The Abundance Curve Is Bending

Most companies that don't dive headfirst into AI right now will be obsolete in less than five years. I don't say that to scare you. I say it because the math stopped being subtle, and I meet too many smart entrepreneurs who still can't feel it.

I've spent the last four years on this. Hundreds of EO and YPO talks, forums, conferences, all aimed at one thing: getting business owners to take this moment seriously. And the single biggest problem I run into isn't a lack of tools. It's a lack of urgency.

People don't see the exponential growth happening around them. We're all busy, all overwhelmed, and our brains are wired with cognitive distortions that let us rationalize things just enough to fall asleep at night. Meanwhile, the cost of capability is falling off a cliff, and AI just put its foot on the accelerator.

Let me be precise, because the precision is the point. We are not living in universal abundance. Housing is brutal. Healthcare is brutal. Childcare is brutal. What we're living in is capability abundance, where the cost of doing things is collapsing: communication, knowledge, media, computation, software, design, translation, education, and increasingly intelligence itself.

And here's the part that should grab you. Abundance isn't just rising. The rate of the rise is rising, and the acceleration itself is increasing. The curve isn't a straight line going up. It bends upward, and it bends harder every year. That's why most people miss it. They're reading a straight line on a graph that's actually starting to curl toward vertical.

This pattern isn't new, and it isn't only digital. The physical world rides the same kind of learning curve, where every doubling of production drives cost down. Batteries are the clearest case.

We're measuring an abundance machine with a scarcity dashboard

GDP is not useless. It tells us what people paid for. But it's a transaction ledger, not a lived-experience ledger, and that gap is now enormous.

The U.S. Bureau of Economic Analysis says plainly that GDP was never designed to be a welfare measure. That's exactly why Stanford built GDP-B, a framework that estimates the welfare from free and digital goods instead of only money spent. Using that approach, researchers found ten digital goods generated more than $2.5 trillion of consumer welfare per year across 13 countries, roughly 6% of their combined GDP, with the free ones helping lower-income people the most.

Sit with that. Abundance usually shows up first as welfare, not revenue. When a thing goes free, GDP often records zero, even when that thing replaced hundreds or thousands of dollars of spending. Your phone ate the GPS unit, the camera, the encyclopedia, the calculator, and a dozen other products, and most of it reads as nothing on the national books.

We keep using a scarcity dashboard to measure an abundance machine. Mike Schwarz, Founder & CEO, MyZone AI

The cost of intelligence is collapsing

AI is not just another technology. It's a cost-down engine for every other technology. It writes software, designs chips, optimizes logistics, tutors students, runs simulations, and improves the tools used to improve itself. And the price of running it is falling faster than almost anything in economic history.

The capability didn't plateau while prices fell. Stanford's 2026 AI Index put generative AI at 53% population adoption within three years, faster than the PC or the internet, and estimated $172 billion in annual U.S. consumer value by early 2026. The infrastructure is bending upward too.

Agents went from broken to doing almost all of my work

This is the part that still surprises me, and I live in it. We went from agents barely working to agents doing almost all of my work, fast. I now have days where my entire productivity jumps 20% in a single day. That's not a metaphor. That's a Tuesday.

We are so close to mass-producing software with limited human intervention, just guidance, judgment, and taste. The proof points keep landing. At its June 2026 launch, Anthropic said Stripe used Claude Fable 5 to migrate a 50 million-line Ruby codebase in a day, work that would have taken a team more than two months.

Then it got taken offline. Three days after launch, Anthropic disabled Fable 5 worldwide to comply with a US government export-control directive citing national security. If that model had stayed live, I'm convinced I'd have banked another 10x leap that week. That's the frontier right now. One day you find something close to magic, the next day the federal government pulls it.

The math people miss: 90% to 99% is 10x

Here's the thing that's easy to miss, and it's the most important number in this whole piece. I've heard credible reports that Anthropic is north of 90% automated on its own software development. People hear that and think the road from there is almost done. It isn't. It's just getting started.

The difference between 90% automated software development and 99% automated software development is 10x faster software development. It's easy to miss, but the math holds up. When the human-handled slice drops from 10% to 1%, the bottleneck shrinks tenfold, and output speeds up by roughly the same factor.

The difference between 90% and 99% automated software development is 10x faster software development. It's easy to miss, but the math holds up. Mike Schwarz, Founder & CEO, MyZone AI

Now stack the deeper unlock on top: AI improving the systems that improve AI. Google DeepMind's AlphaEvolve uses Gemini models with automated evaluators to discover better algorithms. Sakana's AI Scientist runs an open-ended research loop end to end. And METR found the length of tasks agents can complete is doubling roughly every seven months. Automation makes work cheaper. Recursive automation makes progress cheaper. We're not just automating work anymore. We're automating progress. You can see what this looks like in practice in our piece on agentic development.

Why you can't feel it: we get used to miracles in days

So why does almost everyone miss this? Because humans habituate to stimuli incredibly fast. It's how our species is wired.

Think about the first time you got wifi on a plane. You're moving at 700 kilometers an hour, 33,000 feet up, bouncing a signal off a satellite, and for the first half hour it feels like magic. Within a couple of days you're complaining the connection is slow. We get bored of miracles almost instantly, and then they feel normal.

It's the boiling-frog problem. Turn the heat up slowly enough and we adjust to almost anything. We're doing it with technology, and we did it with politics. We just keep quietly accepting things that would have stunned us a year earlier. On top of that, we live in a post-truth, fear-driven media world where scary and new get the clicks. Abundance doesn't get the clicks. So we're sitting in a truth vacuum while the most important story on Earth plays out in the background.

People like Peter Diamandis and Steven Kotler saw this coming. When they published Abundance in 2012, they argued exponential technologies would let us meet and beat humanity's basic needs. The first time I read it, it put words to things I'd been feeling for years. I started my tech career in 1999, building on the web before social media existed, and I watched the power of digital connection grow exponentially. It has not stopped since.

This is bigger than horses to cars, and it's faster

I want to be clear about the scale, because soft language doesn't do it justice. This is ten times bigger than the transition from horses to cars. It's ten times bigger than the shift from pre-electricity to post-electricity. And it's going to happen ten times faster. There's no precedent for what this is.

This is ten times bigger than horses to cars. Ten times bigger than electricity. And it's going to happen ten times faster. Mike Schwarz, Founder & CEO, MyZone AI

That's why the cost of standard software is about to collapse. For prototypes, integrations, migrations, internal tools, dashboards, and support tooling, 90 to 99% cost reductions are plausible at the task level. Mission-critical systems won't drop 99% overnight, because security, compliance, and human review still matter. But the marginal cost of creating working software is cratering, and that reorders entire markets.

It also means the gap between countries and companies is about to widen fast. Canada, where I'm from, is falling further behind the United States on this, and that's going to cost us a lifetime of opportunity. The same is true company to company. This is the biggest financial opportunity in the history of the planet, and most people aren't seeing it. Our AI strategy guide is where I'd start if you want to plan around it.

The real problem isn't the tech. It's inaction.

For the 5 to 10% of leaders who are all in, the pain is simply "where do I go from here?" And they're going to be fine, because they're doing the right thing. They're getting their hands dirty, jumping in, playing with the tools, getting a little scared, putting in the long hours. That's how you learn. That's how I did it. I came out the other side after an intensive first six months of diving deep, and then it all started to make sense.

For everyone else, the barrier is rarely the technology. It's culture, leadership, and resistance. Recent research from Writer and Workplace Intelligence found roughly one in three employees admit to behavior that amounts to sabotaging their company's AI rollout, rising to 44% of Gen Z. This is not an easy journey. I've been at it with my own team for four years.

We run monthly reviews of everyone's AI rocks, their AI learning rocks and automation rocks. We have coaching, monthly all-hands, weekly L10 meetings, and right now we're running morning training sessions five days a week, 8 to 9 am, and we can still barely keep up. I started the AI Wake-Up Call back in 2022 because I felt a moral obligation to get the word out to my fellow SMB CEOs. It wasn't even a business at first. It was camaraderie. The business grew out of it.

So here's the candid version, the way I'd say it to another founder over a beer.

The beer version The exponential productivity jumps that me and the other people I work with, people I'd genuinely consider AI experts, are seeing right now is just astounding. We had a report come in just this morning of a colleague who's seen a 20x increase in productivity in the last six months. Twenty times. And me personally? I'm often doubling my productivity week over week. Today's a perfect example. I found a way to double my productivity again, just like that. It really makes me wonder how the companies that aren't running down this rabbit hole are possibly going to keep up. Mike Schwarz, Founder & CEO, MyZone AI

The worst part, and the crazy part, is the uncertainty at the frontier. It changes day by day. One minute we think we've found AGI, the next minute the government shuts a model down. One minute you're safe, the next you're hacked. One minute ChatGPT is in the lead, then Claude, then Gemini. Planning for this is genuinely hard.

But one thing I know for sure: the work you do on education and preparedness pays dividends, and it's a lot of fun. As more capability goes cheap, human judgment becomes more valuable, not less. Taste, trust, distribution, and leadership are the new scarce goods. The question stops being "can we build it?" and becomes "do we understand the customer well enough to build the right thing?" If you want a feel for that shift, our AI automation guide shows where to begin.

The abundance curve is bending upward. Most people are still staring at the straight line. Stop making excuses about time, or not being technical, or getting to it next quarter.

Saying you don't have time to automate your business with AI is like saying I'm too busy washing dishes, I don't have time to go out and buy a dishwasher. Mike Schwarz, Founder & CEO, MyZone AI

The journey of a million miles begins with one step, and there has never been a better time to take it than right now. So what's the first workflow you're going to hand to AI this week?