By Ganesh Padmanabhan
There’s a growing debate online about what the future of skilled work looks like in a world where AI agents can automate increasingly complex tasks.
Much of that debate has been driven by what we’re seeing in software engineering: AI coding agents handling long-running tasks, dozens of agents operating in parallel, and credible claims that weeks of work are now being completed in hours. Even Boris Cherny, creator of Claude Code, recently noted that essentially all of his contributed code in a given month was written by AI.
It’s easy to see how that triggers anxiety. If AI can write the code, what happens to engineers? And more broadly—what happens to deeply skilled professionals in fields like medicine, clinical operations, utilization management, or care delivery?
Healthcare is where this question becomes most interesting—and most misunderstood.
Healthcare Is Not a “Task Job.” It’s a Judgment System.
At first glance, healthcare looks like a collection of tasks:
- Chart review
- Prior authorization decisions
- Clinical documentation
- Coding, billing, and quality reporting
- Care coordination and follow-ups
Many of these tasks can now be handled—at least partially—by AI systems. But that’s not what healthcare professionals actually do.
Healthcare work is about:
- Integrating incomplete, messy, and delayed information
- Making decisions under regulatory, clinical, and ethical constraints
- Managing risk in high-stakes environments
- Defending decisions months or years later under audit or appeal
- Coordinating across humans, systems, and institutions that don’t naturally align
AI can increasingly perform individual tasks in that system.
The job is knowing how those tasks fit together, what good looks like, and when the system is drifting toward error or harm.
That distinction matters.
AI in Healthcare Is a Force Multiplier, Not a Replacement
There’s a useful way to think about this shift.
As Max Levchin recently put it in another domain:
What we think of today as AI being a 10× productivity multiplier for skilled workers will soon flip—where skilled workers become a 10× multiplier on AI.
Healthcare is already there.
What AI can do well today:
- Extract and normalize data across unstructured clinical records
- Orchestrate workflows across siloed systems
- Draft determinations, summaries, and rationales
- Surface gaps, inconsistencies, and missing documentation
- Execute policy-driven steps at machine speed, without fatigue
What it cannot do alone:
- Decide when an exception matters
- Understand nuanced clinical context
- Balance access, quality, cost, and compliance simultaneously
- Take accountability for outcomes
- Redesign operating models when regulations change
That’s where expertise becomes leverage.
Productivity Multiplication Is Only the First-Order Effect
Most conversations stop at “productivity.” That’s incomplete—especially in healthcare. The real second- and third-order effects matter more:
1. Error Reduction in High-Risk Processes
AI agents don’t get tired, skip steps, or forget documentation requirements. When orchestrated properly, they reduce variability across:
- Utilization management decisions
- Appeals and grievances
- Quality and risk adjustment workflows
That directly impacts compliance exposure and patient outcomes.
2. Time-to-Care Compression
When administrative bottlenecks shrink:
- Prior authorizations move faster
- Care gaps close earlier
- Escalations surface before harm occurs
This is not just abstract efficiency—it’s real access.
3. Cognitive Load Rebalancing
Clinicians and clinical reviewers spend less time hunting for information and more time applying judgment. That’s how you get closer to practicing at the top of your license—without burning people out.
History Is Clear: Skills Don’t Disappear. Expectations Expand.
We’ve seen this before. Lawyers didn’t disappear when research went digital. Engineers didn’t disappear when infrastructure became abstracted. In fact, both professions grew—because the scope and complexity of what was expected expanded.
As Sam Altman has observed in another context, each generation of work looks “too easy” to the one before it—until you realize the bar has simply moved.
Healthcare is no different.
As AI takes on routine execution, the bar moves up—healthcare organizations will be expected to deliver better decisions, faster access, clearer traceability, stronger governance, and more personalized care. That requires more expertise, not less.
The Healthcare Professional of the Future Is an Orchestrator
The most valuable healthcare professionals in the AI era will be those who:
- Operate with a true end-to-end understanding of complex clinical and administrative workflows
- Exercise judgment in what to automate versus what must remain human-led
- Identify and intervene when systems produce outputs that are technically correct but clinically misaligned
- Continuously adapt operating models as regulations, standards of care, and risk profiles evolve
Less time spent executing. More time spent deciding. Full accountability for outcomes.
This Is a Rare Moment
This is a rare inflection point. AI lowers the barrier to entry into complex domains, while simultaneously expanding what excellence looks like at the top end.
For healthcare professionals who are systems-oriented, curious, and willing to adapt how they work alongside intelligent systems, this shift is not a threat—it’s leverage.
Core skills are not being displaced. They are becoming compounding assets. And in healthcare—where judgment carries real consequences and errors have human cost—that leverage may matter more than in any other industry.
If you’re building or operating healthcare workflows today, the question isn’t whether AI will change the work. It already has. The real question is whether we redesign the system—or let legacy processes waste a once-in-a-generation opportunity to improve care.
