Hello, Futurists.
Three ideas about the future are becoming increasingly clear to me.
Not because I think I know exactly what will happen. Nobody does.
But the more I observe technological progress, the money being invested, and the incentives behind it all, the harder it becomes to ignore where we seem to be heading.
1. Progress isn't going to stop
One thing I think we need to accept is that artificial intelligence development isn't going to stop.
Not because we all think every advance is necessarily good. Because the incentives to continue are too big.
Companies compete with each other. Countries compete with each other. Scientists want to discover new things. Investors want to back the winner. Governments understand that falling behind in AI could become a huge economic and geopolitical disadvantage.
AI has become an economic competition. A scientific competition. A military competition. And, increasingly, a competition between countries.
This isn't just an isolated intuition. The CEOs of the companies building this technology openly speak in those terms.
Sam Altman wrote:
“I want the United States to win in artificial intelligence, in both open-source and proprietary models.”
Sam Altman, CEO of OpenAI
And Jensen Huang, NVIDIA's CEO, has summed up the scale of this transformation by saying AI:
“Will transform every industry, power every company, and be developed by every country.”
Jensen Huang, CEO of NVIDIA
That explains why it's so hard to imagine a scenario where we simply decide to stop.
Imagine a company decides not to develop a more powerful model out of caution. Its competitor will probably continue.
Imagine the United States decides to drastically reduce its investment. China has a huge incentive to keep going.
Imagine a university stops researching. Another will do it.
No organization wants to discover five years from now that it decided to stop while the rest of the world kept moving.
That's the problem with technology races. Once they begin, stopping them requires coordination between actors with completely different incentives.
That's why I believe that, whatever our opinions about AI, we should prepare for a world where these systems keep improving.
2. Artificial intelligence will surpass us
The second idea may be uncomfortable.
But if progress continues, I think we'll reach a point where AI systems outperform us at practically every measurable intellectual task.
We're already seeing small versions of this. Models that code better than a large share of people. Models able to analyze enormous amounts of information. Solve math problems. Research. Write. Design. Work for hours without getting tired.
What matters isn't simply comparing an AI to a person. The truly powerful thing happens when we use AI to accelerate the process of creating new technologies itself.
Sam Altman has outlined a fairly simple progression:
“1. AGI accelerating research 2. AGI accelerating businesses 3. Personal AGI accelerating everyone…”
Sam Altman, CEO of OpenAI
This matters because it completely changes the speed of progress.
Using AI to do better marketing makes us more productive. Using it to write better code makes us more productive.
But if we use AI to discover new medicines, design new chips, find new energy sources, or even improve AI systems themselves, we begin accelerating the rate at which we produce new discoveries.
Technology starts helping us create better technology. And that can change the curve dramatically.
That's why I think we'll reach a point where saying “AI is smarter than us” won't be a particularly controversial statement.
The question will be what “smarter” means exactly.
A human can still have different experiences, emotions, relationships, and forms of creativity. But in terms of processing information, solving certain problems, coding, researching, or analyzing possibilities, it's hard to imagine we'll always retain the advantage.
The creators of these systems themselves acknowledge that we're entering territory that's hard to predict.
Altman recently wrote:
“AI is becoming extremely capable. Nobody fully understands the consequences of this.”
Sam Altman, CEO of OpenAI
For me, that sentence captures our current moment well.
We're building something increasingly powerful. We know it can deliver enormous benefits. But we still don't know exactly how an entire society reorganizes around that capability.
Which brings us to the third point.
3. We're going to make mistakes
I think it's practically impossible to go through a technological transformation of this magnitude without making mistakes.
Not necessarily because the people building AI are irresponsible. Because we're doing something new.
Every major technology in history brought consequences its creators didn't fully anticipate.
The internet gave us almost unlimited access to information. It also produced misinformation at a previously impossible scale.
Social media connected billions of people. It also discovered how to capture our attention in ways we're still trying to understand.
Cars gave us mobility. They also brought accidents, pollution, and cities designed around roads.
Technology solves problems. And often creates others.
AI will probably be similar. Except this time, the technology can advance much faster.
We can build a system with a new capability and only afterward discover how millions of people use it.
We can automate a task and then discover economic consequences we hadn't considered.
We can create extremely powerful tools for scientists and discover that people with bad intentions can use them too.
We can replace inefficient processes while displacing people who depended on them for a living.
That doesn't make artificial intelligence bad.
It's simply what happens when technological capability advances faster than our ability to learn how to use it.
Interestingly, even the people leading this race recognize that contradiction. On one hand, they believe AI can produce extraordinary benefits. On the other, they openly acknowledge risks we don't yet fully understand.
Sam Altman has put it this way:
“There's an obvious tension here… we're clearly at a stage of development where we think caution is justified… AI is becoming extremely capable; nobody fully understands the consequences of this.”
Sam Altman, CEO of OpenAI
Dario Amodei, CEO of Anthropic, has expressed a similar position. His argument is that we need to be able to speak honestly about both things at once:
The benefits can be extraordinary and the risks can be real.
In fact, Amodei has advocated mandatory third-party evaluations of the most advanced models for cybersecurity, biological, and autonomy risks—including the possibility of preventing deployment when a model's risks are too high.
For additional context, he also explored many of these concerns in The Adolescence of Technology:
I think that's an important signal. These aren't people outside the revolution saying everything is dangerous. These are the people building the most advanced systems saying:
This can be incredibly beneficial, and precisely because of that, we need to be careful.
Both things can be true.
We can advance and make mistakes. Create abundance and new problems. Cure diseases and create risks that didn't exist before. Make millions of people much more productive and disrupt the labor market.
I don't think the answer is panic. Nor do I think it's pretending there are no risks. I think we'll have to learn as we go.
So what do we do?
Here's where my position becomes much more practical.
I don't think you need to become an AI researcher. You don't need to understand every new model either.
But I do think you should use AI constantly.
There's an enormous difference between observing this revolution from outside and experiencing it from within.
When you use these tools every day, you start developing intuition. You understand what they do well. Where they fail. Which tasks you can delegate. What still needs your judgment. Which skills are starting to lose value. Which new skills are becoming important.
That knowledge will be much more useful than trying to predict exactly what will happen in ten years.
Altman has described something similar when discussing a process where society and technology evolve together:
“An iterative cycle where society and this technology evolve together is what gives us the best chance of getting this right.”
Sam Altman, CEO of OpenAI
I think that's probably the right attitude. Not watching AI from a distance until someone perfectly explains what will happen.
Experiment with it. Learn. Make mistakes. Develop judgment. Adapt as the technology itself changes.
Probably nobody can tell you exactly what the world will look like in ten years. Not Sam Altman. Not Elon Musk. Not Jensen Huang. Not Dario Amodei. Not me.
That's precisely the point.
You don't need to predict the future perfectly.
You need to stay close enough to the technology to recognize where it's heading.
Progress probably won't stop. AI will probably surpass many of our capabilities. And we'll make mistakes along the way.
All three can be true at once. The important question is what you decide to do while it's happening.
This piece took me about three hours of writing, researching, and organizing these ideas so you could read them in a few minutes.
Honestly, that's one of the parts I enjoy most: taking complex topics, spending hours trying to understand them, and then explaining them as clearly as possible.
Thank you for reading this far.
If you have a minute, send me an email and tell me what you thought. If there's a topic you'd like me to explore in depth in a future edition, let me know too.