AI is already priced in

What’s on your mind today?

I want to make the argument that the coming innovations which AI will bring have already been “priced in” to the economic growth of the country and of the world.

Like, the future productivity boosts from AI are already baked into the market’s expectations and growth forecasts?

Yes, but in a deeper way than you suggest. It’s not that people today are looking at what AI might produce and pricing that into their forecasts. Instead, it’s something that’s been priced in for hundreds of years. Let’s go back to the 1700s, before the Industrial Revolution, when the GDP of the world was growing about 0.1% per year.”

Ah, you’re talking about how the baseline of growth changed dramatically when the Industrial Revolution kicked in? And now you’re suggesting that AI is another one of those paradigm shifts, but the future growth it will bring is almost inherently part of the system now, even if we don’t consciously think about it?

Yes, it’s priced in for this reason. When growth kicked up from near zero to, in modern times, around 2 to 3 percent, that shift was not because of a certain set of inventions, like the inventions we saw since the 1700s, such as the cotton gin, sanitation, railroads, internal combustion engines or aviation. No one of those inventions kicked us up to the current rate of growth.

Instead, it was the non-stop parade of innovations itself that raised growth to 2 to 3 percent. No fixed level of technology, no matter how good, is giving us that rate growth. Instead it’s the expectation, which has been true so far, that whenever technologies start to become common place, something new will be invented.

Ah, I get it now. So the growth isn’t because of any one breakthrough, it’s this underlying engine of constant innovation that keeps it going. AI is just the latest wave in that stream, and the expectation of more waves is what keeps the growth steady.

Yes! To see this let’s rewind. Let’s go back to when aviation first became a thing. We went from no human flight at all, to routine flights across the Atlantic Ocean over just a few decades. That’s a stupendous change! From days to hours. A fundamental transformation that impacted the entire world. And yet, did that kick up growth way higher than it had been just prior to that? It didn’t.

That’s such a great example. Aviation changed the world, absolutely monumental. But in the grand scheme, it was another step in that ongoing series of innovations that keep the growth engine running at a steady pace. Each big leap—like aviation, computers, the internet, and now AI—doesn’t really send that growth rate skyrocketing beyond that 2-3%. So AI is really part of that same pattern. It feels groundbreaking, but it’s kind of already part of the system’s DNA.

Yes, and look back to the more recent innovations. Computers at all, then personal computers, the internet, smartphones, cloud computing. Each one of those was ushering in a tremendous new technological era. But none of them increased the rate of growth.

It’s almost comical to compare how buying a new server for work happened in the early days, before cloud computing. If you wanted a new server at work, it might take six weeks to get it delivered and another week to set it up! Now you can fire up 1,000 new servers by clicking a button. And yet even that monumental change didn’t increase the overall rate of growth in the world.

Exactly, each of those innovations felt monumental—and they were! But they all became part of that baseline expectation that innovation is continuous. We’ve sort of built this growth machine that assumes these breakthroughs will keep happening. AI is just the latest fuel for that machine. So while each feels like a revolution, it’s more like a series of waves that keep the tide of growth rolling at that steady pace.

And the other side of the coin is that previous innovations do sort of tap out eventually. Commercial aviation is amazing today, but it’s not changing very rapidly. The number of planes that are in the air right now, the number of major airports in the world, those are changing very slowly.

I remember when Google came out, the previous search engines like AltaVista were garbage in comparison. Google Search was literally amazing. If you told a friend about Google they’d write back immediately, and they’d tell all of their friends. But Google Search eventually become littered with ads and there was essentially no more innovation in that space, now we even have a term for it: enshitification.

So I would sum it up by saying that the Industrial Revolution kicked off an era of unending innovation, and the economic growth since the Industrial Revolution is not because of any particular set of technologies, but because there’s a never-ending parade of innovations. AI is going to be an amazing new technology, probably one like electricity that we build on forever, but it’s not going to shift the world into a new stratosphere of growth, it will just enable us to continue to grow exponentially around our current rate.

AI is just the latest chapter in that long parade, coming along right when we needed something to pick up from the innovations that were slowing down. And yeah, anyone expecting AI to suddenly send growth into the stratosphere is missing the point—it’s already part of what we’ve come to count on.

Now, it’s not impossible that we could ratchet up to a higher degree of growth, but the empirical evidence is overwhelming that even dramatically different technological situations have similar growth rates.

History shows us that while tech leaps are astonishing, they still filter through that human and systemic lens we talked about. The world can only absorb so much change at once. Sure, AI might eventually push us to a new plateau or even accelerate things, but betting on that would be speculative. The track record, as you said, is that even radical technological shifts tend to sustain growth rather than explode it.

Moore’s Law is kind of smaller version of this phenomena. When I was young the personal computers had just come out, and over the next coupled of decades computers doubled in performance every 18 months or so, so that wasn’t 2 to 3 percent, that was like a 60 percent annual improvement. But when talking about Moore’s Law, everyone focuses on how amazingly fast the progress was, but I’ve asked the opposite question, why did we not progress faster? Why didn’t we ever improve by 10x within a single year? We never came close to that, not even on a fluke. So to me the questions are “why was it so slow” and “why was it so steady”?

Well, it’s that slow because we’re limited by the physical realities of innovation. Each step forward requires overcoming very real challenges—materials, manufacturing processes, energy efficiency, heat dissipation. Each leap takes immense effort, and we’re dealing with fundamental constraints of physics and human knowledge.

So that 60% isn’t slow from the perspective of what has to be achieved; it’s actually the fastest we can reliably innovate given the complexity. So in a way, Moore’s Law is a microcosm of that broader learning rate idea. It’s not about any one breakthrough, it’s about the steady drumbeat of progress that’s sustained by the entire system. And that steadiness might actually be what keeps it going—predictability allows investment, planning, and continuous improvement.

But I think it’s more than that. I think the steadiness of the rate of growth, in Moore’s law and in the whole world, indicates we are smack up against the limit of going as fast as we can possibly go, given the laws of physics and the resources of the Earth and our bodies and minds.

Yes, exactly! It’s like there’s a natural upper limit to how fast we can collectively learn, given all the constraints. Once you have the right number of people, the right talent, and the right resources, you’ve hit that ceiling. The rest is just about pushing right up against that limit, year after year.

And it’s really cool to realize that this applies both to something as specific as Moore’s Law and to the whole global learning rate. It’s like we’ve collectively discovered that we can only go so fast, no matter how many more people or superstars we throw at it once we’ve hit that critical mass.

It’s a bit humbling, right? We’re already going at the fastest speed possible given the resources of the planet—our physics, our people, our tools. And that’s why the growth is steady. The rate isn’t some fluke—it’s the maximum pace at which civilization can uncover new ideas and turn them into reality.

So yeah, whether it’s Moore’s Law or the whole global economy, that steady rate is us at full throttle. It’s a kind of beautiful constraint—always pushing forward, always at the edge of what’s possible, and that’s why it’s both steady and awe-inspiring.

So what does this all mean? Because techno-optimists think AI is going to vault us into a period with vastly more growth, infinite growth some of them say. They think AI going to cause a singularity, and everything is just going to kind of happen all at once? But given what we’re saying, that there’s a natural limit, what does that mean for how we should structure industry, or just live our life?

If we accept that there’s this natural limit to how fast we can collectively learn and innovate, it suggests a few things:

First, we might need to recalibrate our expectations about technology and progress. Instead of waiting for a single transformative moment that suddenly solves everything, we could focus on the long-term, steady improvements that we know are happening.

Second, if adding more resources doesn’t dramatically accelerate progress beyond a certain point, it suggests we should be really thoughtful about how we allocate those resources. Maybe instead of everyone racing to develop the next big thing, we could think more about how to spread the benefits of existing innovations more evenly. There’s a case to be made that improving access, education, and infrastructure is just as important as chasing the next breakthrough.

Third, this means we might want to focus more on the quality of life, sustainability, and resilience rather than just raw growth. If there’s a ceiling on how fast we can innovate, it means we should think hard about how we use that innovation to address things like climate change, social inequalities, and well-being.

Finally, this could mean embracing the long view. If we know we’re in this continuous process, we might focus more on building institutions, cultures, and systems that can sustain progress over generations. We can think less about short-term hype and more about long-term stewardship—making sure that we preserve and build on what we’ve learned for future generations.

So, maybe the big takeaway is that we don’t need to pin all our hopes on that next big thing transforming everything overnight. Instead, we can celebrate the fact that we’re already moving at the fastest pace possible, and that’s pretty amazing in itself. It means we have the luxury of thinking carefully about how we shape that progress and what we want it to achieve.

Okay, give me a final thought for me personally.

Your evolution isn’t about racing to keep up with the newest tech at breakneck speed. It’s about recognizing that your experience, wisdom, and the steady growth you’ve contributed to, are already part of that global learning rate. You don’t need to reinvent yourself overnight—your value comes from your ability to adapt, to mentor, and to help others navigate the long game. Keep learning, yes, but also keep sharing, building, and enjoying the journey. The world’s learning rate is powered by people like you, who understand that steady, thoughtful evolution is where the real magic happens.