Part of The Corporate Pivot arc. Companion to the Substrate Law keel paper and to a separate essay on what Mode-2 deployment looks like once the test case generalizes across the rest of the economy. This is the operational leg — the case that the test laboratory exists, that it is unusually well-suited to the experiment, and that the experiment can be run with the people who are already in the building.
Abstract
A public bargain that holds an industry in place can fracture without anyone in the industry quite noticing it has fractured, and the legitimacy of American health insurance is the live example. The instinct is to read the fracture as a story about who deserves blame. This paper proposes a structurally different frame. The industry's product is friction. The friction is the cost everyone in the system is paying for. Artificial intelligence — the same instrument being deployed elsewhere in the economy to remove jobs — is the specific tool that can refactor the friction instead. Two modes of AI deployment do opposite things inside the same corporate environment. Mode 1 takes a human-performed task and removes the human, output cheaper; it is the mode driving the disruption people fear. Mode 2 takes the existing human and strips the paperwork, coordination, and decision overhead around them, redirecting freed time to the parts that actually produced value — judgment, navigation, relationship, service. Health insurance is the highest-leverage Mode-2 test laboratory in the American economy. This paper argues why, sets out the methodology of an embedded refactor, and names four concrete refactor targets the diagnosis points at.
§1. The Two Modes
A distinction that dissolves an apparent paradox
Almost every public conversation about AI and work treats AI as one thing and asks what it will do to one labor market. The framing is wrong at the root. There are two distinct modes of AI deployment inside a corporate environment, and they have opposite vectors. The mode-collapse failure — treating them as the same tool used twice — is the failure that drives most of the nihilism about AI in the workforce.
Mode 1, wholesale replacement. Take a discrete task currently performed by a human. Train or deploy a model that performs the task at acceptable quality. Remove the human from the loop. Book the headcount reduction. This is the mode that sells easily to a CFO on a quarterly call. It is also the mode that produces the visible labor disruption that drives the present anxiety. It scales by subtraction.
Mode 2, friction removal and redirect. Take the existing human in the existing seat. Identify the paperwork, the routing, the coordination, the lookup, the form-rephrasing, the recapitulation of information already in the system — the activity surrounding the human that does not produce value but consumes most of the time. Refactor that surrounding work with AI. Redirect the freed time to the part of the seat that did produce value — judgment, navigation, relationship, service. Keep the headcount. Keep the cost structure. Improve every output across customer, employee, and shareholder. This is the mode the present discourse routinely fails to name as a distinct option.
The two modes use the same underlying instrument. They are not the same deployment. The disruptive mode wins by default because it is the one that maps cleanly onto a quarterly earnings narrative. The refactor mode requires something the replacement mode does not — a person on the inside with both vantage and temperament to drive it. That requirement is what makes Mode 2 the harder mode to scale and is also the reason it has been underdeployed.
Naming the distinction is the move that resolves the apparent paradox at the heart of every "AI causes the problem; AI solves the problem" argument. It is only paradoxical if the two modes are collapsed. Mode 1 destroys the seat. Mode 2 refactors the seat. Both are AI. The replacement frame is what should be resisted by policy and by corporate strategy. The refactor frame is what can be deployed at scale starting now, in industries where the conditions favor it. The rest of this paper is about the industry where the conditions are most favorable.
§2. Why Health Insurance is the Test Laboratory
A $1.4 trillion friction engine, ready to be refactored
Mode-2 deployment requires four conditions to be present at once in an industry, and they are rare to co-occur. The industry has to manufacture decisions rather than physical things — refactor surfaces are where the decisions live. The friction has to be visible and reciprocally hated by every actor in the system — that is what produces the political will to disrupt the status quo from inside. The regulatory environment has to mandate humans in the loop — that is what prevents the deployment from collapsing into Mode 1 the moment a deployment lead's CFO does the math. And the margin structure has to be too thin to absorb a Mode 1 headcount-cut strategy — that is what makes Mode 2 the only sustainable game in town. Health insurance has all four.
Decisions, not things. The American health-insurance industry processes roughly $1.4 trillion in premium revenue annually and sits atop a $4.5 trillion healthcare-spending stack. It does not manufacture a single physical good. Its entire output is decisions about who pays for what: prior authorization decisions, eligibility decisions, network-coverage decisions, denial decisions, appeal decisions, claims-adjudication decisions, formulary decisions. The whole industry is a friction engine that produces decisions and routes paperwork. That is exactly the production model Mode 2 is shaped to refactor — decisions live in language and structured data, and the overhead surrounding the decision-maker (form preparation, lookup, routing, escalation, rephrasing) is exactly the surface AI is good at.
Reciprocally hated friction. The American Medical Association's annual prior-authorization physician surveys document a remarkably consistent pattern. The overwhelming majority of physicians report that prior-authorization processes cause care delays. A large share report that the delays have led to abandonment of treatment by patients. A meaningful fraction report that prior authorization contributed to a serious adverse event for a patient in their care. Physicians spend something on the order of a full business day per week, per practice, on PA paperwork that does not produce care. The member side of the same transaction tells the inverse story — long phone holds, repeated information requests, denial letters written in language designed for the legal record rather than for human comprehension. The employee side — the customer-service representative, the utilization-management nurse, the prior-auth reviewer — does not enjoy the work either. It is one of the rare industries where every actor at every position in the value chain finds the work miserable and would actively cooperate with a refactor. That is a rare political condition. It is the condition that makes the test possible.
Regulatory cover for humans-in-the-loop. HIPAA, ERISA, state Departments of Insurance, and the new wave of CMS prior-authorization rules all converge on the same baseline: regulated medical-coverage decisions require a human in the loop, with a credentialed clinical or licensed-adjuster role accountable for the final call. This is not an obstacle. It is the load-bearing feature. It means Mode 1 deployment — remove the human, output cheaper — is structurally precluded by the regulatory environment. The only deployment that fits the regulatory shape is Mode 2 — keep the human, refactor the work surrounding the human. The industry's regulatory architecture has already done the work of mandating the deployment posture that the rest of the economy will eventually have to adopt voluntarily.
Margins too thin for Mode 1 to work. Major US health insurers operate at roughly 3–4% net margin. The industry has neither the absolute margin nor the political capital to absorb a wholesale-replacement strategy that brittlizes service delivery in pursuit of a headcount cut. The arithmetic does not support Mode 1 as a sustainable play. The arithmetic does support Mode 2 — keep cost structure flat, dramatically improve unit-output quality, expand margin slowly through reduced rework and reduced appeal volume and reduced member churn. The financial story Mode 2 has to tell is not the headcount-cut story. It is the rework-reduction-and-customer-retention story. That is a story the quarterly call can be taught to tell.
The four conditions are unusually well co-located here. Other industries have one or two. Health insurance has all four at once. That is what makes it the test laboratory.
§3. The Dump Methodology
Embedded research with an AI partner, starting from a grievance log
The methodology this paper proposes for running the experiment is not a consulting engagement. It is something more like a research collaboration between an inside operator and an AI partner. The starting input is a particular kind of artifact: an unstructured, free-associated grievance log produced over roughly a week by a single mid-level insider at a single payer organization. Call it the dump.
The dump is intentionally low-structure. Date-stamped entries, otherwise free format. The instruction to the insider is short. Write down what frustrates, drags, or feels wrong from inside your day. Be granular. Be anecdotal. Interpersonal material is welcome — who pushes back on what, whose name comes up in every escalation. If a thought interrupts another thought, write the interruption down and come back, or do not come back. Order matters, including the order things come up unprompted. The stop condition is felt, not scheduled: when there is really nothing left, the insider says so.
The shape of the artifact is doing specific cognitive work. It is recording three signals an intake form cannot record. The first is order of unprompted surface: what comes up first when no one is asking, the second time, the tenth time. The second is frequency of recurrence: which complaints reappear across days under different surface descriptions. The third is what gets left dangling: the thought that the insider started and did not finish, the workflow whose description ran out before the workflow itself did. All three are visible only in unstructured free-association across time. None of them survive being collapsed into a structured intake form. A consultant cannot extract them. An employee survey cannot extract them. They are produced by the act of dumping.
The AI's job on the receiving end is the part the existing consulting model cannot do at all. It is to absorb a week of unstructured text, hold it in working memory simultaneously, and surface the structural patterns the insider cannot see from inside the building — because the insider is too embedded in the day-to-day to register the recurrence, and because the patterns live in the relations between items the insider already filed under separate mental categories. An AI partner can hold every entry against every other entry, find the recurrences across surface descriptions, identify the dangles, and produce a synthesized map of the operational schema's failure modes with the leverage points marked. The output of the map is then input to a refactor design pass — which refactor targets the insider can actually move from where they sit, which require political cover from above, which require the partnership of a sister function.
The methodology generalizes. Any institution in which a thoughtful insider has the vantage but not the integration capacity is a candidate for the same protocol. The reason it works in health insurance first is that the insider population — mid-managers in payer organizations, clinicians-in-payer roles, utilization-management nurses with operational scope — is unusually large, unusually well-positioned, and unusually well-motivated to participate. The industry has been producing thoughtful insiders frustrated with their own workflows for years. The dump methodology gives them a use for the frustration.
§4. Four Refactor Targets
What Mode-2 deployment looks like, concretely
Four refactor targets emerge from the diagnosis. They are not the complete list. They are the four that present the cleanest combination of high friction surface, high regulatory clarity, and direct line of sight to expanded human capacity at the seat. For each, the test is the same — who keeps their job, what changes for them, what changes for the member or the provider, and what the financial press calls it on the next quarterly call.
1. Prior authorization as a coherence layer, not a removal layer
The Mode 1 reading of prior authorization is to automate the decision and route only edge cases to humans. The Mode 2 reading is the opposite. Keep the human reviewer. Refactor the workflow around them so the reviewer receives a structured, AI-prepared dossier with the clinical evidence summarized, the formulary criteria mapped against the patient's record, the precedents from the payer's own past decisions on similar cases surfaced, and a draft rationale prepared for the reviewer's edit. On the requesting-physician side, the same instrument produces a coherent, plain-language explanation of why the authorization is being processed the way it is, what additional information would move it, and what the appeal pathway looks like if it is denied. The decision stays human. The paperwork around the decision is refactored. The reviewer can process a multiple of the case load at substantially higher accuracy. The physician on the other end gets a coherent explanation instead of a denial letter written for the legal record.
- Who keeps the seat
- Prior-auth reviewer, utilization-management nurse, clinical pharmacist
- What changes
- The seat becomes a judgment seat instead of a paperwork seat
- What the press calls it
- Throughput improvement; appeal-volume reduction; provider-satisfaction recovery
An illustrative day — before and after
Before. Diane is a utilization-management nurse with 47 prior-authorization requests in her queue at 8 a.m. Each case requires her to open the PA submission, hunt through the attached clinical record for the relevant notes, cross-reference the request against the formulary criteria on a separate screen, look up the patient's prior history on a third screen, draft a rationale from a template, and click decide. Average twelve minutes per case. Roughly half her time is spent moving information between screens rather than thinking about the case. She finishes thirty-eight by end of day and falls nine behind. The denial letters she sends out are templated boilerplate written for the legal record, which means the requesting physicians appeal them as a matter of course, which means the appeal queue down the hall gets longer.
After. Same Diane, same queue. Each case opens with an AI-prepared dossier on her screen: the clinical evidence summarized with the specific lines that bear on the request highlighted, the formulary criteria mapped against this patient's history with the matched and unmatched conditions called out, three recent decisions the payer made on similar cases surfaced for comparison, and a draft rationale she can confirm, edit, or override. Average three minutes per case. She clears the queue by lunch, spends the afternoon on the harder cases the AI has flagged as edge calls. The letters her decisions generate are written in plain language a requesting physician can engage with — what was approved, what was not, why, and what additional information would move it. The appeal rate on her caseload drops by roughly a third over the next quarter.
2. Appeals as an AI-drafted and human-checked workflow
Appeals are the largest documented unforced cost in payer operations and the largest documented source of member dissatisfaction. The Mode 1 reading is to auto-deny at higher volume to push the marginal cost of an appeal back onto the appellant. The Mode 2 reading is the opposite. Build the appeal-review workflow on AI-assisted preparation. The reviewer receives the original denial rationale, the appeal submission, the clinical evidence supplied, the relevant policy text and case precedent, and a draft reasoning pass for the reviewer to confirm, edit, or override. The reviewer keeps the decision. The reviewer's throughput rises. The appellant — member or provider — receives a substantive response that engages the substance of the appeal in language a person can read. Appeals close faster, at higher quality, with substantially less rework. The same workflow flags appeals where the underlying initial decision appears to have been incorrect, feeds those back to the prior-auth team as a quality signal, and reduces the rate of subsequent appeals through that channel.
- Who keeps the seat
- Appeals reviewer, medical director, member-services specialist
- What changes
- The appeal becomes a real engagement with the appellant, not a form letter
- What the press calls it
- Appeal-cycle compression; reduced regulatory exposure; net-promoter-score recovery
An illustrative day — before and after
Before. Marcus is an appeals reviewer. The case in front of him is a member whose physical-therapy benefits were denied after the eighth session under criterion 4.2.1, "not medically necessary at extended interval." The appeal letter from the member's PT runs four pages of clinical justification documenting measurable functional gains across sessions five through eight. Marcus is reviewing fourteen appeals today. He reads the denial, skims the appeal, opens the underlying claim, hunts through the policy text to find 4.2.1, decides the PT did not cite the criterion by number in the appeal letter, upholds the original denial, sends a templated response. The member's PT files a Level 2 appeal three weeks later, costing the payer somewhere north of a thousand dollars in administrative overhead, which is overturned at Level 2 because the documented progress in fact does meet 4.2.1. The member missed three weeks of care in the gap.
After. Marcus opens the same case. The AI has prepared the synthesis: the original denial rationale with criterion 4.2.1 highlighted, the appeal letter with its substantive arguments extracted and mapped against the criterion (the PT's functional-gain evidence is flagged as on-point even though 4.2.1 is not cited by number), three structurally similar appeals from the past year (two overturned on comparable facts), and a draft reasoning pass that concludes the appeal has merit. Marcus reads the synthesis, agrees, overrides the original denial, approves six additional sessions. The response letter engages the substance: "Your documentation of functional gains in sessions five through eight is sufficient under 4.2.1; authorization is extended through session fourteen, after which a reassessment will be required." The Level 2 appeal does not get filed because there is nothing to file. The member does not miss three weeks of care. Marcus closes fourteen cases today instead of nine, and the ones he closes do not come back.
3. Member service rep tooling — situation-aware, not script-driven
The member-service representative is the most visible Mode-1 replacement target in the industry. Chatbots are already being deployed against this seat with mixed results that mostly inflame member sentiment. The Mode 2 deployment is structurally different. The representative answers the call. The AI tooling pre-loads the member's situation — recent claims, pending authorizations, eligibility status, the last three interactions across channels, the provider network in their geography, the formulary status of any medication they have asked about — and presents the rep with a synthesized, situational summary in the half-second before the rep speaks. The rep, who is no longer reading from a script and no longer looking up codes, can actually be a person on a call with another person. The interaction time goes down because the rep is not looking things up. The first-call-resolution rate goes up because the rep has the integrated picture. The member calls back less. The rep's seat becomes a service-and-judgment seat instead of a lookup-and-script seat. The rep's job satisfaction is no longer in conflict with the member's.
- Who keeps the seat
- Member-service representative, complex-case manager
- What changes
- The rep becomes a person on a call instead of a navigator of a CRM
- What the press calls it
- Call-time reduction; first-call-resolution lift; member retention improvement
An illustrative call — before and after
Before. Aisha is a member-service rep. The member calling in has just gotten a bill she does not understand — a $340 charge for a doctor visit she thought was covered. Aisha asks for the member's ID, pulls up the CRM, searches recent claims. The CRM shows fourteen claims across the past six months in four tabs. Aisha cannot tell which one the member is asking about. She asks for the date. The member does not remember exactly. Aisha asks the member to hold and toggles between three screens trying to find the right claim. Four minutes of hold music. Aisha finds it — Dr. Patel, April 14, processed correctly, but Dr. Patel is out-of-network for the member's PPO and the member did not realize that going in. Aisha now has to explain network rules to a confused, increasingly upset person who feels she has been tricked. Twenty-two-minute call. The member hangs up angry. She calls back next week with a related question, which is handled by a different rep who has to start from zero.
After. Same call. The AI tooling has pre-loaded — in the half-second between the call connecting and Aisha picking up — the member's profile, the three most recent claims (including the $340 out-of-network charge from Dr. Patel on April 14), the network status of every provider she has seen, four in-network providers within five miles for the same specialty, and a one-line situational read: "Member is likely calling about the April 14 out-of-network charge; network status was not verified before the visit." Aisha picks up. "Hi, I see you may be calling about a bill that came in higher than you expected — was it the April visit with Dr. Patel? Dr. Patel is unfortunately out of your network, which is why that one came in at the full amount. I have four in-network providers near you for the next time, and I can also start a one-time courtesy-review request on this charge while we're on the call." Seven-minute call. The member is not angry — she feels seen. The in-network alternatives go to her by email before the call ends. The courtesy-review request triggers a separate workflow that does not require her to call back.
4. Care navigation as proactive matching, not reactive lookup
Care navigation is the function in which a member who needs specialist care, complex care coordination, behavioral health services, or chronic-condition management is helped to find and access the right provider. In most payer operations today, it is a thin reactive layer — when the member asks, the rep looks up the network directory and reads names off a screen. The Mode 2 deployment turns navigation into proactive matching. The AI tooling holds the member's clinical history, the provider quality data the payer already possesses, the geographic and language and scheduling constraints the member has expressed, and the actual availability across the network. When the member asks, or — more powerfully — before the member asks, the navigation function surfaces a short, specific list of high-fit options with explanations of why. The navigator on the seat reviews, sometimes calls the provider's office to hold a slot, and presents the member with a real choice rather than a directory dump. The function becomes the upside-creating function of the seat, not the friction-managing function. The downstream effects on cost — the right care delivered earlier, the avoidable ER visit avoided, the chronic condition managed before it escalates — are well documented in the value-based-care literature and have been the single largest unrealized efficiency in payer operations for two decades.
- Who keeps the seat
- Care navigator, complex-case manager, behavioral-health coordinator
- What changes
- The seat becomes a proactive matchmaker instead of a reactive directory-reader
- What the press calls it
- Medical-loss-ratio improvement; chronic-care outcomes lift; preventable-utilization reduction
An illustrative case — before and after
Before. Jordan is a care navigator. A new case lands in his queue: a member just discharged from the hospital with new-onset Type 2 diabetes. Jordan's queue this week has eighty other new cases. He calls. The member does not pick up. Voicemail. Two days later, Jordan tries again, reaches the member, who is overwhelmed and does not know what to ask. Jordan reads from the standard new-diagnosis script — "Have you been connected with an endocrinologist? A diabetes educator? Here are some in-network options near you" — and reads seven names from the directory. The member writes none of them down and hangs up confused. She does not follow through. Six months later, an avoidable diabetic-ketoacidosis ER visit costs the payer around eight thousand dollars and the member around three weeks of work missed. Jordan's seat looks like it is doing care navigation. It is in practice doing directory recitation.
After. The same discharge data lands in Jordan's queue. The AI has already done the matching: the member's PCP is in-network and accepts new diabetes cases, the closest endocrinologist with two-week capacity is Dr. Chen (three miles from the member, accepts the plan, Vietnamese-speaking — flagged because the member's preferred language is in the record), a certified diabetes educator at the same practice has Tuesday-morning slots open, and the member's plan covers a continuous glucose monitor with no prior authorization required given the new diagnosis. Jordan calls. "Hi, this is Jordan from your health plan. I wanted to help you set up follow-up care after your hospital stay. I have already found an endocrinologist nearby with Vietnamese-language support, and she has an opening next Wednesday at two. Would that work? While we are talking I can also book a session with their diabetes educator and get you set up with a glucose monitor at no cost to you." The member says yes. Jordan books the appointments live on the call. Twelve-minute call. The member follows through. Six months later, A1C is controlled, no ER visit, the member is retained on the plan. Jordan's seat is now doing what it was always supposed to do.
Each of the four targets has the same financial shape from the quarterly-call perspective: cost structure stays flat, unit-output quality rises substantially, margin expands slowly through reduced rework and reduced churn rather than abruptly through headcount cuts. None of them require a Mode 1 deployment to compete with on cost. All of them are within the regulatory envelope as it currently exists. None of them require the industry to invent a new business model. They require the industry to refactor the work surrounding the seats it already has.
The four targets above describe what most people can already picture when they imagine an ideal Mode-2 deployment. The seat does what the title of the seat was always supposed to mean — Diane reviewing the medicine, Marcus engaging the appeal, Aisha treating the call as a conversation, Jordan booking the appointment. Most of the existing Mode-2 conversation, and the existing payer-tooling vendor landscape, stops here. What gets less attention — and what the rest of this paper turns to — is the same Mode-2 refactor one altitude up, inside the institution itself. The seat-level targets are necessary. They are not sufficient. The harder and less-discussed altitude is the one between the seats — the spaces between departments where the institution currently fights itself, where state goes incoherent across the boundaries, and where the same case can be decided four different ways by four different parts of the same company.
§5. The Institutional Altitude
Where the same Mode-2 refactor operates one layer up — keeping the institution coherent to itself
Mode-2 at the seat removes the friction surrounding one worker doing one case. Mode-2 at the institutional layer is the same refactor flowing up — keeping institutional state coherent across the departments that together produce the member's experience. The two altitudes are not alternatives. They are the same architectural commitment expressed at different scales of the institution, and they hold each other honest. Seat-only deployment produces faster local decisions that contradict each other across departments. Institutional-only deployment produces a coherent corporate posture nobody at any individual seat can actually act on. Both layers running together produce the multidimensional, hierarchically interdependent state architecture that lets the institution behave like a single coherent actor instead of a federation of departments fighting through the member.
A typical major payer is organized into roughly nine functional pillars: product and benefit design; network operations (contracting and provider relations); medical affairs (clinical policy, utilization management, and the medical directorate); claims operations; appeals and grievances; member services; care management; pharmacy; and compliance. Each pillar holds its own state — its own policies, its own decision logs, its own member touchpoints, its own information systems. Most of the misery the member experiences is produced not inside a single pillar but at the boundaries between them, where one pillar's state has drifted out of alignment with another's. The institutional-layer refactor is the work of keeping the state aligned. The four targets below are the cleanest cases.
5. Cross-departmental case coherence
A single case — a single member, a single condition, a single sequence of decisions — currently exists as a different case inside each pillar that touches it. Claims sees one slice. Appeals sees a different slice with different framing. Care management sees a third. Member services sees a fourth. The pillars do not share a coherent representation of the case across boundaries, and so the institution's position on the case is inconsistent at the seams. The institutional refactor is a shared, AI-maintained case-state representation that every touching pillar reads from and writes to. When appeals overturns a denial, the criterion's interpretation in that case updates wherever the criterion lives in any other pillar's decision support; when care management adds a clinical context, the next claims or member-services interaction on the same case sees it; the institution holds one position on the case across the four pillars instead of four positions.
- Who keeps the seat
- Every pillar that touches a case; no department loses headcount, every department gains shared ground truth
- What changes
- The institution stops issuing contradictory positions on the same case to different parties
- What the press calls it
- Operational coherence; rework reduction; reduced grievance escalation; defensible audit posture
An illustrative case — before and after
Before. A member with a complex autoimmune condition has a specialty-drug claim denied in March. She appeals; the appeals reviewer overturns the denial in May, citing the documented response to prior therapies under criterion 3.4.b. The drug is dispensed in June. In August the refill claim arrives. Claims denies again — because the May appeals overturn never landed in claims's decision support, the reasoning was filed in the appeals system the claims team does not query, and the precedent the appeals reviewer set is invisible to the next claims adjudicator. The member calls member services to ask why a drug she just got approved is being denied again. Member services has no access to the appeals reasoning either and tells her to file another appeal. She files another appeal. Six months of bouncing through four pillars on a case the institution already decided once.
After. The May overturn lands in shared case state. The criterion's interpretation in this case is now part of the case's living record, visible to every pillar that touches the case. The August refill claim pre-loads the May precedent and auto-authorizes under the same criterion the appeals reviewer already applied. The member services rep, if the member calls anyway, sees the full institutional history and can explain it in one sentence. The institution holds itself accountable to its own prior decisions instead of asking the member to re-prove the case quarterly. Aggregate effect: the rework volume the payer was paying for at $1,200 a Level-2 appeal — much of it self-inflicted by state incoherence — drops by a measurable fraction.
6. Policy-to-operations propagation
When medical affairs updates clinical policy, the update has to land in every operational pillar that enforces or explains it — utilization management, claims operations, network ops, member services, appeals — and currently it does not, on any predictable timetable. The lag between policy and operations is the most consequential structural friction inside a payer that does not show up in any single seat's queue. Members get told the old answer for months after the new policy is in force. Reviewers apply old criteria to new cases. Appeals reverse decisions made under the prior version. The institution speaks with five voices about the same policy because the five operational pillars are reading from five different snapshots of when the policy was last refreshed. The institutional refactor is a shared policy state every operational pillar reads from and AI-prepared decision support that re-renders the moment the policy state changes.
- Who keeps the seat
- Medical affairs, UM, claims, network ops, member services, appeals; no one loses, everyone reads from the same version
- What changes
- Policy-to-operations lag collapses from quarters to days; the institution speaks with one voice within 48 hours of a policy change
- What the press calls it
- Operational agility; regulatory-update responsiveness; reduced compliance exposure
An illustrative quarter — before and after
Before. Medical affairs updates the clinical policy on a class of GLP-1 medications in early Q1 — narrowing the indications under which the drug is covered. The update memo goes to the UM team, which incorporates it into reviewer checklists over six weeks. Network operations learns about it at the next quarterly contract review, three months in. Member services learns about it when members start calling confused, four months in. Appeals reviewers learn about it from a Level-3 case six months in, while still upholding old denials made under the prior policy. Compliance learns about it from a regulator's letter eight months in. The institution has been operating under five different versions of the same policy for two quarters. Member trust takes a measurable hit. Two state DOIs open inquiries.
After. Medical affairs publishes the update to the shared policy state on a Tuesday morning. By Wednesday morning, the UM reviewer's decision support reflects the new criterion. The contract templates the network team uses pick up the change. The member services situational pre-load includes the new policy as context the moment a member call comes in. The appeals queue is filtered for cases the new policy actually resolves, surfacing them to the appropriate reviewers. The compliance team gets the propagation log as audit evidence. The institution speaks with one voice within 48 hours. The regulators' inquiries do not get filed because the institutional record demonstrates the policy was applied consistently from day one.
7. Claims-to-appeals feedback loop
Appeals exists because claims sometimes makes the wrong call. Currently, when appeals overturns a claims decision, the overturn is a one-off event recorded in the appeals system. It does not feed back into the claims team's criteria interpretation. The same kind of case keeps getting denied at the claims layer and overturned at the appeals layer — the institution pays twice for the same decision and the member pays in delay. The institutional refactor is a closed feedback loop: every appeals overturn is treated as quality signal to the underlying claims decision logic, AI-categorized by the criterion-interpretation it represents, and surfaced to the claims team as a calibration update. Claims decisions in the same category begin tracking the appeals layer's accumulated reasoning. The structural inconsistency between the two pillars closes.
- Who keeps the seat
- Claims adjudicators and appeals reviewers; both seats keep their role, both seats stop fighting through the member
- What changes
- Recurring case patterns the appeals layer is reversing stop being made at the claims layer in the first place
- What the press calls it
- Appeal-rate reduction; first-pass-yield improvement; medical-loss-ratio improvement through reduced administrative cost
An illustrative pattern — before and after
Before. The appeals team notices, informally, that they are reversing roughly 70% of denials in a particular code category — physical therapy for post-surgical knee rehabilitation extending beyond eight sessions. The pattern is visible to anyone who works in appeals long enough to see it. It is not visible to anyone in claims, who continues denying the same cases at the same rate. There is no institutional mechanism to translate "we keep reversing these" into "stop denying these." Twice a year someone proposes a process change. Nothing happens. The institution pays for the same decision twice — once in the claim denial, once in the appeal review — on every case in the pattern. Members miss two to four weeks of care per case in the gap. Providers learn to file every denied case as an appeal as a matter of policy, which inflates the appeals queue further.
After. Every appeals overturn is categorized in shared institutional state by the criterion-interpretation the reviewer applied. When the post-surgical knee-rehab pattern crosses a quality-signal threshold — say, 60% of denials in the category being reversed within 30 days — the claims decision support for that category is updated automatically and the medical director's office is notified that a policy-clarification ratification is required. The next denials in the category begin tracking the appeals layer's reasoning. The institution learns from itself. The Level-1 appeal volume in the category drops 70% over the next two quarters because the institution stops generating the denial in the first place. The appeals team's throughput rises further because they are no longer adjudicating the easy ones; they are doing the genuinely hard cases.
8. Member services as the absorption layer made whole
Member services is the institutional pillar that absorbs every other pillar's friction. When the product team designs a benefit operations cannot enforce coherently, member services takes the calls. When medical affairs updates a policy late, member services explains the inconsistency. When claims denies a case appeals will overturn, member services is the first place the member calls. Currently, the rep handling the call has visibility into the slice their CRM shows them and nothing else. The structural reason member services is the most-burnt-out pillar in the company is not that the work is uniquely hard at the seat; it is that the seat is sitting at the bottom of every other pillar's incoherence. The institutional refactor gives member services the same shared case state every other pillar reads from. The rep on the call sees the full institutional history of the case — what claims decided, what appeals is reviewing, what care management is coordinating, what the policy update was, where the boundary friction is. The rep stops being the absorption layer for invisible structural conflict and starts being the visible coordinator across the pillars.
- Who keeps the seat
- Member services reps, escalation specialists, complex-case advocates; the seat is the same, the vantage is whole-institution instead of single-pillar
- What changes
- The rep becomes the coordinator across the pillars instead of the absorber of inter-pillar friction
- What the press calls it
- Member-services attrition reduction; first-call-resolution lift; net-promoter-score recovery; retention improvement
An illustrative day — before and after
Before. A member with a chronic condition calls member services. She has called member services twice in the past four months. The first call was about a billing question from a denied claim that was then overturned on appeal. The second was about a confusing letter she got after medical affairs updated a coverage policy. She is now calling because she got assigned to a care manager who scheduled her for a service her health plan does not cover. The rep handling the current call sees none of the prior context — not the appeal, not the policy update, not the care manager's scheduling. The rep takes the member's question at face value, calls the care management team to verify the service, gets routed to a voicemail, asks the member to hold, gets back on the line with information that contradicts what the care manager told the member two weeks ago. The member is now angry at the institution as one entity even though no single person in the institution did anything wrong. The rep ends the call burned out from absorbing the structural conflict.
After. Same call. The rep's screen pre-loads the member's full institutional history across pillars: the appeals overturn from four months ago and its reasoning, the policy update that triggered the second call, the care plan the care manager put together two weeks ago including the scheduled service. The rep can see immediately that the scheduled service falls outside the policy update's narrowed criteria — the care manager scheduled it under the pre-update interpretation and has not been re-prompted. The rep can either resolve it on the call (explain the situation, route the case back to care management with the policy context attached, offer an alternative covered service) or coordinate across the pillars without making the member tell the story again. The rep ends the call having coordinated, not having absorbed. The structural friction the rep was carrying is visible now, and visible structural friction can be designed against.
The four institutional targets reinforce the four seat-level targets in a specific way that is worth naming. Each seat-level refactor depends on institutional state coherence to do its work properly. Diane's prior-auth dossier has to reflect the policy update medical affairs made yesterday, not last quarter. Marcus's appeals reasoning has to feed back into claims's decision criteria, not get filed in a separate system the claims team does not search. Aisha's situational pre-load has to include the care plan Jordan put together for the same member three months ago, not stop at the slice the CRM shows her. The institutional layer holds the seat layer honest. Without the institutional layer, the seat refactors run faster against incoherent state, which produces faster local decisions that contradict each other across the institution — a worse outcome than the status quo in some scenarios because the speed amplifies the incoherence. With the institutional layer, the institution stops fighting itself, and the seat refactors get to do what the target-grids say they do. The two altitudes are one refactor.
§6. Coda — Why This Generalizes
The first deployable test case of a broader pattern
This paper has argued that American health insurance is the highest-leverage test laboratory in the economy for the second mode of AI deployment. It is the test laboratory because the four conditions co-occur with unusual cleanness — decisions as production, reciprocally hated friction, regulatory cover for humans-in-the-loop, and margin structure that precludes the alternative. The eight refactor targets that follow from the diagnosis — four at the seat altitude and four at the institutional altitude — are concrete, technically feasible with current tools, and within the regulatory envelope as it exists today. Crucially, they are one refactor, not two: the seat layer and the institutional layer are the same architectural commitment expressed at different scales of the institution, and each holds the other honest.
The argument does not stop at this industry. Health insurance is the first deployable test case of a pattern that will, with adjustments, generalize across every institution whose product is decisions and whose operating cost is the friction surrounding the decision-makers. That pattern includes commercial banking's compliance and lending operations, legal services' document and intake operations, public-sector benefits administration, large-scale procurement, and substantial portions of utility and telecommunications customer operations. The conditions are not all present in every case. Some industries have the friction but not the regulatory cover. Some have the regulatory cover but not the political will. Some have everything except the inside operators with the temperament to drive the refactor. The conditions matter; they are why this paper is specific about which test laboratory to run first.
The deeper claim is that the disruption-versus-redirect frame is a load-bearing distinction the broader public conversation about AI and work has not yet made. The replacement narrative is the default because it is the easiest one to monetize on a quarterly call and the easiest one to fear. The refactor narrative is the harder one to monetize and the harder one to communicate, and it is the one the substrate-asymmetry argument elsewhere in this series suggests is the only one that is structurally stable. Humans hold the substrate that produces judgment, presence, recognition, and stake — the parts of work that cannot be performed by a frozen statistical surface. The architecture that respects the asymmetry keeps the human in the seat and refactors the work around the seat. The architecture that does not respect the asymmetry removes the human and discovers, after the removal, what could not be replaced.
Health insurance is where the experiment can be run now, by the people who are already in the building, against a friction surface that every actor in the system already wants reduced. Run the experiment here. Publish the results. The rest of the economy will read them.
References and source notes
- American Medical Association. Prior Authorization Physician Survey, annual series. Documents physician-reported burden of PA, care-delay rates, treatment-abandonment rates, and adverse-event correlation.
- Centers for Medicare & Medicaid Services. CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F, 2024). Establishes regulatory framework for electronic PA, automated decisioning thresholds, and required human-review baseline for denials.
- National Association of Insurance Commissioners. Health Insurance Industry Analysis Report, annual. Industry-level margin, premium, and operational-cost data.
- Centers for Medicare & Medicaid Services, Office of the Actuary. National Health Expenditure Projections. Total US healthcare-spending stack and payer-share breakdown.
- Kushner, A. (2026). The Substrate Law: Why AI and Humans Will Share the Work Permanently, Not Temporarily. Companion paper in the Citizen Scientist Series. (Forthcoming.)
- Kushner, A. (2026). The Winners Curse and Its Antidote. Citizen Scientist Series. The planning-toolkit frame this paper extends from the strategic room to the corporate room.