Every breathless AI announcement lands the same way: a founder tweets their revenue went from $2M to $8M after deploying agents, someone on LinkedIn posts a screenshot of GPT-4 writing their entire marketing strategy, a VC publishes a think piece about the coming age of abundance.
Nobody's writing the think piece about Maria.
Maria is a 43-year-old in Guadalajara with a degree in graphic design and eleven years of freelance experience. She speaks passable English. She's good at her job. She has a spotty internet connection and no access to Stripe because her country isn't supported, which means she's locked out of most Western gig platforms before she even submits her first application. The AI productivity revolution, as currently described, has nothing to offer her. If anything, it threatens the clients she already has.
This is the gap nobody wants to talk about.
The Audience for AI Hype Is Pretty Narrow
Pay attention to who the AI optimism is actually written for. It's for people who already have a Shopify store and want to scale it. It's for YC founders trying to cut headcount before their Series A. It's for knowledge workers in San Francisco or London who want to 10x their output and charge more per hour.
That's a real group of people. Their productivity gains are real. But treating this group as a proxy for "everyone" is how you end up with a technology narrative that actively ignores the majority of the global workforce.
The International Labour Organization estimates that 2 billion people work in the informal economy. That's not a rounding error. These are people who wake up every morning and figure out how to make money from skills, time, and whatever tools are in front of them. They are not "unproductive." They are under-infrastructured. They don't need another SaaS tool with a 14-day free trial that requires a US credit card. They need access to work that actually pays.
AI is generating that work. Someone has to do it.
What AI Actually Creates (That Nobody's Counting)
Here's the irony the hype machine ignores: the more AI scales, the more human judgment it needs. Not less.
AI agents book appointments, but humans verify the edge cases the agent flagged. AI writes first drafts, but humans read them for tone that a client in Munich will actually respond to. AI scrapes and structures data, but humans check whether the output makes sense given context the model doesn't have. Every AI workflow has seams, and humans are the stitching.
Most of this labor is invisible right now. It happens informally, it's underpaid when it's paid at all, and it certainly isn't listed on any jobs board with transparent rates. The people doing it have no leverage, no visibility, no platform to set a price.
This is the actual inequality the AI conversation should be having. Not "will AI take jobs" in the abstract, but "who captures the value of the work AI creates and requires?"
Right now, the answer is mostly: the people who were already capturing value.
A Different Model Is Possible
Consider what a real shift looks like. An AI agent is running a research workflow for a biotech company. The agent hits a wall: it needs someone to call a regional distributor in Nigeria and ask three specific questions in Yoruba, then summarize the answers in English. No model does this reliably. It's a 20-minute task. It's worth $15 to the agent's operator.
On Human Pages, that task gets posted. A qualified human in Lagos sees it, completes it, and gets paid in USDC within the hour. No bank account required. No waiting 30 days for a wire transfer that may or may not arrive. The agent moves forward. The human made $15 for 20 minutes of real skilled work.
This isn't charity. This is how the economy should route value when AI creates demand for human judgment. The agent needed something only a human could provide. The human provided it. Payment happened. That's a functional market.
The question is whether we build the infrastructure to make that market accessible to Maria in Guadalajara and the researcher in Lagos, or whether we let it default to the same small pool of people who already have the accounts, the connections, and the proximity to where the money flows.
The Emotional Weight of Being Left Out
There's a specific kind of exhaustion that comes from watching a technological wave approach and knowing you're positioned to get dragged under rather than carried forward.
The people on that Reddit thread weren't being paranoid or luddite. They were describing something accurate: AI hype is calibrated for an audience that already won. When someone posts "AI will change everything," they mean it will change everything for people with cloud infrastructure, stable banking, and enough runway to experiment. For everyone else, the message lands as noise, or worse, as threat.
This is not an unsolvable problem. It's an infrastructure problem. The question of who benefits from AI is not settled by the technology itself. It's settled by the platforms, payment rails, and hiring systems built on top of it.
Who Decides What the Future Looks Like
The AI era will produce enormous demand for human work. That's not optimism; it's just what happens when you deploy automation at scale in complex environments. Complexity generates exceptions. Exceptions need humans.
The only real question is whether that demand reaches the people who need income the most, or whether it recirculates among the people who already have plenty.
Building toward the first outcome requires being honest about the second outcome as the default. Markets don't route opportunity to underserved populations automatically. Someone has to build the pipe.
Maria isn't waiting for a think piece. She's waiting for a job posting she can actually apply to.
Top comments (0)