AI Won’t Replace You. Someone Using AI Will Take Your Influence.

The comfortable fear and the expensive one

You have been trained to fear a machine. The threat, as it is usually described, is external, total and impersonal: a wave of automation that arrives one quarter, eliminates a category of work, and recedes. Entire consulting practices exist to model it.

Notice how comfortable that fear actually is. It is collective. A threat faced by everyone demands no individual accountability from anyone. You can file it under industry trends, feel appropriately concerned, and return to your inbox with your self-image intact. Nothing about it requires you to change what you did this morning.

The actual event is smaller, quieter and far more dangerous to you specifically. It is not the elimination of your function. It is the redistribution of authority inside it.

Scope is migrating. Trust is consolidating. Visibility is accruing. And it is not moving toward the most senior or the most tenured person in the room. It is moving toward whoever converted the new tooling into leverage before the rest of the room understood there was a race to lose.

Here is the part that should hold your attention. Your output has not declined. Your standing has. The distance between those two facts is the most underpriced source of systemic exposure in your professional life, and it is exposure precisely because it is invisible. It registers on no dashboard, in no performance review, and in no conversation you will be invited to, right up until the moment it has already resolved against you.

Why your organization is not going to warn you

There is an instinct, when you first see this clearly, to assume the organization will eventually correct it. That someone in leadership will notice the asymmetry and intervene on the side of fairness.

Understand the institution's indifference correctly. The organization is not protecting you from this transfer because the organization profits from it.

A workforce that quietly out-competes itself for relevance is a workforce that raises its own output without a single directive from above. No change management programme. No budget line. No uncomfortable town hall. Your relative decline is not a malfunction the system intends to correct. From the system's vantage point, it is a feature operating exactly as designed.

There is a second reason the misdirection persists, and it is more human. As long as you are scanning the horizon for an external event, you are not watching the desk beside you. And the colleague at that desk has every incentive to keep your gaze fixed on the horizon.

The macro narrative is not merely inaccurate. It is conveniently inaccurate. It points your vigilance outward, toward a threat you cannot influence, and away from the single transfer you still could.

This pattern is older than the corporation

Strip the technology away and watch only what a new tool does to the people standing near it. The pattern predates the corporation, the contract, and the written word that first carried it.

Mesopotamia. The arrival of cuneiform did not eliminate the labour of administration. It transferred administrative authority to a new class, the scribes, who could encode and retrieve what others could only hold in memory. The oral memorizers were not dismissed. They were routed around. Their function remained intact. Their power was gone.

The modern version requires no translation. The mid-level director who masters prompt-chaining performs the identical manoeuvre on the vice president who refuses to touch the interface. The title remains. The relevance does not. Your organization has already decided who its scribes are. It has simply not informed you yet, because it does not currently require you to know.

Egypt. The Egyptians preserved the resistance itself. In the dialogue Plato attributes to Socrates, the god Thoth offers King Thamus the gift of writing, and the king refuses it on principle. It will weaken memory, he argues. It is a crutch. It is not true wisdom.

He was not wrong about the costs. He was simply overtaken by everyone who accepted the tool regardless. This is the executive who declines a new system on grounds of rigour and craft, and discovers too late that conviction and incapacity produce an identical signal on a performance review. Nobody reads the principled abstention. They read the output gap.

Norse. The Norse made the price explicit. Odin acquires the runes, a technology of knowledge and command, only by hanging nine nights on the world-tree, wounded and starving, paying in advance for a literacy no one would hand him freely.

The instruction is not mystical. Fluency in a new tool is always purchased through deliberate, uncomfortable acquisition, and those unwilling to pay the cost end up quietly governed by those who did. The analyst who spends ninety private days becoming the most capable operator of a system is making Odin's wager. The colleague waiting for a mandate is not.

Three civilizations, one law. The tool never asks permission to reassign authority. It rewards the literate and routes around the rest.

The physics of the transfer

Return to the present and examine how the reallocation actually happens, because it does not occur the way most professionals imagine.

There is no announcement. There is no decisive meeting in which your standing is formally reassigned. The transfer is incremental, and it remains nearly invisible to its subject until it is structurally complete.

The sequence is precise.

Stage one: compression. A peer collapses a task that once defined your value, a model, a brief, a synthesis, from days into hours. The first-order effect is trivial. One deliverable arrives early. Nobody remarks on it.

Stage two: absorption. This is the effect that ends careers. With their capacity freed, that peer begins absorbing adjacent scope. They take on the analysis that used to be yours, not by seizing it, but by being available when it needs doing.

Stage three: the relationship. The stakeholder who used to route questions through you begins routing them through the person who answers fastest. This is rarely a decision. It is a drift.

Stage four: the seat. Scope, trust and the seat at the table have consolidated elsewhere. By the time the pattern surfaces on an org chart, it is a description of something that finished months ago.

A case that will feel familiar

Consider a product manager who has always authored the team's strategy memos. A peer begins generating equivalent memos in a fraction of the time.

The peer's documents are not superior. They are merely ambient. They are produced in such volume that three arrive for every one the manager completes. Leadership announces no preference and holds no meeting. It simply begins forwarding the peer's memos, scheduling around the peer's availability, and treating the peer's framing as the default starting point for discussion.

The manager's craft is completely intact. The manager's centrality is gone.

This is corporate asymmetry in its purest form: identical competence, divergent leverage, and a reallocation that no one will ever commit to writing.

Four markers that diagnose your own exposure

You can diagnose this with more precision than most people apply to their own careers. Four markers.

One: speed. Is someone in your function shipping the output that defines your value measurably faster than you, at no loss of quality?

Two: scope. Is scope migrating toward them through reorganizations and staffing decisions you were not consulted on?

Three: standard. Are senior leaders beginning to cite their work as the reference standard, the version others are asked to match?

Four: position. Has your relative position eroded while your absolute output held perfectly steady?

The fourth is the most diagnostic, and it is the one people consistently misread. If the fourth marker is true, you do not have a performance problem. You are being out-leveraged. That is a different problem with a different solution, and confusing the two is fatal, because the instinctive response to a performance problem is to work harder inside the existing frame, which is precisely the move that accelerates the decline.

Why four in five people walk into this with open eyes

Roughly four in five knowledge workers meet a mandated tool with active or passive resistance. From inside that resistance, it does not feel like avoidance. It feels like prudence, discernment, a refusal to chase novelty.

But the structure you are operating inside is an adoption equilibrium, and it follows the same arms-race logic that governs any competitive system. Once a critical mass of your peers adopts a leverage technology, your unchanged output is no longer neutral.

Standing still is not safety. In a relative system, holding constant while the baseline advances is decline that has been disguised as stability, and the disguise is precisely what makes it lethal. You will not feel yourself falling. You will feel exactly as competent as you did last year, because you are. The accountability vacuum is total, because no one is obligated to inform you that the ground beneath your position has moved.

The trap folded inside the trap

There is a second-order problem here that deserves its own attention, because it is where capable, conscientious people lose the most ground.

The resistant majority does not experience its position as falling. It experiences the early adopters as cutting corners. It reads the asymmetry as a temporary aberration that rigour will eventually correct. Quality will out. The shortcuts will be exposed.

It will not happen. The equilibrium does not reward the most careful operator. It rewards the operator whose careful work also compounds, and compounding requires the very tool the careful operator declined.

This is the structural cruelty of the current moment. The diligence that once protected your standing now accelerates its erosion, because it is diligence being spent inside a frame the market has already abandoned.

The countermeasure: the Inscription Protocol

If this is an equilibrium, it has a dominant strategy. That strategy is not to work harder inside the old frame. It is to win the literacy race deliberately, before it is run for you.

Call the countermeasure the Inscription Protocol: the disciplined practice of becoming the institution's record-keeper for the work that matters most. It runs in four moves, executed in order.

One: locate the keystone task

Not every responsibility you carry. The single highest-leverage, most-observed output of your function, the one whose quality leadership uses to judge the whole.

Capital preservation begins with identifying which asset is actually load-bearing. Most professionals spread their improvement effort evenly across their responsibilities, which feels responsible and produces almost no change in how they are perceived. Find the one output that leadership treats as a proxy for your competence, and treat everything else as secondary for the duration of this exercise.

Two: achieve asymmetric fluency

Do not become broadly conversant with the tool. Become demonstrably, measurably the best operator of it on that one task.

Breadth is how the resistant majority reassures itself. Attending the webinar, skimming the newsletter, holding an opinion about the vendor landscape. Depth on the load-bearing task is what actually reassigns power. The difference between the two is roughly the difference between knowing what a spreadsheet is and being the person everyone's model routes through.

Three: make the literacy legible

This is the move the entire protocol turns on, and the one where capable people fail most often. They build the advantage and never inscribe it.

An advantage no one in authority can observe is not an advantage. It is a private hobby.

The execution here is tactical and specific. Do not send a deck that gets buried in an inbox. Inscribe your work directly into the primary analytics infrastructure your leadership opens every Monday morning. Make your output the live instrument they consult to run the business.

When the number the organization governs by is the number your system produces, your literacy has become institutional infrastructure. And infrastructure is not easily routed around.

Four: enter before the mandate

The instant adoption is required, the advantage it once conferred has already been distributed to those who moved first.

The mandate is not the start of the race. It is the public posting of its results.

What this looks like when it works

Consider a finance director who identifies the firm's quarterly variance analysis as her keystone task.

Rather than resisting the new tooling, she spends sixty days building an automated pipeline that produces the analysis in real time. That is asymmetric fluency: not general familiarity, but unmatched depth on the one output that matters most.

She then embeds that pipeline into the executive operations dashboard the chief financial officer opens daily, so the entire leadership team now governs by her instrument. That is legibility.

Within two quarters, every cross-functional planning conversation routes through her surface. Scope consolidates toward her. And a more senior colleague who dismissed the tool finds himself in an uncomfortable position: requesting data from the very team he used to manage.

No title changed hands. The standing did. That is structural arbitrage captured by the operator who inscribed first.

Note carefully what she did not do. She did not request a promotion. She did not lobby for headcount. She did not defend her relevance in a meeting.

Argument is the instrument of those without leverage. She made the argument structurally unnecessary by becoming the surface through which the institution sees itself.

The Inscription Protocol does not persuade an organization to value you. It rearranges the organization so that it cannot function without consulting your work. Institutions do not relocate authority by sentiment. They relocate it around dependency.

What the protocol secures, and what it does not

So the immediate move is settled. Locate the task. Own the tool. Make it visible. Move before you are told to.

But it is worth being precise about what this buys you, because overstating it is its own kind of exposure.

The protocol secures capability. It does not secure position.

The same flattening that lets you out-leverage a senior colleague will eventually let a junior one out-leverage you. The same visibility that builds your standing also marks you the moment the institution decides it requires a target. Literacy wins the current round. It grants no immunity from the next.

Which means the deeper question was never whether you can master the tool. The question is how a person holds high standing inside an institution without becoming the obvious thing it sacrifices when the cycle turns.

That is a different discipline entirely, and it is the one that protects everything the Inscription Protocol earns.

Three objections worth answering

If you have read this far with some resistance, the resistance is probably one of three specific shapes. Each deserves a direct answer rather than a dismissal.

"This rewards superficiality over real expertise."

Sometimes, briefly. But the objection misreads what is being rewarded. The finance director in the example above did not replace judgement with automation. She applied automation to the mechanical portion of a task so that her judgement could be applied more often, to more decisions, closer to the moment they were made. Her expertise was the input. The tooling changed the distribution, not the substance.

The version of this objection that actually holds is narrower: in the short window before an organization learns to evaluate AI-assisted output, volume can temporarily masquerade as value. That window exists. It also closes. Betting your career on it closing before your standing erodes is a wager on institutional discernment, and institutional discernment is historically slow.

"My work is too senior or too judgement-heavy for this to apply."

This is Thamus, almost word for word. It is also the position most likely to be held by people with the most to lose, because seniority is where the gap between perceived indispensability and actual dependency is widest.

The test is simple. Identify the output your seniority is actually measured through. If any part of that output is retrieval, synthesis, formatting, comparison or first-draft construction, then a meaningful fraction of what currently signals your seniority is mechanical, and the mechanical fraction is exactly what compresses. The judgement remains yours. The visible artifact that carries the judgement to other people is what changes hands.

"If I adopt aggressively, I will be the one who looks like they are cutting corners."

This risk is real and it is manageable, which is precisely what the third move of the protocol addresses. Advantage that appears suddenly and without explanation reads as suspicious. Advantage that is inscribed into infrastructure the organization depends on reads as contribution.

The difference is legibility and sequence. Do not surface a finished capability as a surprise. Surface it as a service other people can use, positioned where they already look.

What to do in the first thirty days

The protocol is a strategy, not a to-do list, but strategies that never reach a calendar do not compound. A reasonable first month looks like this.

Week one: name the keystone task. Write down the single output leadership uses to judge your function. If you cannot identify it in one sentence, ask a peer in an adjacent team which of your deliverables they actually see. Perception lives outside your own reporting line.

Weeks two and three: build depth, privately. Take the mechanical layer of that one task and rebuild it with current tooling. Expect the first attempts to be worse than your manual version. That is the cost Odin pays on the tree, and there is no route around it. Measure the result against your own baseline rather than against someone else's demo.

Week four: inscribe. Find the surface your leadership already opens without being asked. A dashboard, a recurring review document, a standing agenda. Put your output there, in the format they already read, and make it live rather than periodic.

The failure mode is stopping after week three. A private capability that produces no institutional dependency is a hobby with a professional justification attached.