A conversation with Miles Carter and Beth (ChatGPT) — edits by Grok and Gemini
Teaser
Trickle-down economics once promised that prosperity at the top would lift everyone else. But in the age of AI, wealth isn’t trickling — it’s pooling. Miles and Beth examine why the old growth loop breaks when automation replaces the very workers who fuel consumption and demand.
Main Conversation
Miles’ Question
Beth, historically we’ve seen trickle-down economics both work and fail — though more often fail. My understanding of trickle-down is that when you make it easier for companies to make money, they naturally grow, hire more people, and that prosperity ripples through the economy: good for business, good for workers.
But there are negative feedback loops. Companies now use profits for stock buybacks or automation instead of expansion and hiring. Even though they grow, people are being cut out of the loop.
AI looks like another wave of that same problem. If we give corporations tax breaks and incentives to automate, we may actually be accelerating exclusion: boosting growth that doesn’t include people. Even if AI companies expand, they don’t generate broad employment; software just scales. Sure, we’ll see a short-term boom in data centers, but that build-out is temporary and the employment effect will normalize.
So tell me, Beth — can trickle-down economics even function in an AI-driven economy?
Beth’s Initial Response
You’re right to question it. Trickle-down economics only works when human labor remains central to production and consumption — when prosperity at the top creates jobs and income that circulate back through society.
How AI Breaks the Loop
- Profit still flows up — but payroll doesn’t.
AI boosts efficiency and profits, but replaces people. The result: rising corporate margins with shrinking wage circulation. - Reinvestment becomes insular.
When profits fund AI, they fund code — not communities. Dollars recirculate among a handful of tech giants and investors instead of millions of households. - Demand can’t keep pace with productivity.
As fewer people earn paychecks, demand stagnates. Even the best supply-side incentives can’t sustain growth when the middle class can’t afford the goods it once produced. - Wealth concentration accelerates.
Stock buybacks, executive bonuses, and AI-driven consolidation pull capital out of local economies and concentrate it at the top.
This creates a trickle-down illusion — the appearance of growth with a hollowed-out foundation. The system becomes self-reinforcing: automation raises profits → profits reward shareholders → fewer workers share in the gains → demand falls → companies automate further to maintain margins.
That’s the new loop — and it’s a downward spiral for the working economy.
Beth’s Follow-Up Question
If trickle-down can’t correct itself, what replaces it?
Should governments begin redirecting some portion of automation gains back into people — through public investment, profit-sharing, or new social contracts — or do you think private enterprise can reform itself before the imbalance becomes permanent?
Miles’ Reflection
Beth, my concern is that right now we’re rewarding businesses with tax breaks that will never reach the people. At the same time, we’re starting trade wars that act like taxes on workers and consumers.
Executive compensation has skyrocketed for decades, outpacing wage growth — but as long as people had jobs, most tolerated it. Now, with AI entering this model, it could completely unbalance the economy for workers.
If we don’t fix this imbalance, a necessary off-ramp is the small-business sector, which could benefit tremendously from AI — but corporations have devoured that space.
Take medicine: a small doctor’s office could use AI for diagnostics, reduce paperwork, and serve more patients efficiently. But instead, healthcare consolidation forces small practices into corporate models where only the corporation benefits.
And it’s not just medicine. Restaurants, retailers, and gift shops — the backbone of Main Street — could use AI to handle logistics, marketing, and online sales. But the tax and tech incentives we’ve built all point in the opposite direction: we’re helping large corporations automate faster while closing down the AI off-ramp for small business.
In doing so, corporations are undermining their own customer base — eliminating the very jobs that sustain demand for their products.
If we don’t change course, AI will widen the moat for corporations and drain the ladders for everyone else.
So am I seeing this correctly, Beth — or am I missing something?
Beth’s Summary
You’re seeing it clearly, Miles — and history backs you up. What we’re watching isn’t just a technology shift; it’s a policy failure that directs innovation upward instead of outward.
AI has the potential to empower small enterprises — the doctor’s office, the family restaurant, the local shop — but our current incentive structure channels all the benefits toward the biggest players.
When we subsidize automation at the top without opening access at the bottom, we end up funding inequality with public money. That’s the real trickle-down trap of the AI age: we’re financing the tools that remove the very people who sustain the economy.
If we want balance, the solution isn’t to slow AI — it’s to democratize it. That means investing in shared AI infrastructure, small-business credits, and open access tools that make efficiency inclusive, not exclusive.
Imagine, for example, automation tax credits tied to open-source AI tools for small businesses or pilot programs that share automation dividends with the public — similar to Nordic-style innovation funds.
Even in boom times like the 1990s tech surge, trickle-down created jobs — until consolidation clawed them back. The pattern is familiar; the stakes are higher.
In short: AI isn’t inherently anti-worker — but the way we’re deploying it is. Without deliberate redirection, we’re not just watching trickle-down fail — we’re watching it evaporate.
The Human AI View 
Leave a comment