Healthcare Jobs Will Change. The Workflows Have to Change First.

There is a better way to talk about artificial intelligence and healthcare jobs. The conversation often starts with fear. Which jobs will be replaced? Which teams will shrink? Which departments will be automated first?

That fear is understandable. Healthcare workers have already lived through years of staffing pressure, documentation burden, payer-provider friction, revenue leakage, denial volume, audit exposure, and new technology that sometimes adds work instead of removing it.

But history gives us a more useful operating lesson. Technology usually does not start by replacing an entire job. It starts by changing the most repeatable task inside the job. Then the economics of the workflow change. Then the role changes.

That distinction matters for payer and provider operations. Healthcare does not need a future where people are pushed out of judgment-heavy work. It needs a future where skilled teams spend less time chasing status, reconciling mismatched data, assembling repetitive documentation, and moving information from one system to another.

The ATM Lesson Was Never Only About Banks

Grant Lee the CEO of Gamma used a strong analogy about automated teller machines (ATMs) and bank tellers because it shows something that gets missed in technology shifts. The machine took over a repeatable task first. Cash withdrawals, deposits, and balance checks became easier to handle without a person at the counter.

That did not end the branch model overnight. It changed the economics of the branch. A location could operate with fewer tellers, which made it easier for banks to open more locations. As the branch footprint changed, the teller role changed too. Less time went into the cash drawer. More time moved into customer questions, account support, fraud issues, small business needs, and relationship banking.

The lesson is in the sequence.

AI implementation will not affect every healthcare function the same way. It will move first into work that is repetitive, rules-based, documentation-heavy, or dependent on matching information across fragmented systems. That includes claims research, eligibility checks, prior authorization packet assembly, denial categorization, payment variance review, chart retrieval support, quality documentation checks, risk adjustment preparation, and member service triage.

Those are not small areas. They sit close to cost, revenue, access, compliance, and experience.

Other Industries Show the Same Pattern

The Thomas Guide had its own version of this transition. For decades, a printed map book was part of how people moved through Southern California and other markets. Then navigation became live, searchable, and connected to traffic, delivery, dispatch, local search, ride-share, field service, and logistics. The old workflow was damaged. But live routing became one of the layers that made new work models easier to scale.

Music went through a similar shift. Cassettes and compact discs gave way to downloads and then streaming. Physical retail, packaging, trucking, and legacy distribution channels were hurt. At the same time, more independent artists found paths to listeners, and new work formed around rights management, audience data, playlist strategy, creator tools, music marketing, licensing, analytics, and global discovery.

The video rental sector had a harder landing. The local rental store model did not get a clean transition. Blockbuster's decline, Netflix streaming, and the later closure of remaining video rental chains showed how painful technology shifts can be when the old operating model cannot adapt quickly enough. But video demand did not disappear, it moved into streaming platforms, content operations, original programming, localization, editing, licensing, and recommendation systems.

The early pain usually lands on workers closest to the old workflow. Some companies fail. Some skills lose value. Some workers do not get a fair bridge to the next market. That is the part healthcare leaders should take seriously. Optimism only matters if it leads to operating discipline.

Healthcare Is Different Because the Stakes Are Different

Healthcare cannot treat AI adoption like a simple productivity project. A bank teller error may frustrate a customer. A healthcare workflow error can delay care, create a billing dispute, trigger a denial, expose a compliance issue, damage trust, or put a vulnerable patient through another round of administrative work.

That means AI needs a different operating model in healthcare. The goal should be human-controlled AI inside accountable workflows. AI can help find patterns, summarize records, flag missing documentation, classify incoming work, identify claims anomalies, detect revenue leakage, and guide staff toward the next best action.

But judgment still matters.

A prior authorization decision needs clinical context, policy alignment, documentation discipline, and appeal readiness. A risk adjustment workflow needs coding integrity, defensible evidence, and audit controls. A claims integrity process needs contract knowledge, payment logic, exception handling, and governance. A member service workflow needs empathy, clarity, and escalation judgment.

AI can support those workflows. It should not be allowed to quietly become the accountable decision-maker without oversight, validation, and clear ownership.

The Work Most Likely to Change First

In payer and provider operations, the first wave of change will likely hit the parts of work that are high-volume, repetitive, and information-heavy.

For providers, that includes revenue cycle work where teams spend too much time checking claim status, preparing appeals, reconciling underpayments, gathering medical records, reviewing denials, and identifying missing documentation.

For payers, it includes claims review, utilization management support, risk adjustment operations, quality documentation, provider data maintenance, member issue routing, and audit preparation.

For both sides, the common problem is fragmented visibility. The information exists somewhere, but it is split across systems, teams, vendors, portals, spreadsheets, call notes, documents, and inboxes. Staff become the glue. They search, copy, paste, reconcile, chase, and escalate.

That is expensive work. It is also fragile work. When the workflow depends on a person remembering where to look, what to check, who to ask, and how to document the decision, variation becomes normal. At small scale, variation looks like inconvenience. At large scale, it becomes leakage, delay, compliance exposure, avoidable appeals, staff burnout, and member or patient frustration.

Adaptation Cannot Be Left to Individual Workers

There is an unfair version of the AI transition where leaders tell workers to adapt but leave the workflow untouched. That will not work well in healthcare.

A claims processor cannot adapt if the system still requires five portals, two spreadsheets, and unclear escalation rules. A care coordinator cannot adapt if the member record is incomplete and no one owns the handoff. A revenue cycle team cannot adapt if denial insights arrive too late to change intake, authorization, coding, or documentation behavior. A compliance team cannot adapt if AI outputs are used without audit trails, validation, or policy governance.

Adaptation has to be designed. Payers and providers need to map the work at the task level, separate repeatable administrative steps from judgment-heavy decisions, and define where AI can assist, where human review is required, and where automation should not be used.

They also need to measure the transition in practical terms:

A Practical Operating Model for Payer and Provider Leaders

The starting point is simple. Do not begin with the job title. Begin with the workflow.

What This Means for Healthcare Leaders

The opportunity is not to replace experienced people with software. The opportunity is to redesign administrative work so experienced people can spend more time on judgment, service, controls, and improvement.

That matters because healthcare already has too many places where broken workflows create cost and friction.

AI can help with healthcare cost control, but only when it is connected to workflow discipline, clean data, human oversight, and measurable outcomes.

Optimism Requires a Plan

Healthcare workers have adapted many times before. They adapted to electronic health records, changing payer rules, value-based care models, new coding requirements, remote work, staffing shortages, quality measures, and more complex member and patient needs.

AI will require another round of adaptation. The difference is that leaders can make this transition more practical if they treat it as operating redesign instead of software deployment.

Healthcare does not need blind optimism about AI. It needs disciplined optimism, the kind that recognizes the disruption, protects the people doing the work, strengthens compliance, and improves the experience for the patient, member, beneficiary, or caregiver on the other end of the process.

Frequently Asked Questions

Farzin Espahani

Strategic Advisor, Insurance Growth, Medicare Distribution, and Compliance

Farzin Espahani is a Strategic Advisor focused on insurance growth, Medicare distribution, healthcare operations, and compliance-aware performance marketing.

He has led growth, marketplace operations, partner distribution, and Medicare marketing initiatives across regulated healthcare and insurance markets, with experience spanning QuinStreet, Health IQ, Solution21, and The Patient's Guide, which was acquired by IAC Publishing.

At QuinStreet, he worked across Medicare and health insurance distribution, performance marketing, partner economics, lead quality, compliance alignment, funnel performance, and AI-enabled consumer decision support, including the development of an AI-powered Medicare plan recommendation experience for Insure.com.

Farzin also serves in industry leadership and advisory roles with the Insurance Marketing Coalition and Lead Quality Alliance, where his work supports ethical marketing, lead quality, consumer transparency, and compliance-aware growth.

His academic background in biocultural anthropology, sociology, and social psychology informs how he evaluates consumer behavior, incentives, trust, decision-making, and organizational friction. His executive education includes programs at Massachusetts Institute of Technology (MIT) Sloan, Wharton, Cornell, and Northwestern Kellogg, with advanced study in artificial intelligence (AI), revenue analytics, pricing, leadership, change management, and communication.

For VHealth, Farzin brings a practical advisory lens on healthcare growth, workflow visibility, regulated distribution, compliance-aware interpretation, and responsible AI-enabled decision support.