Every Human in the Loop

The AI Divide (and Why It Matters)

Two developers. Same skills. Completely different futures.

One lives in San Francisco. Uses Claude, GitHub Copilot, and every AI tool he wants. His productivity has tripled. His salary: $400,000 a year.

The other lives in Lagos. Unstable internet. AI tools are expensive in his local currency. His freelance contracts are drying up because clients prefer “AI-augmented” developers who deliver faster. His income has dropped 60% in two years.

This isn’t a story about talent or work ethic. It’s about something more unsettling: AI isn’t just changing who does the work. It’s changing who gets paid.

Why This Gap Is Different

Every technological revolution has created winners and losers. The printing press replaced scribes but created editors. The industrial age displaced artisans but gave rise to engineers. The internet killed local stores but built global markets.

Each transition was brutal. But it also created new categories of work—usually more jobs than it destroyed.

AI feels different. It doesn’t just automate tasks. It concentrates economic value among those who can access, afford, and effectively use AI tools—while making everyone else’s work worth less.

What’s Happening Right Now

Writing: Agencies in the U.S. use AI to produce more content with fewer writers. Those who remain earn more. Freelancers in the Philippines and India are watching their rates collapse.

Customer service: Entire call centers are being replaced by AI assistants. The displaced workers aren’t moving to “better” jobs—they’re leaving the workforce altogether.

Software development: Entry-level positions are disappearing. Companies are hiring fewer people but paying senior developers much more, because AI amplifies what experienced talent can do. The bottom rung of the ladder is vanishing.

Access Isn’t Equal

Previous technology gaps could be closed with education and infrastructure. Learn to code, get online, upskill—the path was clear, if difficult.

The AI gap is more insidious:

Access costs money. Premium AI tools run $20–200 a month—affordable in wealthy countries, prohibitive elsewhere. Free versions are limited. The best models need stable internet connections much of the world still lacks.

The learning curve never stops moving. Using AI effectively isn’t just access—it’s knowing how to prompt, when to trust, and how to integrate it into your workflow. That meta-skill compounds over time, giving a huge advantage to those with consistent access.

Network effects amplify the gap. Early adopters build reputations, attract better opportunities, and join communities that share techniques. Those without access fall behind—not because they’re less capable, but because they’re excluded from the ecosystem of learning.

Which Humans Stay “in the Loop”?

We talk about keeping “humans in the loop” as if it’s reassuring—as if AI will simply enhance human judgment and keep us essential.

But here’s the uncomfortable question: which humans get to be in the loop?

Right now, being “in the loop” means:

• Having reliable access to advanced AI tools
• Living in a region with stable internet infrastructure
• Speaking fluent English (most AI training data is in English)
• Having the education to understand AI’s limits and capabilities
• Working in industries where AI augments, not replaces, humans

For everyone else? They’re not in the loop. They’re being left out entirely.

AI Isn’t the Answer to Everything

The optimistic view says access will naturally democratize. Costs will fall. Infrastructure will improve. New jobs will appear.

The pessimistic view sees a permanent underclass: a small elite capturing most of the value, a shrinking middle class, and developing countries left further behind.

The realistic view? We’re heading toward that pessimistic scenario—unless we actively change course.

Because AI doesn’t have to be just a wealth concentrator. It can support someone battling addiction. It can comfort someone who feels alone. It can teach. It can connect people who need each other.

The technology already exists. The issue, as always, is that there may not be a financial incentive to build those kinds of solutions.

So Where Do We Start?

Maybe the question isn’t just about documenting the gap—it’s about closing it.

How do we use AI as a tool for inclusion, not just for profit?
How do we keep every human in the loop?
Do you think this gap is inevitable—or can it be closed?
What concrete solutions can you imagine?
Who should lead this change—governments, companies, or communities?

AI is terrifying—but also deeply exciting. And we can still use it to make humanity better. We just have to start now, start small, and start together.