Reality Is Malleable

Essays on systems, society, and constructed futures.

Deflation Activism

A short subtitle or one-sentence framing line goes here.

Abstract of a neighborhood of apartments on a hill top

Preamble

Prices are not shaped only by materials, labor, and scarcity. A surprising amount of what people pay comes from friction: search costs, paperwork, opaque pricing, weak coordination, administrative drag, and intermediaries that exist because systems are hard to navigate. This essay argues that AI agents could begin stripping away those hidden costs by helping ordinary people compare, coordinate, negotiate, switch, and route around inefficiency at scale. The result would not be utopia or full post-scarcity. It would be something quieter and more practical: a consumer-side pressure system that pushes more of the economy closer to its real cost structure.

TL;DR

The Hidden Tax of Friction

The price of something rarely reflects only the cost of making it. Materials and labor are part of the story, but they are not the whole story. A large share of the price people pay comes from everything surrounding the transaction itself.

Someone has to search for suppliers. Contracts must be written and enforced. Risk has to be managed. Payments need to be processed. Records must be kept. Entire industries exist to coordinate these steps.

Each layer adds cost.

Over time those layers accumulate into something that shapes the modern economy as much as production does. One way to think about it is as a tax embedded inside almost every purchase. The tax does not appear on receipts, yet it is present in the final price of groceries, insurance policies, airline tickets, and medical bills.

This collection of hidden costs can be described as the friction economy.

Friction appears in many forms. A consumer searching for the best price spends time and effort. A company managing compliance with regulations hires staff and consultants. Brokers connect buyers and sellers who would otherwise struggle to find each other. Legal review reduces the risk of disputes. Administrative workers move information between systems that do not communicate easily.

None of these functions produce the underlying good or service. They make the transaction possible.

Technological change occasionally removes large portions of this friction. When railroads spread across the nineteenth century, the cost of transporting goods over land fell dramatically. Container shipping later reduced the labor and coordination required to load cargo ships. The internet exposed information that had once been controlled by specialists and intermediaries.

Each shift changed the structure of prices across entire industries.

A similar shift may be beginning to emerge from a different direction. Artificial intelligence agents are gradually becoming capable of performing tasks that revolve around coordination. They can search markets continuously, compare options, track contracts, and handle routine administrative work.

If these capabilities mature and spread widely, the cost of coordinating economic activity could fall sharply.

For most of modern history the tools required to optimize complex systems were available only to large organizations. Corporations built departments devoted to procurement, logistics, and data analysis because individuals could not easily perform those functions on their own.

AI agents may begin to change that balance. When individuals gain access to tools that search, negotiate, and coordinate on their behalf, the optimization power that once belonged to institutions becomes widely distributed.

That shift raises an interesting possibility.

Consumers might begin using these tools deliberately to squeeze inefficiencies out of the systems they interact with. The goal would not be protest or regulation. It would simply be the systematic reduction of unnecessary costs.

One could think of this emerging behavior as a form of deflation activism.

What Deflation Activism Means

The phrase deflation activism may sound confrontational at first. It is easy to imagine protests, campaigns, or regulatory battles aimed at forcing prices downward. The idea discussed here is quieter than that.

Deflation activism would operate through tools rather than political pressure.

The central idea is simple. If individuals gain access to capable AI agents, those agents can begin performing the kinds of optimization tasks that were once limited to corporations and large institutions. They can watch markets continuously, compare options, monitor contracts, and look for inefficiencies that raise prices.

The goal is straightforward. Remove unnecessary costs wherever they appear.

In practice an agent acting on behalf of a consumer might spend its time doing small but persistent forms of economic housekeeping. It could monitor grocery prices across several stores and recommend the cheapest option. It could watch internet and phone plans and suggest switching providers when prices change. It might examine insurance renewals, search for alternatives, and handle much of the paperwork involved in changing policies.

Administrative work offers another example. Many routine interactions with companies and governments involve forms, documentation, and repeated communication between systems that do not easily exchange information. An agent could handle much of this background work automatically. The consumer would simply approve the outcome.

Agents could also help individuals coordinate with each other. A single household has little leverage when negotiating prices for services such as electricity, internet access, or insurance. A large group of households acting together has more influence. Software agents could organize that coordination quietly in the background.

Another area involves transparency. Pricing often varies widely because buyers cannot easily see the full range of available options. Agents that continuously scan markets could expose those differences and route purchases toward the most efficient providers.

Each of these actions seems small when viewed individually. Together they represent something larger.

Instead of demanding that markets become fairer or more efficient, consumers would begin using software tools that gradually push markets in that direction. Prices fall not because institutions are forced to change, but because inefficiencies become harder to maintain.

In that sense deflation activism would be less a political movement than a shift in how people interact with the economic systems around them.

Why AI Agents Change the Game

For most of modern economic life, individuals have faced a basic problem. Even when better deals or more efficient arrangements existed, finding and acting on them took too much effort. The market may have contained cheaper options, better contract terms, or ways to coordinate with others, but the cost of discovering and using those options was often too high to bother.

That is one reason inefficiency persists. It is not always protected by law or by monopoly power. Sometimes it survives simply because people are busy.

Large firms have long had an advantage here. They can hire analysts, negotiators, lawyers, procurement specialists, and software teams to keep searching for savings. They can compare suppliers, renegotiate terms, model risk, and optimize operations on a scale that ordinary households cannot match. In that sense, optimization itself has been a kind of luxury good.

AI agents may begin to change that by lowering three constraints at once.

The first is the cost of intelligence. Work that once required a trained specialist can increasingly be done by software, at least for routine cases. That does not mean every expert disappears or that every judgment can be automated. It does mean that the first pass of analysis becomes much cheaper. A household that would never hire a consultant to compare insurance plans or audit utility bills could have an agent do something similar in the background.

The second change is persistence. Human beings search in bursts. They compare prices when a bill comes due, when a contract expires, or when a purchase becomes urgent. Agents do not have to work that way. They can monitor prices, policies, wait times, inventory levels, and service terms continuously. That matters because many inefficiencies survive only because no one is watching all the time.

The third change is coordination. This may be the most important one. A single consumer has limited leverage. A thousand consumers acting together can negotiate better terms. Historically, organizing that kind of cooperation required time, trust, and administration. AI agents could make it far easier to form temporary bargaining groups, compare shared needs, and act in concert without requiring everyone to spend their evenings in meetings or email chains.

Taken together, these shifts do more than make shopping easier. They change the economic position of the individual.

A consumer with a capable agent is no longer just a passive participant moving through markets designed by others. That consumer becomes part of a network that can search, compare, coordinate, and respond at scale. The isolated buyer starts to look more like a node in a distributed system.

That is why AI agents could matter in a way earlier consumer software did not. A coupon app or price comparison site helps at the margins. An agent that combines analysis, persistence, and coordination begins to approach the strategic behavior of an institution.

Once that happens, a different kind of market pressure becomes possible.

The Economics Behind the Idea

This line of argument has an older economic foundation than it may first appear. In 1937, Ronald Coase published a paper called “The Nature of the Firm,” asking a deceptively simple question: if markets are so effective, why do firms exist at all?

His answer was that using the open market is not free. It carries transaction costs. Someone has to find suppliers, compare terms, negotiate agreements, monitor performance, and enforce contracts. When those costs are high enough, it becomes cheaper to pull activity inside an organization rather than constantly bargain with outsiders.

That basic logic explains a great deal about the modern economy. Firms are not only production machines. They are coordination machines. A company hires employees, builds departments, and develops internal procedures partly because it is easier than recreating a miniature market for every task.

Seen in that light, many corporate structures are really responses to coordination difficulty.

This is where AI agents become economically interesting. Their most important effect may not be that they make people more informed or more productive in a general sense. It may be that they make coordination itself much cheaper.

An agent can compare vendors, draft routine communications, monitor compliance requirements, keep track of deadlines, and search for alternatives continuously. It can also do these things for many people at once. Tasks that once demanded staff time, institutional knowledge, or specialized departments begin to look more like software functions.

If transaction costs fall far enough, some activities that were once easiest to manage inside firms may become easier to coordinate through networks instead. A consumer group might negotiate insurance collectively without needing a traditional broker. A small business might join an agent-managed purchasing pool rather than maintaining a large procurement function. A neighborhood energy network might allocate demand and storage through software rather than relying on a more centralized model.

This does not mean firms disappear. Large organizations still provide capital, absorb risk, and manage long-term projects in ways decentralized groups often cannot. But some of their functions may weaken. The more those functions depend on routine coordination, information processing, or administrative management, the more exposed they are.

That possibility matters because a large share of price in modern economies comes from exactly those activities. When coordination grows cheaper, some of the institutional layers built to manage it no longer justify their cost. They shrink, specialize, or vanish.

Deflation activism follows naturally from that shift. It is what happens when individuals and small groups start using low-cost coordination tools to route around expensive structures that once seemed unavoidable. The politics of the idea may be novel, but the economic mechanism is familiar. Lower transaction costs change the shape of markets.

Historical Friction Collapses

The basic pattern described here is not new. Economic history contains repeated examples of technologies that did not merely make production more efficient. They made coordination easier. When that happened, prices often fell, old intermediaries lost power, and markets reorganized themselves.

The Sears Roebuck catalog is one of the clearest early cases. Before mail-order retail reached scale, many rural Americans depended on local general stores with limited competition and opaque pricing. The catalog changed that by making prices visible across distance and giving households a direct route to a wider market. It did not eliminate retail, but it weakened the grip of local middlemen and narrowed the room for arbitrary markups.

Container shipping provides a more industrial example. Before standard containers, moving goods through ports required labor-intensive loading, unloading, sorting, and repacking. Each stage introduced delay, error, and cost. Standardization changed the entire system. Once goods could move in predictable units across ships, trucks, and rail, the coordination burden of global trade fell sharply. Shipping costs dropped, transit became more reliable, and entire supply chains were rebuilt around the new logic.

Travel aggregators did something similar in a different domain. For a long time, travel agents held an advantage because they controlled access to schedules, fares, and booking systems that ordinary consumers could not easily navigate. Once websites exposed that information directly, much of the value of the intermediary disappeared. Consumers could compare options for themselves, and commissions that once seemed normal began to look like friction.

Craigslist pushed the same logic into classified advertising. Newspapers had long enjoyed a lucrative business matching buyers and sellers for jobs, apartments, used goods, and local services. The revenue looked stable until a low-cost digital platform made peer-to-peer listings far easier. What collapsed was not the desire to advertise. It was the old fee structure attached to the act of matching local demand and supply.

The microchip revolution widened this pattern even further. As computing power became cheaper and more abundant, it reduced the cost of design, planning, inventory management, communication, and analysis across countless industries. The falling price of electronics was only the most visible result. Less visible was the way cheap computation improved the coordination of production itself, allowing firms to operate with more precision and less waste.

Fracking is a less obvious example, but it points in the same direction. New extraction techniques mattered on the engineering side, yet part of their impact came from changing the speed and flexibility with which energy producers could respond to market conditions. Supply increased, bottlenecks loosened, and prices came under pressure. The details differ from mail-order catalogs or booking sites, but the general mechanism is familiar. Better coordination reshaped the cost structure.

These examples vary in scale and sector, but they suggest a recurring sequence. A technology lowers the cost of searching, matching, standardizing, or synchronizing activity. Friction that had once supported high margins or entrenched intermediaries begins to weaken. Prices fall, sometimes gradually and sometimes with surprising speed.

That sequence matters here because AI agents appear to target the same layer of the economy. They do not simply promise better production. They promise cheaper coordination. History suggests that when coordination costs drop far enough, the effects do not stay confined to back-office operations. They show up in prices.

High-Leverage Arenas for Deflation Activism

The broad theory matters, but the real test of the idea is practical. Where would agent-driven deflation actually show up first, and where would it matter most?

The obvious targets are sectors where coordination costs make up a large share of the final price. These are areas where the underlying good or service may not be especially scarce, yet the path between provider and customer is cluttered with search costs, paperwork, opaque pricing, fragmented demand, and layers of intermediation. When those layers thin out, prices can move more than people expect.

Retail is the simplest place to start. Price transparency networks could give consumers a far more aggressive version of comparison shopping. Instead of checking a few stores manually, agents could continuously scan prices across chains, online marketplaces, local vendors, warehouse clubs, and any other viable competitors for a given product, including time-sensitive promotions and substitute goods that meet the same need. Once enough consumers are using tools like that, opaque pricing becomes harder to sustain. Retailers that rely on customer inertia, weak comparison habits, or geographic ignorance lose some of that advantage. The likely result is not perfect uniformity, but less room for price discrimination and fat margins built on obscurity.

From there, the next step is coordination among buyers themselves. A single household has little bargaining power with utilities, internet providers, insurers, travel platforms, or even grocery distributors. A large bloc of households is in a different position. Agents could make it much easier to form temporary bargaining groups around shared needs, while also surveying the full field of viable providers rather than the handful most visible to human consumers. That kind of collective consumer bargaining would not require a formal organization in the old sense. The software could identify overlapping demand, compare competing offers across the market, aggregate it, and negotiate on behalf of the participants. Large firms have long enjoyed those scale advantages. Agents could extend a version of that logic to ordinary people.

Insurance provides an especially revealing example because it is already a business built on statistics, pooling, and administration. In principle, much of what an insurer does is coordinate risk. Agents could help organize small mutual-style insurance pools among participants with similar risk profiles, while also handling actuarial modeling, claims oversight, and the purchase of reinsurance for catastrophic losses after surveying viable reinsurance and coverage options across the market. Traditional insurers would still matter in some areas, especially where capital depth and regulatory compliance are essential, but parts of the system could shift from centralized institutions to agent-managed networks. A portion of the premium that currently goes to overhead might simply stop being necessary.

Healthcare may be one of the most politically sensitive arenas, but it is also one of the most friction-heavy. In many systems, especially the American one, a remarkable share of healthcare spending has less to do with treatment itself than with billing, coding, appeals, network negotiations, and administrative processing. Agents could help patients compare providers, prices, drug options, and treatment pathways across all viable competitors within a region or network. They could challenge billing errors, automate insurance appeals, and coordinate group negotiations for drugs or routine services. The effect would not be to solve healthcare in some sweeping sense. It would be to attack the administrative thicket that inflates so much of the cost.

Legal and compliance work sits in a similar position. A large amount of economic activity carries legal overhead because trust is expensive and disputes are costly. Contracts must be drafted, vendors vetted, requirements checked, and conflicts mediated. Much of that work is necessary, but much of it is also repetitive. Agents could standardize routine agreements, compare legal service providers and compliance options where alternatives exist, verify counterparties, monitor compliance conditions, and handle lower-level dispute resolution. Law would not disappear. The routine coordination layer around law could become far cheaper.

The same is true of bureaucracy more broadly. Administrative systems absorb an extraordinary amount of labor because forms must be filled out, documents assembled, deadlines tracked, and information translated between institutions that do not share clean interfaces. Agents could take over much of this work. Permit applications, insurance claims, tax filings, billing disputes, and standard compliance reporting all fall into this category. In settings where citizens or firms must choose among service providers, filing channels, contractors, or processing options, agents could also compare the viable alternatives rather than defaulting to the path of least resistance. None of this sounds glamorous, but it is exactly the sort of hidden effort that silently raises costs across the economy.

Real estate is another sector where friction is often mistaken for necessity. Finding property, comparing listings, evaluating prices, preparing documentation, coordinating inspections, and managing negotiations all involve genuine work, but much of that work is informational and procedural rather than deeply scarce. AI Agents could automate a meaningful share of it. They could search across all viable listings and competing properties, model local valuations from public data, compare financing options, assemble documentation, and coordinate routine transaction steps. That would not eliminate every broker or lawyer, but it could put pressure on commissions and fees that have long persisted partly because the process is cumbersome.

Energy and utilities offer a more technical but potentially important frontier. As households acquire batteries, electric vehicles, smart appliances, solar panels, and dynamic pricing plans, energy consumption starts to look less like a simple monthly bill and more like a continuous coordination problem. Agents could decide when to charge a vehicle, when to draw from storage, when to sell excess power back to the grid, and when to buy electricity during low or negative price windows after comparing all viable suppliers, tariffs, and exchange opportunities available to the household or neighborhood. At a larger scale, local networks could coordinate these decisions collectively. In that world, utilities still matter, but more as infrastructure operators than as one-way pricing authorities.

Small businesses face a parallel challenge on the procurement side. Large firms can negotiate favorable supplier contracts because they buy at scale and invest in purchasing systems. Smaller firms usually pay more because demand is fragmented. Agents could help form dynamic procurement networks that pool orders across many small businesses, compare viable suppliers across a much wider field, predict near-term demand, and negotiate on that basis. That would reduce distributor margins, smooth inventory flows, and narrow the gap between corporate purchasing power and everyone else.

Some of the earliest consumer wins, though, may come from less dramatic places. Subscription and fee elimination is a good example. A surprising amount of modern household spending leaks away through recurring charges that people forget to review. Agents could monitor these expenses continuously, compare all viable alternatives in the market, cancel unused services, renegotiate contracts, flag price increases, and route users toward cheaper or free substitutes. Each individual saving may be modest. Across millions of users, the effect becomes significant.

The longest-horizon possibility may lie in open design and local manufacturing. Here the deflationary effect comes from making knowledge more portable and production more distributed. Agents could help generate, refine, compare, and distribute open-source designs for household goods, replacement parts, tools, and modular products suited to local fabrication, while also evaluating viable local production and repair options. Combined with CNC machines, 3D printers, repair shops, and small-scale manufacturing networks, this could reduce dependence on traditional branded supply chains for certain classes of goods. The key shift would be that design intelligence becomes easier to share than industrial coordination once was.

What ties all of these arenas together is not simply automation. It is the reduction of hidden coordination costs in sectors where those costs have accumulated for decades. Some of these changes would save consumers money directly. Others would weaken the institutional layers that currently extract value by controlling information, limiting visibility into competitors, aggregating demand, or managing complexity. In each case, the pattern is the same. Agents do not need to abolish a sector to force it to become leaner. They only need to make its friction more visible, and easier to route around.

Adversarial Optimization

Part of what makes deflation activism interesting is that it does not depend on institutional goodwill. Consumers do not need firms to volunteer lower prices, simplify their fee structures, or make switching easier. They only need tools capable of navigating around the obstacles.

That logic resembles what Cory Doctorow has described as adversarial interoperability. The basic idea is straightforward. A new tool does not wait for an incumbent system to invite it in. It finds ways to connect, translate, scrape, compare, or otherwise interact with that system in the service of the user. In practice, this often means restoring options that were technically possible but commercially discouraged.

AI agents fit naturally into that pattern.

An agent does not need a retailer’s blessing to compare prices across competitors. It does not need an insurer’s enthusiasm to evaluate policy alternatives or challenge a renewal quote. It does not need a subscription platform’s cooperation to notice that a household is paying for three overlapping services when one would do. It can simply observe, compare, and act in the user’s interest.

That is where the adversarial element enters. The goal is not sabotage, and it is not necessarily confrontation in the dramatic sense. The conflict arises because many profitable business models depend on friction remaining in place. Opaque pricing, difficult cancellation flows, weak interoperability, confusing contracts, and high switching costs all help preserve margins. A capable agent treats those features as problems to be solved.

In that sense, optimization itself becomes adversarial.

The consumer is no longer limited to the pathways firms prefer. If a company builds its pricing model around the assumption that most customers will not compare all viable competitors, the agent undermines that assumption. If a service relies on user inertia, the agent chips away at it. If an intermediary survives because the underlying process is tedious, the agent turns tedium into software.

This point matters because it clarifies why deflation activism may produce resistance even when it remains entirely peaceful and ordinary in appearance. No protest is required. No dramatic political campaign is necessary. A household simply deploys a tool that works harder and more broadly than the institutions around it would prefer.

From the consumer’s point of view, this looks like ordinary self-interest aided by better software. From the point of view of firms that depend on hidden friction, it looks like a direct attack on their margin structure.

That tension is likely to define much of the next phase.

The Swarm Effect

One consumer using an agent to shave a few dollars off a monthly bill is not especially dramatic. It looks like a better budgeting tool. Even a few thousand users doing this would register more as a niche software trend than an economic shift.

The picture changes once agent use becomes widespread.

What matters is not only what a single agent can do, but what happens when millions of them are searching, comparing, negotiating, and switching at the same time. At that scale, the behavior stops looking like isolated consumer choice and starts to function as a distributed pressure system acting across the economy.

A retailer can ignore some comparison shoppers. It has a harder time ignoring a market in which a large share of customers are continuously checking all viable competitors. An insurer can count on a certain amount of customer inertia under normal conditions. That becomes less reliable once renewal quotes are being evaluated automatically and alternatives are surfaced immediately. A subscription business can benefit from forgetfulness and fragmentation for a while. It has less room to do so when agents are watching for duplication, price creep, and cancellation opportunities across millions of accounts.

The important point is that the pressure is cumulative. Each consumer may be acting independently, but the aggregate effect is coordinated even when no central authority is directing it. Firms start to encounter a different kind of market environment, one in which hidden fees are exposed faster, inflated quotes are punished more quickly, and complexity is less effective as a moat.

That kind of environment changes business behavior.

Services become easier to compare because comparison becomes unavoidable. Prices become clearer because obscurity stops working as well. Margins narrow where they were previously supported by customer passivity, weak search, or limited visibility into competitors. Providers simplify contracts, cancellation flows, and product structures because friction that once protected them now drives users away more efficiently than before.

In other words, the market begins to reorganize around the expectations of agent-mediated consumers.

This is why deflation activism may matter even if most agents are not especially dramatic on their own. The force comes from scale. A swarm does not need every unit to be extraordinary. It only needs enough of them acting in ways that reinforce each other.

Under those conditions, deflation does not arrive as a decree or a campaign promise. It emerges from the interaction between broad search, rapid switching, persistent comparison, and distributed coordination. What looks like a collection of private optimizations begins to behave like a public economic force.

Resistance from the Friction Economy

If deflation activism begins to work, resistance will follow.

That should not be surprising. A large share of modern profit does not come from producing fundamentally scarce goods. It comes from controlling access, managing complexity, limiting comparison, and preserving the small frictions that keep customers from moving too easily. Any tool that reduces those frictions threatens established revenue models.

Some industries will respond by trying to preserve opacity. Others will try to redefine optimization itself as improper behavior.

One obvious front is technical access. Companies can close APIs, restrict data portability, redesign websites to frustrate scraping, and deploy increasingly aggressive bot-detection systems. What had once been a relatively open consumer environment may start to resemble a defended perimeter. The practical message is simple: compare less, switch less, see less.

Another front is legal. Firms may push for anti-bot rules, stricter terms of service, or regulations that limit automated negotiation and machine-mediated access to pricing, contracts, or inventory. These arguments will often be framed in terms of security, fraud prevention, or system stability. Sometimes those concerns will be real. Just as often, they will serve as a more respectable language for protecting margins built on friction.

There is also a subtler form of resistance that may prove just as important. Institutions can redesign their services so that comparison becomes harder even without explicit exclusion. Pricing can be bundled, categories made less comparable, cancellation flows made more conditional, and offer structures tailored to confuse broad market scanning. In effect, businesses may begin optimizing against the agents in the same way agents optimize against them.

That is where the conflict starts to look like an arms race.

Consumers and the tools acting on their behalf will seek wider visibility, easier switching, cleaner interoperability, and broader access to viable competitors. Firms that benefit from weak comparison will seek narrower visibility, stickier workflows, and more barriers between users and alternatives. Neither side needs to announce a war for the pattern to emerge. It follows directly from the incentives.

This matters because it shows that deflation activism is not just a technical possibility. It is also a political and institutional challenge. Lowering transaction costs may be economically efficient, but efficiency does not always align with the interests of the organizations currently extracting value from complexity.

The friction economy will not disappear quietly. In many sectors, it will fight to remain frictional.

Structural Risks and Failure Modes

Any serious argument for deflation activism has to admit that lower friction is not automatically an uncomplicated good. Reducing coordination costs can make markets leaner, but it can also produce new distortions, new concentrations of power, and new kinds of vulnerability.

One obvious issue is the Jevons paradox. Efficiency gains do not always translate into lower total spending. Sometimes they simply make it easier to consume more. If agents save households money on groceries, insurance, or subscriptions, that money may not remain saved in any meaningful sense. It may flow into other categories instead. The result could be lower costs in friction-heavy sectors paired with rising demand, and rising prices, in scarcer ones. Land, prestige services, scarce housing, and other supply-constrained goods may absorb some of the gains.

There is also the risk of algorithmic herding. If large numbers of agents are trained on similar objectives and draw from similar datasets, they may begin steering consumers toward the same providers at the same time. That could create a strange outcome in which tools designed to increase competition end up concentrating demand. A supermarket, insurer, or platform that wins enough agent-driven traffic could gain a stronger market position than it otherwise would have had. In the worst case, optimization produces a new path toward local monopoly.

The digital divide presents another problem. Tools that reduce friction do not automatically benefit everyone equally. Early gains are likely to accrue first to people with the time, trust, literacy, and infrastructure needed to use these systems well. Households that are older, poorer, less connected, or less comfortable with automation may be left outside the optimization layer. If that happens, deflation activism becomes uneven. Some people would gain access to a more efficient economy, while others remain stuck paying the old friction tax.

Privacy and security risks are also hard to ignore. Agents that compare providers, negotiate contracts, optimize household spending, and coordinate across many users will inevitably handle sensitive information. Spending patterns, insurance details, health data, purchasing plans, and business operations all become valuable targets. A network of highly capable consumer agents could easily become a surveillance asset, a hacking target, or both. Even where the economics are favorable, trust may prove fragile.

Regulation could slow or block progress in some of the sectors where the potential savings are largest. Healthcare, insurance, finance, energy, and housing are not simply inefficient markets waiting to be cleaned up. They are dense institutional territories with licensing rules, compliance burdens, liability frameworks, and political interests layered on top of them. In such sectors, the friction is not always accidental. Sometimes it is embedded in law. Agents may still help at the margins, but the path to major cost reduction could be much slower than the technology alone would suggest.

Then there is the question of quality. If agents are trained too narrowly on price, they may drive users toward providers that are cheaper for bad reasons. Lower durability, weaker labor standards, hidden environmental costs, and minimal customer support can all look efficient in a narrow optimization frame. A friction-heavy system can certainly protect waste, but some margins also support safety, resilience, and long-term quality. An agent that strips away cost without distinguishing between waste and value may leave users with goods and services that are cheaper in the short term and worse in the long term.

These risks do not invalidate the broader argument. They do, however, make the picture more complicated. Deflation activism is not a magic solvent that dissolves inefficiency and leaves everything else intact. It shifts incentives, redistributes leverage, and changes the terrain on which firms and consumers meet. Some of those changes will be genuinely beneficial. Others will require governance, design choices, and social judgment.

Lower friction can be a powerful economic force. It is not the same thing as wisdom.

Labor Disruption: The Hollow Middle

The promise of deflation activism is lower cost. The social question is who absorbs the loss when those costs disappear.

A great many jobs in modern economies do not exist to make physical goods or deliver irreplaceable expertise. They exist to move information between systems, reconcile mismatched rules, manage paperwork, guide people through institutional complexity, or stand in the gap between buyers and sellers. These roles are often treated as normal features of economic life. In many cases they are really the labor expression of friction.

That is why the transition could be so disruptive.

Administrative staff, brokers, compliance workers, claims processors, support representatives, schedulers, and a wide range of back-office personnel all sit inside the coordination layer that agents are most likely to compress. The issue is not that every such worker is doing meaningless labor. Many of these people are solving real problems inside systems that were built in clumsy, fragmented ways. The problem is that if software becomes good at handling those routine coordination tasks, the number of humans required to do them may fall sharply.

This creates what might be called a hollow middle. At the top end, highly specialized experts may remain valuable because judgment, liability, trust, and complex problem solving are still hard to automate. At the bottom end, many physical and in-person jobs remain tied to the stubborn realities of bodies, places, and materials. The pressure lands most heavily in the middle, where large numbers of workers perform structured cognitive and administrative tasks that are important, but also repetitive.

That kind of labor shock would not necessarily appear all at once. In some sectors it may begin as slower hiring, smaller departments, and growing expectations that fewer people can handle the same load with better software. In others it may show up as a sudden collapse in the value of intermediary roles that had long survived on search costs, weak comparison, or procedural complexity.

The economic logic is straightforward. If a system once required ten people to process forms, compare providers, handle renewals, monitor compliance, or coordinate routine decisions, and agents can reduce that need to three, the missing seven do not vanish from the social ledger just because the system became more efficient.

This is one of the central tensions in the whole idea. The same friction that inflates prices also supports livelihoods. Remove it, and consumers may benefit while large categories of workers lose bargaining power, income, or entire career paths.

That does not mean the friction should be preserved for its own sake. Economies cannot sensibly justify every layer of waste by pointing to the jobs attached to it. But it does mean the transition cannot be discussed honestly as a simple win. If deflation activism succeeds, one of its first visible effects may be a broad employment shock concentrated in the administrative and intermediary middle of the economy.

The long-term outcome could still be positive. Lower costs free up household income. Simpler systems reduce daily stress. New forms of work may emerge around agent oversight, trust verification, exception handling, local service provision, and areas where human judgment still matters. Even so, there is no guarantee that these new roles will appear quickly enough, or in the same places, for the people displaced by the old ones.

A cheaper economy is not automatically an easier society to live in during the transition.

Cultural Implications

If deflation activism develops into something real, its effects will not be limited to prices. It would also reflect a cultural shift in how people think about economic life.

For a long time, optimization has mostly belonged to institutions. Corporations optimize supply chains, staffing models, procurement systems, logistics routes, and pricing strategies. Consumers, by contrast, are usually expected to navigate whatever environment those institutions create. They can comparison shop, read reviews, clip coupons, or switch providers once in a while, but they rarely act with anything like institutional sophistication.

AI agents could begin to change that relationship. They would not simply help consumers make better choices within existing systems. They would allow ordinary people to participate in the optimization process itself.

That gives the idea a certain family resemblance to earlier movements that treated capability as something worth distributing widely rather than hoarding. Open source software did this with code. Free knowledge movements did it with information. Maker culture did it with tools, fabrication, and technical literacy. In each case, part of the cultural appeal came from the sense that ordinary people should have greater access to capacities that had once been limited to large organizations or specialists.

Deflation activism could express a similar impulse in economic form.

The underlying claim would be that efficiency should not belong only to firms seeking higher margins. It can also be a public-facing good. Lower search costs, lower switching costs, and lower coordination costs make everyday life easier to manage. They reduce the amount of time and attention people must spend fighting through systems that were never designed with their convenience in mind. In that sense, optimization stops being only a corporate strategy and becomes part of the civic environment.

That shift matters because it changes the role of the consumer. Instead of being treated mainly as a target for pricing, retention, and behavioral nudges, the consumer starts to look more like an active node in a broader network of economic coordination. People are no longer only choosing among offers placed in front of them. They are using tools that compare, negotiate, and route around unnecessary complexity on their behalf.

There is also a subtler cultural effect. Once people become accustomed to seeing friction as something contingent rather than natural, their expectations begin to change. Fees that once seemed inevitable may start to look arbitrary. Administrative burdens that were long accepted as part of adult life may begin to feel more like bad design. Entire sectors could come under pressure simply because people no longer assume that cumbersome processes deserve to exist forever.

That does not mean every form of complexity disappears, or that every market becomes clean and rational. It does mean that a broader share of the population may start thinking like systems critics. The question shifts from “Why is this expensive?” to “What, exactly, is making this expensive, and can it be routed around?”

That is more than a technical adjustment. It is a change in economic consciousness.

Toward Post-Scarcity Lite

If the central argument of this essay holds, the larger significance of deflation activism is not just that it saves people money here and there. It is that it pushes more of the economy closer to its underlying cost structure.

A surprising share of what people pay for is not the thing itself, but the difficulty of reaching it. Search costs, coordination costs, administrative drag, weak interoperability, and institutional complexity all create distance between production cost and final price. When those layers shrink, more goods and services begin to move closer to what they actually cost to provide.

That does not create a fully post-scarcity world. Scarcity does not vanish because software gets better. Land remains finite. Skilled labor remains finite. Time remains finite. Some resources will always be constrained by physics, geography, or social preference. Even so, a society can move a long way toward greater abundance without abolishing scarcity in the absolute sense.

The important shift is that many forms of scarcity turn out to be partly artificial. They are not produced by a genuine shortage of matter or energy. They are produced by bad coordination, weak information, fragmented bargaining power, administrative bloat, and systems that evolved under older technological limits. If AI agents begin stripping away those inherited frictions, a portion of what once looked structurally expensive may start to look unnecessarily expensive.

That is where the connection to a broader abundance framework comes in.

Automation pushes down the labor cost of producing goods and services. Cheap manufacturing and better fabrication tools push down the cost of making physical things. AI-assisted design lowers the cost of planning, iteration, and customization. Deflation activism works on a different layer. It pushes down the cost of coordinating all of this. Production becomes cheaper, and access becomes cheaper at the same time.

When both pressures operate together, the result can be more significant than either one alone. Goods approach their production cost more closely. Essential services become less burdened by layers of administration and intermediation. Everyday life becomes less dominated by tollbooths disguised as complexity.

That is why a phrase like post-scarcity lite may be useful. It avoids the fantasy of total abundance while still naming a real directional change. A society does not need infinite goods to feel radically more abundant than the one before it. It only needs the cost of many important things to fall far enough that access stops being a constant source of pressure.

In that sense, deflation activism may be one of the quieter routes toward abundance economics. It does not promise a utopia. It simply suggests that as coordination becomes cheaper, more of the economy may finally behave as though it belongs to an age of high capability rather than one still organized around inherited friction.

The Quiet Revolution

Economic change does not always announce itself with a slogan or a crisis. Sometimes it arrives through a series of small adjustments that seem ordinary in isolation. A bill gets lowered. A subscription disappears. A renewal quote is challenged. A group of households negotiates better terms. A process that once required a broker, a form, and three phone calls is handled quietly in the background by software.

None of this looks revolutionary at first.

That is precisely the point.

If AI agents become capable of persistent search, broad market comparison across viable competitors, collective coordination, and routine negotiation, they will not need to overthrow markets to change them. They will work within markets while making many of their older frictions harder to sustain. Hidden fees become easier to detect. Intermediaries face pressure where their value rests mainly on obscurity or procedural burden. Administrative costs that once seemed unavoidable begin to look more like leftovers from an earlier technological era.

As those pressures spread, sectors do not disappear all at once. They reorganize. Services become easier to compare. Contracts become simpler because complexity stops paying as well. Margins narrow where they were being protected by customer inertia, weak visibility into alternatives, or the difficulty of coordinating buyers. The result is not the end of economic life as we know it. It is a market environment that behaves differently because consumers are no longer operating alone.

That shift may prove uncomfortable for many institutions. It may also be uneven, politically contested, and socially disruptive. Some jobs will be squeezed. Some firms will fight to preserve the friction on which they depend. Some efficiencies will create new problems rather than cleanly solving old ones. Even so, the underlying direction remains important. Lower coordination costs change what kinds of structures an economy can support.

That is why deflation activism is worth taking seriously. It points to a future in which ordinary people gain access to tools that were once available only to large organizations. They do not need to stage a revolt against the economy. They only need to optimize their way through it more effectively than before.

The revolution, if it comes, will be quiet.

It will arrive through efficiency.

- Iarmhar

March 24, 2026

Follow on X to be notified when new essays are posted.

Related

The Goblin Economy
A vision of AI-enabled microbusinesses turning global commerce into a hypercompetitive auction house.

What is Money When Nothing Costs Anything?
An essay on how money changes when survival goods become cheap and value becomes expressive.

Attention Collapse and the Rise of Cognitive Minimalism
A diagnosis of attention collapse and a case for clarity as the next form of wealth.