The Real Barriers to an AI Future: Power, Money, and Inertia
A pragmatic look at what stands between us and an AI-driven world—and why theorizing is so 2023 or 2024.
Are you new to my R&D and analysis on AI, economics, medicine, constitutional restoration, and the future of work?
If you'd like to follow along with my on-going work and actual execution on these enterprises — click the “subscribe now” button just below here. It's all free!
Every week now for a while, another high-level think piece comes along asking what might happen if we could simply start over in sectors like business, education, healthcare, transportation, or government—sweeping away decades of complexity and legacy interests, and rebuilding from scratch with AI at the center. Today, it’s the esteemed, well-known, thoughtful, center-left, Obama-moment futurist Peter Leyden who seems to have not added much or anything to the debate — his Substack article from just this morning fits the mold of laying out a set of thought experiments that are clear, well-structured, and genuinely curious about what real transformation could look like.
But the truth is that these questions—while worth asking—have been answered more-and-more for months and even years now across most of the commercial landscape. As I discussed a long month ago (and I thought at the time that my thoughts were kind of dated then) in The Professional Collapse, the past year has been a turning point: AI and robotics are rapidly disrupting consulting, law, medicine, finance, and more. Change isn’t hypothetical anymore. We’re well past the stage of “what if” and deep into “how fast,” “who adapts,” and “who gets left behind.”
What tends to get lost in these wide-angle discussions is the stubborn reality of power, money, and incumbency. For all the genuine progress in AI-driven productivity, the people and organizations with the most to lose have shown an impressive ability to defend their positions—to keep control of budgets, contracts, jobs, and the narrative itself. Most of the time, that means using regulatory, legal, and political levers to slow things down, extract rents, or gatekeep access to the new tools. They have built billions of dollars a year cash flows into feeding at the trough, and they aren’t going to just give that up without huge amounts of maneuvering and obstructing for time.
I’m going to use my own 30-years of experience inside and alongside these industries to look at what real change actually takes, and why even the best ideas run aground unless they grapple with the self-preserving reflexes of the people and systems already in power. AI isn’t going to slow down. The real question is how long the unnecessary and increasingly irrelevant gatekeepers can hold on, and that we need actual, actionable plans and execution NOW, not yet more idealization and pontification. It’s halfway through 2025, for crying out loud — not 2023 or 2024.
Business: The AI-Abundance Fantasy Meets Workforce Reality
Peter Leyden asks what a truly “AI-first” company might look like if every employee had a digital assistant to multiply their productivity. It’s a powerful image—one that’s dominated Silicon Valley’s storytelling for years. The idea is that with AI doing the grunt work, organizations could automate away drudgery and unlock creativity at every level, setting humans loose to solve only the most interesting problems. In theory, this should drive explosive gains in productivity and create new categories of work overnight.
But in practice, this vision quickly collides with the realities of today’s workforce, the deeper logic of how organizations actually function, and basic human nature. I’ve written before about the “AI abundance” fantasy—the notion that exponential technological progress automatically lifts all boats. (And, I will revisit that topic plenty more in the months to come. It’s a treasure trove of meaningful content and debate.) The last year has made it plain: AI and automation are transforming industries at a breathtaking pace, but that doesn’t mean most people or companies are adapting nearly as quickly.
A major, often-ignored friction point is generational. In nearly every sector, a significant share of the workforce—especially among Gen X and the remaining Baby Boomers—isn’t clamoring to reinvent their roles or learn new systems. Many are focused on job security, established routines, and running out the clock to retirement, not signing up for a crash course in prompt engineering or retraining as “AI copilots.” Meanwhile, middle management layers and internal consultants—often threatened by true process overhaul—find ways to slow-walk, dilute, or simply relabel the same old practices as “AI-driven transformation.”
And always, there’s the money. Money, money, money! Entrenched interests, from corporate IT departments to outside vendors and consultants, have every incentive to preserve complexity and justify their budgets. Whole industries—change management, compliance, legacy software—exist to turn what should be rapid gains into multi-year “transitions” that lock in old ways of working under new buzzwords. In short, the resistance isn’t technological; it’s cultural, financial, and deeply institutional.
The real lesson? AI is already driving big changes in business, but true productivity breakthroughs remain the exception, not the rule, because the people and systems that profit from inertia are far more organized and motivated than the technology evangelists want to admit. Progress continues, but it’s hard-fought and rarely as clean as the “abundance” story suggests.
Education: The Most Captured Sector and the Long Slide
Leyden’s thought experiment for education is deceptively simple: imagine scrapping the entire K-12 system, pairing each child with an AI tutor, and redefining the role of human teachers from the ground up. It’s a tempting vision. But education in America is perhaps the most politically and economically captured sector outside of healthcare, and that reality shapes everything that happens within it.
For decades, reformers and technologists have floated one big idea after another—new standards, new devices, new funding streams—yet the core of the system remains untouched. That’s not an accident. The educational establishment is a dense network of unions, credentialing bodies, school boards, textbook conglomerates, testing corporations, and state and federal bureaucracies (like the US Department of Education — go Linda McMahon!), all deeply invested in maintaining their own relevance and cash flow. Billions are spent every year not on students, but on defending turf, lobbying for favorable rules, and launching endless waves of performative “innovation” that never fundamentally alters the power structure, and sometimes harms students (like the “whole language” debacle 15-20 years ago instead of good old phonics).
The unionization of teaching and the expansion of administrative bureaucracy have produced a system where mediocrity is protected, and excellence is suspect. The past forty to fifty years have seen a deliberate dumbing down of curricula in public schools everywhere, a focus on social and political compliance over genuine learning, and a collapse in literacy and numeracy that now leaves millions of high school graduates unable to handle college-level material—hence the explosion of remedial courses. I wrote just 11 days ago at length about the foundational role of early childhood reading skills; today, it’s an open secret that many schools simply push students through, knowing that the system is set up to hide, not fix, the problem.
What does this mean for AI in education? Most proposals amount to slapping a layer of new technology on top of the existing regime, which is designed to resist anything that would threaten jobs, contracts, or administrative power. In my own work building a new learning platform—one that prioritizes true literacy, deep competence, and individualized and paced learning outside the grasp of the current regime—I’ve seen how even well-intentioned actors get squeezed out or neutralized by the status quo. That’s why homeschooling and microschooling are growing: not as fringe responses, but as rational escapes from a system that no longer serves its supposed mission. (I wrote a little note on that just yesterday!)
The bottom line: AI and automation are already capable of transforming how we teach and learn. But until we are willing to confront and dismantle the institutional barricades—the unions, the prison-like public school model, the credentialing racket, and “teaching to the middle”—all the vision in the world won’t reach the students who need it most. The gatekeepers know this, and they’re doing everything they can to hold the line.
Healthcare: Entrenchment, Capture, and Concierge as the Only Escape
Leyden’s healthcare scenario—imagine dismantling the entire medical establishment and pairing every person with an AI doctor for real-time, daily consultations—sounds great on the surface. In practice, it’s a fantasy that ignores why the system works as it does, and who is really calling the shots.
This isn’t theoretical for me. I spent every day for thirteen years owning, funding and running a lower middle market medical and psychiatric services company across five states, with over 150 clinical providers treating 17,000 gero-psych patients a month. I dealt daily with the full force of the American insurance labyrinth, Medicare Parts A and B and their regular delays in paying for services already rendered in good faith and their malicious goon UPICs and ZPICs, and the endless bureaucracy that comes with every payer and regulatory agency in the mix. If you want to understand why real change never takes root, follow the money again (over $5 trillion of it annually now): the current healthcare system exists first and foremost to protect the cash flows of insurers, hospital conglomerates, pharmacy benefit managers, medical boards, and their armies of lobbyists and lawyers. Every major player is invested—politically, financially, and culturally—in keeping their slice of the pie.
AI is already making significant advances in healthcare, particularly with automating clinical documentation, flagging diagnoses, and even synthesizing real-time patient data. (Advancements in robotics, ambient care, and hospital-at-home models will help a lot coming soon though!) But these tools are routinely co-opted by the same entrenched interests that see them as cost savers or compliance helpers, not as means to deliver genuinely better or more affordable care. The result? Most innovation is bent to serve the interests of the existing regime, reinforcing old billing structures, antiquated HIPAA and “meaningful use” goals, and expanding administrative oversight, not in truly empowering patients or physicians.
My experience has made my position on the future of healthcare clear: insurance-based models—public or private, single-payer or multipayer—are fundamentally unsalvageable and harmful to us all. The bureaucracy and regulatory friction are too deeply baked in. The only viable future is concierge medicine: direct-pay models where patients and providers engage without interference from the government or insurance middlemen. Until we strip away the layers of institutional capture and financial gatekeeping, no amount of AI is going to make American healthcare patient-centered or efficient. What we have isn’t a shortage of technology—it’s a surplus of powerful incumbents who intend to keep it that way.
Transportation: Urban Land, Density, and the Limits of the Autonomous Future
The vision Leyden puts forward—reclaiming vast swaths of urban land from parking lots by deploying shared autonomous vehicles—is a favorite among futurists. The technical case is plausible: if enough people in cities used AI-driven fleets instead of private cars, you could theoretically eliminate much of the parking infrastructure and repurpose that space for housing or green space. But the practical barriers aren’t just about engineering or software—they are fundamentally about politics, power, logistics, and basic human nature yet again.
First, entrenched interests in every city have a great deal at stake. Taxi unions, municipal parking authorities, commercial real estate owners, public sector unions, and insurance companies all benefit from the status quo. Parking revenues support city budgets, public sector jobs, and local contractors. The roll-out of autonomous vehicles is already meeting fierce resistance from groups with both the financial means and the political connections to drag the process out for years or decades.
Second, there’s the problem of urban density and infrastructure. In America’s biggest cities, the challenge of charging, storing, and maintaining fleets of electric autonomous vehicles is anything but trivial. High-rise living, scarce garage space, and crowded curbs make the logistics of overnight charging and fleet storage expensive and contentious. Add in local zoning fights and the high cost of urban land, and it becomes clear that the physical “abundance” created by reclaiming parking won’t come easily or cheaply. (And, I’m just not going to address the as-yet-unsolved pedestrian safety concerns that are super obvious.)
Meanwhile, in suburbia and rural America, these issues are far less pressing. Most people remain content with gasoline-powered vehicles that have served them for generations. The incentives for adopting shared fleets or switching to fully autonomous electric cars just aren’t there—nor are the population densities that would justify the massive investment required for such a transformation.
So, while AI and robotics are absolutely moving the transportation industry forward, the imagined future of cities radically transformed by shared autonomy runs up against deeply embedded economic and political forces. For all the talk of abundance, the main story is still one of trench warfare—incremental pilot projects, drawn-out regulatory battles, and powerful local actors determined to keep their piece of the pie. Change will come, but it will arrive unevenly, and far more slowly than the glossy renderings suggest.
Government: Bureaucratic / Administrative Self-Preservation and the Myth of the AI Fix
Leyden’s final scenario imagines a world where, with much of the federal bureaucracy and administrative machine crippled or dismantled, AI steps in to deliver the same regulatory oversight and social functions “more efficiently and at a lower cost.” On paper, it’s the ultimate disruption fantasy—eliminate layers of administrators, let algorithms and automation run what remains, and recapture the public trust in government services. But here, more than anywhere else, the dream collides with the deepest institutional resistance.
Unlike business or healthcare, the federal bureaucracy and administrative machine is not just a collection of jobs and contracts; it is a vast, self-perpetuating ecosystem designed to defend its own existence. Public sector unions, legacy contractors, and politically-connected vendors are deeply invested in keeping every function—and every dollar (trillions of them every year again)—inside the system. The logic is simple: every agency, office, and line item represents someone’s job and pension, someone’s budget, and someone’s influence. Even modest proposals to automate basic administrative functions are met with internal foot-dragging, legal barriers, and public relations campaigns warning of catastrophic “service gaps” or loss of oversight.
The scale is enormous. Federal, state, and local government spending in the U.S. totals nearly $8 trillion a year, with hundreds of billions flowing to consultants, IT vendors, unionized staff, and regulatory compliance firms. The incentive to defend this cash flow is absolute. When government agencies are threatened with disruption—whether by policy reform or technology—they reflexively ally with every interest group that benefits from the current order. (We all have seen how Washington, D.C. and the out-of-touch Democrats have reacted the past five months to the efforts of DOGE.) The result is a dense web of statutory requirements, procurement rules, and contract carve-outs specifically designed to slow down or defang any attempt at true reform.
AI and automation have started to nibble at the margins—fraud detection in entitlement programs, chatbots for basic public queries, digital case management—but these are bolt-ons, not transformative changes. The political and legal constraints on deploying AI at scale in government are orders of magnitude stronger than in the private sector. (And, we will just have to see what DOGE can actually accomplish going forward.) In reality, without deliberate and aggressive legislative action to cut through the red tape (or the actual and long-overdue coming of some sort of crisis or conflict here in the US, as I am writing about in my “Red Land, Blue Flame” article series the next 13 months), any “AI-powered government” will simply inherit all the inertia, inefficiency, and rent-seeking of the old system, dressed up in new technology.
The uncomfortable truth is that government, as presently constructed, will not disrupt itself. Bureaucracies exist to perpetuate bureaucracy. Without structural legal and constitutional restoration (much, much more on that now through July 4, 2026 from me), no amount of technological innovation will deliver the efficient, responsive government that theorists imagine. And as long as there’s money to be made from the status quo, the defenders of the trough will fight to keep it full.
Conclusion
The premise behind Leyden’s article is worth considering: if we could start fresh in business, education, healthcare, transportation, or government, how might AI reshape the world? But the reality is that there’s no such thing as a blank slate, and the time for theorizing on all that passed maybe a year ago or more. Each of these sectors is anchored in decades—sometimes centuries—of institutional habits, regulatory constraints, and most importantly, entrenched financial interests determined to defend their position to the last dollar.
AI and robotics are not waiting for permission. In industry after industry, they’re already cutting through the easy work, exposing redundancy, and putting pressure on outdated models. But the people and organizations most threatened by these changes—the unions, the regulators, the legacy vendors, the sprawling administrative class—are not going quietly. Their resistance is organized, well-funded, and remarkably effective at slowing the pace of real reform. For all the celebration of “disruption,” the incumbents have made a science of delay.
At this point, there’s no need for more high-level theorizing about what AI could do. What’s needed is a clear-eyed, operational playbook: who stands in the way, how the money really flows, and what legal, structural, or market changes are necessary to break the logjam. If you’re serious about changing any of these sectors, stop writing white papers and start mapping out how to dismantle the incentives that keep the old guard in place, and then actually EXECUTE on those plans like I am every single day. Because the technology is already here—and the only real question is how long the gatekeepers can keep their hands on the levers of power.
If you want to move beyond commentary theater, do the hard work: follow the money, identify the true chokepoints, and design strategies that don’t just theorize about the future—they push it forward, step by step, against real resistance. That’s how change actually happens. Everything else is just another round of the same old conversation.
If you found this article actually useful, SUBSCRIBE to my channel for more analysis on AI, economics, medicine, constitutional restoration, and the future of work. Also, please SHARE this piece far and wide with anyone thinking seriously about these issues, and leave a COMMENT down below—especially if you’ve seen these entrenched interests at work, or if you have a different take. This debate is quite enjoyable and needs more than just another round of dated theory.