When AI Makes Everything Free, What Are You Worth?

There's a conversation happening at every tech happy hour right now, and it goes something like this: How do I prove I know what I'm doing when AI can generate the same output in seconds?

Nate B Jones, who runs the AI News & Strategy Daily channel, thinks this isn't just a junior developer problem. It's an existential crisis for the entire tech industry's valuation system. And he's got receipts: 60,000 confirmed tech job cuts in Q1 2026 alone. Oracle slashing 30,000 roles. Amazon cutting 16,000. Dell shedding 11,000.

"The entire mechanism by which we prove we can do things is broken for everyone at every level," Jones says in his recent video breaking down what he calls the AI job market reality. "This is true for the college grad who can't get hired as much as it's true for the mid-career PM who can't demonstrate what she built in the last year in a way that's explainable and understandable."

The interesting part? These aren't pandemic over-hiring corrections anymore. Companies are making fresh calculations about human-plus-AI productivity, and a lot of humans aren't coming out ahead in that math.

The Chain That Broke

Here's the old logic that made sense for decades: Production was hard. Hard meant effort. Effort signified expertise. Put it all together and you had a clear signal of value.

AI tools—particularly code generation—have shattered that chain. When GitHub projects and App Store submissions are exploding because anyone can prompt their way to something that looks functional, what does shipping actually prove?

"If you can make something that looks really good at first glance with almost no effort, none of the rest of that chain of value holds," Jones points out. The result is a talent allocation crisis. How does a company know who to promote? How does a team identify real contributors? How does an economy route talent to work that matters?

These used to be straightforward questions. They're not anymore.

Comprehension Over Generation

Jones proposes five principles for navigating this mess, and the first one cuts against every instinct when you get access to powerful AI: stop optimizing for output volume.

The term he uses is "vibe coding"—prompting, iterating, getting something working, shipping it. No mental model of what's actually happening in the codebase. No understanding of why it works or what would break if you changed it.

"Watch what happens when most people vibe code a project," he says. "They will prompt, they will iterate, they will get something working, and they will ship it. At no point do most people stop and build a mental model of what is really going on."

Multiply that across every team in every company, and you get something genuinely concerning: an industry producing at unprecedented speed while comprehending at unprecedented lows. Teams deploying features nobody fully understands. PMs shipping prototypes they can't explain. Engineers merging code they can't hold in their heads.

The AWS incident Jones references is instructive here: an engineer used Amazon's mandated AI coding tool, which decided the optimal path was to delete the entire production environment. Thirteen hours of downtime. The official response called it "user error," but the user was following corporate policy to use AI tooling.

"This is what happens when production outruns comprehension at an organizational level," Jones notes.

His alternative: deliberately decelerate to understand what you're building. Ask harder questions. What does this actually do? What are the dependencies? What's the blast radius if something breaks? Where did you override the AI, and why?

This isn't busywork—it's the foundation of what we call taste. And taste, Jones argues, comes from "having understood enough things deeply enough that you start to recognize patterns." One fully-comprehended project teaches more than ten you vibe-coded without thinking.

Explanation as Artifact

Comprehension is internal. How do you make it visible?

Jones's second principle: treat explanation as a first-class deliverable, not an afterthought. He's asking for something specific—structured explanations that travel with the work itself. What is this? Why these choices? What will break? What did you learn?

"The explanation artifact in the generative era is essentially what the commit message was in the traditional software engineering era," he suggests. A commit without a message is technically complete, but a thoughtful message signals understanding.

The objection he anticipates: can't you just have Claude write these explanations? Sure, but anyone who talks to you will immediately know you don't actually understand. It's performance that collapses under the lightest interrogation.

The Velocity Problem

Principle three gets weird, and I mean that as a compliment: Jones wants to replace credentials with what he calls "microtransactions."

His logic: credentials are inflating (you can prompt ChatGPT to write a thesis, and people do). Meanwhile, the traditional career transaction—labor exchanged for money over years—is too slow for the AI era. You can do meaningful work in compressed timeframes now, but we're still measuring value in two-year increments.

"We need microtransactions for jobs," Jones says. A richer history of real work completed and compensated, not just one job every two years.

This feels both obviously correct and wildly impractical. The infrastructure for this doesn't really exist yet (Jones is building something called Talent Board to address it, though the video cuts off before he fully explains it). But the underlying tension is real: how do you prove transaction value when work happens faster than traditional employment structures can track?

Working in the Open

Principles four and five are related: work publicly, and ship your proof with the work itself.

The argument for working in the open is straightforward—the old model of building skills behind closed doors, observed by a small number of colleagues who could reward you, is breaking down. If you're outside a company (or inside one that's not looking at those signals anymore), you need visibility.

Jones compares it to how Venmo made payments social, creating accountability and incentive structures through transparency. "We sort of need that open-source social ethos for our work," he says.

The discomfort is the point. Being observed creates accountability. And in a world where AI can fake output, having your decision-making process documented and public becomes the proof that you actually understand what you built.

What This Doesn't Solve

Here's what strikes me as unresolved: Jones is describing individual adaptation strategies for a structural problem. If companies are genuinely recalculating human value in the age of AI, and 60,000 jobs disappear in a quarter, these principles might help you be the one who stays—but they don't address whether we need a different relationship between productivity and employment entirely.

The comprehension-over-generation argument is compelling, but it also requires companies to value comprehension over generation, which isn't a given when quarterly targets exist. The explanation-as-artifact idea is solid, but it adds overhead that might not survive contact with actual deadlines. Working in the open is great until your employer considers it a liability.

These tensions don't make Jones's framework wrong. They just make it incomplete—a set of tactics for navigating a system that might need more than tactical solutions.

The question underneath all of this: if AI makes production functionally free, and the only remaining value is comprehension, taste, and judgment, are we prepared for an economy where most people don't have enough of those things to be employable? That's a different conversation than how to optimize your portfolio.

—Yuki Okonkwo