoring, such as cameras tracking workers; algorithmic management, in which software evaluates or coaches workers; and task or job automation, where AI takes over parts of a role. She said these systems can carry risks including job loss, lower wages, work intensification, discrimination, reduced autonomy and suppression of labor organizing. These were risks she identified, not findings that every AI system causes those outcomes.
In her research on healthcare, Khan singled out deskilling as a concern: when automated tools take over specialized tasks, workers may lose practice and confidence. Stanford biomedical informatics professor Tina Hernandez-Boussard, speaking in a separate summit session, described medical residents who traveled to Uganda to support a hospital cardiac ward. After electricity was lost, they did not have AI available to read electrocardiograms. Hernandez-Boussard said the residents were not able to synthesize the information with enough confidence to make recommendations.
Khan also raised concerns about patient relationships, liability and worker data. Automated handoffs could affect how staff relate to patients, she said, while ambient documentation tools may collect information about employees as well as patients. She questioned how responsibility would be assigned if a clinical mistake followed an AI recommendation. The report did not identify a specific case or establish how often such problems occur.
How AI Could Reshape Care Work
The concerns go beyond whether software can perform a task. If clinicians lose regular practice interpreting tests or making decisions, the effects could become apparent when systems fail, power is unavailable or a case does not fit a tool’s recommendations. Hernandez-Boussard’s account illustrates that risk, though it does not establish that AI caused the residents’ difficulty or show how widespread the problem is.
Workplace tools can also affect control over schedules, pay and performance. Khan warned that algorithmically managed nursing platforms might let nurses bid for shifts while using predictions about the lowest wage a worker will accept. That is a potential model she described, not a confirmed description of a particular platform. Her broader point was that benefits to workers depend partly on whether they have power over how tools are introduced and used.
For patients, questions about accountability and continuity of care matter alongside potential efficiency gains. If AI informs a clinical decision, employers, clinicians and developers may need clear rules about oversight, data use and responsibility. The summit discussion raised those issues but did not resolve them.
AI monitoring system for healthcare workers
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
State Laws Address Specific Uses
Khan said some state legislatures are responding to concerns about AI in care. According to her remarks, laws passed in Illinois and Vermont prohibit replacing mental health workers with AI, while a California law bars companies from advertising an AI chatbot as a health professional. The report did not provide the statutes’ effective dates or detailed definitions of the restrictions.
Khan also said a California bill under consideration would prevent workers from being required to train their replacements, including an AI system. That proposal was not described as enacted. The examples show that policy responses are developing, but they address particular practices rather than settling every question about AI oversight or employment effects.
“Even if they don’t lose a single job to AI, it doesn’t mean there are no harms to workers.”
— Mishal Khan, PhD, senior researcher at the UC Berkeley Labor Center
How Widespread Are These Risks?
The summit report does not quantify how many healthcare workers have experienced deskilling, wage effects or increased monitoring, nor does it establish that AI caused the residents’ difficulty in Uganda. Khan’s examples describe potential harms and concerns; they do not prove that every healthcare AI system produces them. The report also does not specify which ambient scribe products collect worker data or how employers use it.
It remains unclear how liability would be divided when an AI recommendation contributes to an error, and what safeguards are in place across different workplaces. The status and scope of the California bill Khan mentioned are also not detailed. Those questions would require further reporting on specific systems, laws and workplace practices.
Track Rules and Workplace Evidence
The next developments include whether the California worker-protection bill advances and how states implement or enforce the laws Khan cited. Healthcare employers and technology providers may also face more scrutiny over monitoring, documentation data, clinical oversight and worker training as AI tools spread.
For now, the summit discussion is a warning about possible consequences, not a definitive assessment of AI’s effects across healthcare. Evidence from specific workplaces, along with details about safeguards and worker input, will be needed to establish which risks are occurring and how often.
Key Questions
What did Mishal Khan warn about?
Khan said workplace AI could harm healthcare workers through surveillance, deskilling, lower wages or reduced autonomy, even when it does not eliminate jobs. She described these as risks, not universal outcomes.
What is deskilling in healthcare?
Deskilling is the loss of specialized knowledge or practice when automated systems take over tasks that workers previously performed. Khan said this could affect healthcare workers’ skills and scope of practice.
What happened to the medical residents in Uganda?
Hernandez-Boussard said residents sent to help at a hospital cardiac ward faced a power outage and had no AI available to read electrocardiograms. She said they were not confident synthesizing the information to make recommendations; the report does not establish that AI caused the difficulty.
What AI-related laws did Khan cite?
Khan said laws in Illinois and Vermont prohibit replacing mental health workers with AI, and that California bars companies from advertising an AI chatbot as a health professional. She also described a California bill on requiring workers to train AI replacements as under consideration, not enacted.
Is AI already replacing healthcare workers?
The report does not provide figures on AI-driven job losses. Khan’s warning was that harms can occur even without job cuts, including changes to skills, monitoring and working conditions.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
