The Transition Is the Problem

§1. The Transition Is the Problem

Futurism about artificial intelligence has produced a great many destinations. Some of them are dystopian — the hollowing of knowledge work, the concentration of capital in the hands of whoever owns the compute, the breakdown of the social contract that has held the post-war American economy together. Some of them are utopian — abundance for all, the end of drudgery, the citizen empowered with capacities no court or guild could ever have ratified. The two camps argue about the destination. Neither side spends enough time on the route.

The gap in the discourse is the transition. A post-AI economy in which Mode-2 deployment has become the dominant institutional pattern — the friction layer of the firm absorbed by the tool, the substrate-qualified work of presence and judgment and recognition expanded into the room the friction used to occupy — is a real and plausible end-state. But the country does not get there in one step from where it now stands. It gets there by the inside of every firm performing some specific work that has not yet been named. The boss of a regional manufacturer that runs everything on Excel does not have a coherent picture of what his firm looks like after the pivot, or a method for taking the first step toward it. Neither does his accounts-receivable clerk, who has been quietly afraid of the news for a year. Both are right to feel the gap. Neither has been handed a useful answer.

This is the load-bearing fact that most futurism gets wrong by omission. Naming the destination is the easy part. The destination has been described, in print, in adjacent variations, for at least three years now. What has not been described — what nobody seems to want to write down, because writing it down requires committing to a specific mechanism that can be inspected and disagreed with — is the route. The route is the part that has to fit on the inside of an existing institution that does not have a tech budget, a dedicated AI team, or an executive on its board who has ever used a frontier model for anything more demanding than a sentence rewrite. The route is the part that has to land in the hands of the accounts-receivable clerk by Thursday without requiring her to understand the architecture of a transformer.

The companion paper of this series — the corporate-pivot paper that takes health insurance as its test laboratory — names the institutional half of the answer. Mode 1, the wholesale replacement of the human by the tool, is the deployment pattern that destroys the wage base, generates the political backlash, and produces the disruption everyone is justifiably afraid of. Mode 2, the friction-removal-and-redirect, is the pattern that strips the paperwork around the human and redirects the freed time to the parts of the role the substrate of a chemical body that has lived through time is structurally qualified to perform: judgment under consequence, presence with another person, refusal-integrity, recognition, the carrying of relationships across time. The two modes are not the same tool used twice. They are the same technology pointed in opposite directions. Mode 1 wins by default because it is the easier sell on a quarterly call. Mode 2 wins by design, by somebody on the inside choosing to do the harder work.

That paper answers the institutional question. This paper answers the question one level closer to the ground: what does the worker, the manager, the small-business owner, the public-sector clerk actually do, week by week, to walk Mode 2 into their actual job? What is the mechanism that closes the gap between the executive who has not yet decided which mode to deploy and the clerk who is waiting to learn whether her work will be respected or eliminated? The answer turns out to be specific enough to write down. It is the rest of this paper.

§2. The Substrate-Law Architecture, Applied Institutionally

The argument requires a brief detour into the keel of this series. A separate paper, scheduled for release later in the same arc, sets out what I have come to call the substrate law: the structural claim that the human mind and the language model are two different kinds of substrate doing similar-looking work by different production paths, and that the asymmetry between them is not a current limitation engineering will dissolve but a feature of what the two things are. The human mind is a chemical substrate running on a body that has lived through time. The language model is an electronic substrate running on a frozen statistical surface that was trained on the corpus and then stabilized at inference. The two paths produce outputs that, in many narrow tasks, are functionally indistinguishable. The mechanism underneath them is not.

The consequence — and this is the part that matters for the worker's question — is that the two substrates are structurally specialized for different classes of work. The frozen statistical surface absorbs work that is reducible to pattern-completion: synthesis of a large corpus, generation of a draft from a prompt, calculation, exhaustive option enumeration, lateral retrieval, schema-fitting, search of a structured space, the production of code against a specification. This is an enormous domain, much larger than skeptics admit. The chemical substrate, on the other hand, retains work that requires something the frozen surface does not have: presence under consequence, judgment that carries its own stake, refusal-integrity, the recognition of another person by a creature that has itself been recognized, the carrying of relationships across time, the navigation of an institution by a body that has spent years inside it. This too is an enormous domain, larger than enthusiasts admit. The two overlap at the edges, which is where most of the public debate gets bogged down. They are distinct at the center.

The institutional implication of the substrate law is the architectural claim of this paper. The post-Mode-2 worker is not, in the relevant sense, doing less. The post-Mode-2 worker is doing more of the structurally human work. The friction that used to dilute the substrate-qualified parts of the role — the form-filling, the cross-referencing, the looking up of the policy number, the drafting of the rote letter, the searching of the prior case files — was never the reason the role existed. The role existed because somebody in the loop had to make a judgment, hold a conversation, register a constituent's actual situation, or carry the institutional memory of why a thing had been done the way it had been done. The friction had accumulated, over the previous generation, as the cost of doing the substrate-qualified work inside an institution that had no other way to capture the supporting information. Mode 2 removes the friction. It does not remove the role. The role, freed of the friction, becomes more of what it was supposed to be in the first place.

This is the reframe the worker has been waiting for and has not been given. The accounts-receivable clerk who has been afraid of the news for a year is afraid because the public conversation, dominated by the Mode-1 framing, has implied that her job is the friction. Her job is not the friction. Her job is the judgment underneath the friction — the recognition that this customer's invoice is late because the customer just lost a parent, and the appropriate response is a phone call, not a dunning letter. The friction was the part she dispatched in order to get to the judgment. Mode 2 takes the friction and leaves the judgment. The Mode-1 framing tells her she will be fired. The Mode-2 framing, correctly understood, tells her she will finally be allowed to do her actual job.

The institutional architecture that respects the substrate law is therefore not optional decoration on top of a Mode-2 deployment. It is the deployment. Constitutional AI, the human-in-the-loop primitives, the human attestation layer, the refusal-of-final-judgment by the model on consequential decisions — these are not policy preferences imposed from outside. They are the engineering consequences of the law that defines what the two substrates are. A deployment that respects the law produces a Mode-2 outcome by construction. A deployment that ignores the law produces Mode 1 — because the only way to ignore the law in practice is to pretend the model can do the substrate-qualified work, and the pretense, played out in policy, becomes the wholesale replacement.

§3. AI as Its Own Onramp

The remaining question is the practical one. If Mode 2 is the architecturally correct deployment, and the substrate law is the structural reason it is correct, what makes Mode 2 actually happen inside a firm that has none of the apparatus — no consultancy engagement, no dedicated AI team, no executive who has used a frontier tool for more than a sentence rewrite, no budget — that previous industrial transitions required to land their target technology inside an institution?

The historical analogy that makes the question vivid is the previous industrial revolution the United States is widely understood to have missed. The diffusion of internet-of-things sensors, integrated enterprise software, just-in-time inventory tracking, and the broader package of cyber-physical integration that academic writers have called the Fourth Industrial Revolution landed unevenly across the developed world over the 2010s. China — by virtue of a centralized authority capable of telling a manufacturer to install the sensors and a labor force conditioned to accept the instruction — landed it more thoroughly inside its industrial base than the United States did. The United States, lacking that centralized lever, did not land it. A great many American small and mid-sized firms still run their operations, in 2026, on the same Excel workbooks they were using in 2009. The Fourth Industrial Revolution required a hand on the throttle that the American system was constitutionally unable to provide. It produced exactly the outcome that its preconditions predicted.

The crucial fact about the AI industrial revolution is that it does not require the same hand. The tool that does the work in a Mode-2 deployment is the same tool that explains the work, costs the upgrade, walks the implementation in, trains the operator, debugs the integration, and answers the question about why the integration broke three weeks later. The tool is its own onramp. This is the first industrial revolution in the history of the republic whose technology of diffusion is identical to the technology of work. The previous revolutions all required a separate apparatus — the consulting engagement, the trade journal, the trade school, the credentialed implementer — to bring the technology into the institution. The AI revolution does not. The accounts-receivable clerk can install the upgrade by asking the upgrade what to do.

This is the load-bearing claim of the paper, and it is worth being concrete about what it looks like in practice. Three institutions, three different refactor surfaces, three different worker archetypes. None of the three has a McKinsey engagement. None of the three has an AI department. All three have a person inside who has decided to walk Mode 2 in by hand.

The manufacturer running on Excel

A regional manufacturer of, say, custom architectural metalwork. The firm has thirty-five employees, the founder still walks the floor, and the sales pipeline lives in a shared Excel workbook that has been organically extended over twelve years and is now a quiet horror to anyone who has to touch it. The accounts-receivable workflow involves cross-referencing the shop floor's job-completion records against a separate accounting workbook, manually reconciling job numbers that have drifted across three numbering conventions, and producing invoices that are routinely late because the reconciliation takes a junior bookkeeper an entire afternoon. Mode 1 would be to fire the bookkeeper and put a chatbot on it. Mode 2 is to sit the bookkeeper down with a frontier model and ask her to describe the workflow in her own words for ninety minutes.

The model reads the workbooks she names, asks her clarifying questions about the numbering drift, and writes — over the course of an afternoon, in dialog with her — a small script that ingests the shop-floor records and the accounting workbook, reconciles the numbering, and outputs a stub invoice for her review. The bookkeeper's role is unchanged in title but transformed in content. She no longer spends afternoons doing the reconciliation. She spends afternoons calling the three customers whose invoices are unusually large, building the relationship that is the actual point of a small B2B firm's accounts-receivable function. The financial-press story on the quarterly call, were the firm public, is that the founder retained his bookkeeper, watched days-sales-outstanding drop by twenty percent, and learned that he had been understaffing the relationship-building part of the function by an order of magnitude. He did not get a consultant. He did not buy software. He asked his bookkeeper to ask the model what to do.

The county clerk's office

A mid-sized county in the rural South, population roughly eighty thousand, operates its land records, vital records, and voter registration through a clerk's office of fourteen full-time staff, most of whom have been there for decades. The office's largest pain point is records retrieval: a title attorney, an heir, or a journalist arrives in person or by phone with a request that requires somebody to walk to the basement, retrieve a microfilmed roll from 1973, read it on the one functioning reader, and either photocopy or transcribe the entry in question. The work is unevenly distributed across the staff. Two clerks bear most of it. One of them is six months from retirement. The Mode-1 framing would have a vendor digitize the rolls and replace the two clerks with a search index. The Mode-2 framing — and this is the version the office's elected clerk actually walked in — is to digitize the rolls in dialog with the two clerks themselves, asking the model to describe the file structure, identify the inconsistencies in how the 1973 entries were indexed versus the 1991 entries, and produce a unified search layer that the clerks themselves operate and continue to extend as new records come in.

The clerks' roles, post-deployment, include the substrate-qualified work the office had never had time to do under the old workflow: actually helping the family that has come in to settle a contested estate, walking them through what the entries mean, advising them on what to do next, making the public service the office had always been supposed to be. The clerks did not lose their jobs. The office did not hire a vendor. The elected clerk paid a contract programmer for two weeks and produced an outcome that the previous decade's digitization vendors had quoted at eighteen months and four hundred thousand dollars. The county commission noticed. The next county over has begun asking how it was done.

The regional bank

A mid-Atlantic regional bank with roughly four billion dollars in assets — call it forty branches, four hundred employees, a compliance department of twelve. The bank's pain point is its Bank Secrecy Act and Anti-Money-Laundering compliance review workflow. Every transaction over a threshold triggers a manual review. The reviewers are reading the same eight templates of customer activity for the thousandth time, flagging the unusual ones for escalation. The flagging accuracy is, in the unhappy way these things are quantified by the regulator, about as good as it has ever been — which is to say, mediocre — and the manual workload is, in the equally unhappy way these things are felt by the staff, soul-destroying.

The Mode-1 framing replaces the twelve compliance reviewers with a model that scores every transaction and surfaces the top half-percent for human review. The framing fails — and it fails specifically because of the substrate law. The compliance reviewer's substrate-qualified contribution is not the templated scan. It is the judgment-with-stake that says: this looks like the template, but my body has spent seven years in this branch and I know this customer's family, and the transaction that statistically fits the templated pattern is, in this case, a daughter helping her father transfer assets before a hip replacement. The model, lacking the substrate, would have escalated the family transaction. The reviewer, possessing the substrate, would not.

The Mode-2 framing reverses the workflow: the model does the templated scan, surfaces every potentially-flagged transaction with a paragraph of context, and asks the reviewer to ratify or override. The reviewer, freed of the rote scan, now reviews eight hundred transactions a day instead of eighty, and applies actual judgment to the ones that need it. The compliance department's headcount holds. The flagging accuracy, on a generation of regulator audits, improves measurably. The financial-press story on the quarterly call is that the bank's compliance reviewers became the best in their peer group — not by being replaced by an AI but by being given an AI to dispatch the templated work the reviewers had never been the right substrate for in the first place.

What the three cases share

In none of the three cases was the upgrade landed by a consultancy. In none of the three cases did the firm have an AI department or an executive with a frontier-model résumé. In each case the upgrade was landed by a person inside the firm asking the tool what to do, in plain English, with the tool willing and able to explain what it was doing and why. This is the self-onboard property that previous industrial revolutions did not have. It is what makes the American AI transition, in principle, performable at the speed of decision rather than at the speed of consultancy. It is the mechanism by which Mode 2 can become the dominant deployment pattern without a centralized authority forcing it from above. The tool teaches its own use. The tool is the onramp.

A note on what is not being claimed. The claim is not that the self-onboard property removes the need for judgment about which Mode-2 refactor to attempt. Choosing the right refactor is the substrate-qualified part of the deployment — the part that requires the person inside the firm to know which workflow is rotten, which colleague is six months from retirement, which customer's relationship has been understaffed. The tool does not make that choice. The tool, presented with the choice, executes it. The reframe is not that the technology removes the human judgment from the deployment. The reframe is that the technology removes the consultancy from between the human judgment and the deployment. The judgment is the worker's. The execution is the model's. The institution does not need a third party in between.

§4. Distribution of Benefit

Every industrial revolution has produced disruption. None has been absorbed without dislocation. The question is never whether to allow the new technology — that question is, in any meaningfully competitive economy, already answered before it is asked — but whether the benefits of the new technology flow back to the ordinary people whose lives the disruption affects, or get captured at the top of the curve by whoever owned the apparatus that produced the technology. The First Industrial Revolution produced both the modern factory and the modern factory worker, and the slow construction over two generations of the institutions — the labor union, the public school, the social-insurance state — that turned the worker's productivity into the worker's standard of living. The Second produced both the modern corporation and the modern middle class. The Third, the digital revolution, produced both the modern platform and the modern productivity wedge between owner and worker, and the institutions that would have closed the wedge are still under construction. The Fourth, by the conventional count, mostly skipped the United States. The Fifth — the one this paper is about — is where the question of whether benefits flow back to ordinary people becomes a live political decision the country is presently making.

Mode 2 is the deployment pattern that preserves the wage base, which is what makes the question soluble in the affirmative. The wage base is what funds the consumer demand that turns enterprise productivity into national prosperity. Mode 1, taken at scale, eats the wage base. Mode 2, taken at scale, preserves it — and, because Mode 2 redirects the freed time toward the substrate-qualified parts of the role, increases the productivity of the wage base on the same headcount. The resulting equilibrium is the one in which the technology pays for itself in margin without the country having to pay for the technology in disruption. The tax base survives. The consumer market survives. The political coalition that funds the rest of the institutional infrastructure — the schools, the pensions, the deferred maintenance on the highways and the grid — survives.

The load-bearing condition on the equilibrium is not the technology. The technology produces both modes equally. The load-bearing condition is the architectural commitment that the technology will be deployed with the human in the loop, with the model as the dispatcher of friction rather than the substitute for judgment, with the constitutional rail in place that prevents the model from being asked to make the consequential decision the substrate law says only the human can make. The technical name for the rail is constitutional AI. The political name for the rail is the deliberate commitment of the country to the Mode-2 pattern as the deployment standard the rest of the institutional architecture is built around. The two names refer to the same thing. They are the same engineering decision considered at two different scales.

This is what The American Compact, the larger arc of which this paper is a part, has been trying to articulate as a deliberate national choice. The country is not condemned to Mode 1 by the technology. The country is not delivered to Mode 2 by the technology either. The choice between the modes is the political decision the country is making, in distributed form, one institution at a time, this year and next. The Compact is the proposal that the choice be made deliberately, in public, with the architecture explicit, and with the wage base and the substrate law named as the load-bearing constraints the policy is built around. The individual off-ramp this work has elsewhere called the citizen-scientist stack is the parallel commitment at the level of the person — frontier tools in individual hands so the displaced become independently productive, not merely re-employed. The institutional off-ramp this paper has just sketched is the parallel commitment at the level of the firm. Both are required. Neither is sufficient alone. Together they describe a transition that is neither the dystopia the loudest critics fear nor the abundance the loudest enthusiasts promise. They describe an actual route from where the country now stands to the work-sharing world the substrate law makes possible.

§5. Coda

The closing observation belongs to the work that frames the rest of the series. The substrate-asymmetry paper, in its current form, ends with the observation that the country does not really have a choice — that things cannot stay as they are, that the technology has already arrived and is already reshaping the institutions whether the architecture is named or not, and that the honest posture is engagement-with-eyes-open rather than disengagement-as-precaution. The same closing belongs here.

Disengagement is not safer than engagement. It is a different bet, with the same downside and a worse expected payoff. The architecture this paper has described — Mode-2 deployment, substrate-respecting workflow, the self-onboard property of the tool itself, the distribution of benefit through the preserved wage base — is not guaranteed by inaction. The default outcome of inaction is the wholesale-replacement mode that wins by being the easier sell on a quarterly call, and the wholesale-replacement mode produces the disruption everybody is justifiably afraid of. The choice between the two modes is being made, distributed across millions of conversations between executives and clerks and consultants and founders, this year and next. The paper's argument is that the right choice is available and inexpensive, that the technology itself is the mechanism that carries the choice into the institution, and that the missing piece is not capability but commitment.

It could, the closing observation continues, be extremely awesome. The work-sharing world the substrate law describes — calculation and synthesis absorbed by the tool, presence and judgment retained by the human, the institutions of the country running on a wage base whose productivity has been freed by the friction-removal that the previous generation of work had no instrument to perform — is not a fantasy. It is the equilibrium the architecture produces when the architecture is deployed deliberately. The country does not have to be told to do it. The country has to decide to.