AI Alleviated Us from Toil. Now Define Your Value.

Every time a wave of technological change arrives, someone reaches for a historical analogy.

The Industrial Revolution. The assembly line. The internet. The PC. Each one, we're told, looked terrifying and turned out fine. Jobs disappeared and new ones appeared. The economy absorbed the shock. People adapted. Progress won.

It's a comforting story. But this time, I don't think it's enough.

Not because AI is going to destroy everything — I don't believe that. But because the comfort of historical analogy lets us off the hook. It suggests the outcome is inevitable, that the arc bends toward okay on its own, that we just need to be patient and trust the process.

That's not a plan. That's a posture.

This moment is different enough that it deserves to be treated as new. And if we treat it as new, we have to ask a different question — not will we survive this? but what are we choosing to build?

What's Actually Different This Time

Previous technological revolutions automated physical labor or specific, bounded cognitive tasks. The loom replaced hand-weaving. The calculator replaced the abacus. The spreadsheet replaced rooms full of accountants doing manual calculations.

Each time, the automation had edges. It did one kind of thing very well, and humans moved to the work adjacent to it — the creative work, the judgment work, the relationship work that the machine couldn't reach.

AI is different because the adjacency is collapsing.

It's not automating a task. It's automating a layer — the execution layer across nearly every knowledge domain simultaneously. Writing, analysis, coding, design, research, customer service, legal drafting, financial modeling. Not perfectly, not without human oversight, but well enough and fast enough to fundamentally change the calculus of how organizations staff and operate.

But here's the part that doesn't get said clearly enough — the part that changes everything about how we should respond.

Best Practice Is Now the Floor

AI can now perform at the level of best practice across almost any knowledge domain.

Not mediocre output. Not a rough draft that needs heavy editing. Best practice. The synthesis of what's known, applied correctly, produced faster than any human can match and at a fraction of the cost.

This is genuinely new. And it has a consequence that most upskilling conversations completely miss:

Best practice is no longer a human differentiator. It's the baseline.

For decades, being good at your job largely meant knowing and applying best practices reliably — the right framework for the situation, the right structure for the document, the right methodology for the analysis. That was valuable because it was scarce. It took years to develop. It separated the capable from the novice.

AI just democratized all of it. The novice with a good prompt can now produce output that looks like the work of a ten-year veteran. Best practice is no longer scarce. It's a commodity.

Which means humans who are competing primarily on best practice — who built their careers on knowing the right way to do things — are competing on the wrong terrain.

And here's the turn: what AI cannot do is everything that lives above best practice.

It cannot judge. It can present options, model scenarios, surface considerations — but it cannot make the call that requires weighing incommensurable values, reading the room, and putting your name on the outcome.

It cannot feel empathy. It can recognize emotional patterns and generate compassionate-sounding language. But it does not have stakes. It does not lie awake at 3am because a decision it made affected someone's livelihood. The felt weight of consequence — that's a human thing.

It cannot build relationships. Not real ones. Relationships are built through time, through conflict and repair, through showing up when it wasn't required, through trust accumulated in both directions. A model can simulate warmth. It cannot be a friend, a mentor, a trusted partner. Those bonds live in the human domain and they always will.

These are not soft skills. They are not nice-to-haves or cultural add-ons. They are the only category of capability that AI has structurally, permanently left to us — and they happen to be the things we've systematically undervalued for decades.

That's the reversal hiding inside this moment. We built organizations that rewarded execution over judgment, speed over depth, output over wisdom. We optimized humans to behave more like machines. Then we built machines that do it better. And now the capacities we deprioritized — empathy, judgment, relationship, accountability — are the most economically scarce and valuable things a human being can offer.

Two Futures, One Fork

This is the fork. And we are choosing between these two paths right now, mostly without acknowledging that a choice is being made.

The first future treats AI as a headcount reduction opportunity. You automate the work, you reduce the people, you capture the margin. Rational from a narrow financial perspective. Also a catastrophic misreading of what's possible — because organizations that eliminate humans aren't capturing a competitive advantage, they're hollowing out the judgment, context, and relationships that make strategy executable. They'll look lean and efficient right up until they don't.

More importantly: the aggregate effect of that choice, made by enough organizations, is a labor market that concentrates gains at the top and eliminates the middle. Not a revolution. A contraction.

The second future treats AI as a labor shift. The work that machines do well gets handed to machines. The humans move to the work that requires what humans specifically bring — judgment under genuine ambiguity, relationships built on trust and history, creativity that comes from lived perspective, ethical courage, contextual wisdom earned through actual failure. Organizations get more capable, not just leaner. People do more meaningful work, not less.

This future doesn't happen automatically. It requires decisions — by leaders, by policymakers, by workers, by all of us — about what we're actually optimizing for.

The Shared Responsibility

It would be simple to say this is a corporate governance problem. Or a policy problem. Or an individual problem. All three are partly true. None is sufficient alone.

Organizations have to make an active choice not to default to headcount reduction as the primary value capture from AI. That means rethinking what productivity measures, what gets invested in, what gets preserved even when it doesn't optimize cleanly. The companies treating this as a replacement opportunity are making a short-term bet with long-term costs they haven't fully priced — because you cannot automate your way to the judgment and relationships that earn trust at scale.

Governments and institutions have to create conditions where the labor shift is actually navigable. Education systems that develop judgment, not just technical skills. Transition support for workers in roles that are genuinely disappearing. Policy frameworks that keep the gains from AI from flowing entirely to capital while labor absorbs the disruption without a net.

Individuals have to stop waiting for someone else to define what's valuable about them. The skills that matter most — depth over breadth, genuine expertise, relational trust, the willingness to take a position — are developed through practice and investment. That is squarely in your control.

None of these actors can solve this alone. And the mistake we're most at risk of making is assuming that because the technology advances fast, the social and organizational response will automatically keep pace. It won't. Not without intention.

What Up-skilling Actually Means Now

So yes: up-skill. But be precise about what that means, because the old definition is exactly wrong.

Up-skilling used to mean: get better at executing best practices. Learn the methodology. Get certified. Know the right way to do the thing.

That's the floor now. AI already owns it. Chasing it harder won't help.

The new up-skilling means investing in the things that compound over time and cannot be replicated at scale.

Get deeper in your domain. Surface-level knowledge is being democratized — which means the value of genuine depth is going up, not down. The person who has spent years in a specific industry, who has the scar tissue and the relationships and the intuition that comes from actually doing the work — that person is not being replaced. They're being made more powerful, if they use the tools well.

Develop your judgment in public. Write. Take positions. Be willing to be wrong. The people who will matter most over the next decade are the ones whose thinking is legible — who have a track record of calling things clearly, of reasoning transparently, of saying what they actually think. That reputation cannot be automated.

Build relationships based on real reciprocity. Not collecting contacts. Actual relationships, where you know what people care about, where you've delivered something real, where trust has been earned in both directions. That network is not replaceable by any tool.

Learn to work with AI without losing yourself in it. Use it to execute your ideas faster. Don't use it to replace the work of forming ideas. The people who collapse into AI-generated output aren't becoming more capable — they're becoming more interchangeable. The people who use AI as a lever for their own distinct thinking are building something that compounds.

The Choice We're Making Right Now

We are in the middle of making a collective decision, mostly by not making it consciously.

Every organization that defaults to replacement instead of shift is voting for one future. Every institution that fails to update its frameworks is casting a vote. Every individual who competes on best practice instead of developing what lives above it is making a choice too.

The next industrial revolution doesn't have to cut people out. The technology doesn't require it. There is no law of nature that says the gains from AI have to flow to capital at the expense of labor, or that best practice becoming a commodity has to mean humans become one too.

But it will go that way — if we let the defaults run.

The machines have raised the floor for everyone. That's real and it's remarkable. The question is whether we use the space that creates to finally invest in what's always been most human about us — the judgment, the empathy, the relationships, the courage to be accountable.

Not because the machines can't do those things yet. Because those things are worth doing for their own sake. Because they're what make organizations trustworthy, decisions defensible, and work meaningful.

AI alleviated us from toil. Now define your value.