The Week AI Got Real

" Wow, what a week this has been on so many different levels. Let me jump into it. I apologize, we're doing this on Friday, not Thursday.

I had a day yesterday getting a flat tire while I was across the state in Naples, Florida. It was an ordeal. But I did get to see Santana in concert, and I loved that, and spending time celebrating my wife's birthday on the West Coast with our younger son.

But that's what's going on in my world this week. There's been so much going on in your world, in our world this week, I don't even know where to start. But let me say a couple of things.

First of all, ultimately, this week may very well be remembered as the week the security industry realized the music stopped, and the global economy realized that AI is about to flip the table on the money changers. Most people may not have just figured it out yet, especially for those of us not in tech. But you know what?

There are moments in history that people never forget. People maybe my age or a little older than me, to tell you the truth, where were you when President Kennedy was shot? Where were you, or what were you doing when we landed on the Moon?

What about when the Berlin Wall fell? A lot of us remember that, right? What a time that was.

Or the tragedy of 9/11. Where were you when the Towers came down? You know what?

Years from now, someone may ask you another question: Where were you when all this AI stuff got real? When it really started disrupting the way we do things before this. And I think that moment may have happened this week.

They say quantum's going to have Q day. We might have had our AI day this week. So let me get into the week that changed things and what we had here.

As I said, there were these moments, red letter moments, that define us, and this week felt like that. There wasn't just one story, not just one announcement. I want to talk about two different stories, two different announcements, but they're connected.

Right? And when you look back and you look at the bigger picture, you realize that they're on the same plane. We crossed a line this week in terms of AI in our lives.

The first one comes out of cybersecurity. The other one deals more with the broader implications of AI in society and the infrastructure. And as I said, they both point to this new reality that artificial intelligence is very real, and it is going to have this profound impact we've all been saying.

I know a lot of us have been saying, "Yeah, I don't-- We'll see it. " But no, it's real. Let's first start in on the security side because it's something that's near and close to me.

The entire cybersecurity industry has been built on the fact that we have code that's insecure. And the cybersecurity industry is built to defend against insecure code. We find insecurities, and we fix them.

We try to stop people from taking advantage of these bugs. All software has bugs. We've known this.

It always has. Will it always have it or always will? I don't know.

But those bugs became vulnerabilities, and finding them required a special class of bug hunter. Smart people, highly skilled people. I've met so many of them over the years at Black Hat in the cyber industry.

They dig through code. They look for weaknesses. We had fuzzing, all kinds of scanning.

But there was a limit. There was a governor on the system. There was a human limitation built in.

People could only generate so much code. We had 20,000, 40,000 coders. How much code were they generating?

And then code can only be scanned at the direction of people. How much scanning can we do? And so that put brakes, it put gravity, inertia into the system.

But AI has come in now, and AI just blew the governor off. It turned the brakes into a flywheel. Right?

And we saw tools like Mythos demonstrated this week that can analyze enormous code bases. They could reason about how software behaves and identify vulnerabilities at a scale no human team could match. So think about that for a second.

Instead of one vulnerability at a time being founded by a researcher, you have machines at AI scale now scanning millions, billions of lines of code continuously. The bottleneck, as I said, the brakes have been removed. It's a flywheel.

It's not about finding vulnerabilities anymore. It's about fixing them before someone else exploits them. And we can't possibly fix them at human scale when they're being found at AI scale, and that's the problem.

It's a math problem. It's asymmetry. The attacker only needs to get it right once.

The defender has to get it right every time. We've known this in the cyber world. AI just blew that gap into a mile-wide road, super highway.

Machines discovering vulnerabilities faster than organizations can patch them. And this applies, and Mythos and Project Glasswing coming out of it, it applies to every operating system that they tested, every browser they tested. It's going to apply to our banking systems, our healthcare systems, our energy grids, our transportation, probably the nuclear codes that govern our nuclear weapons.

You want to talk about critical infrastructure, it's allIn play right now, because it all runs on software, and all that software has bugs. And if those bugs start getting discovered at machine speed, we've got to defend against it at machine speed, and that's going to take a reckoning, a complete revolution. Everything we've...

Not everything, but 80% of what we're doing in security just became meaningless. And that's why some people are starting to compare this moment to something like a Q Day scenario. Not because encryption suddenly broke overnight, which will happen in a few years with quantum, but because the discovery of weaknesses just got supercharged.

And that's this first story. And eventually we're going to be better for it because it's going to force us to make better code. But this may be a case of the cure worked, but unfortunately the patient died.

And I'll be talking more about that. Just did a podcast that's coming out today with Rich Mogel and Mitch Ashley on it. " But let me hop over now to the second thing that happened.

And that was coming out of OpenAI. And look, I've said it before, Sam... Oh, boy.

Sorry about that. Sam, Slippery Sam, as we call him, over at OpenAI. Look, he's a self-serving individual, there's no doubt about it.

He's one of these tech bros. He's done pretty well for himself and his company. But putting that to the side, this new deal for AI that they came out with has, regardless of the motives, there's a lot of good in there.

You know what? We've got billions of dollars flowing into AI infrastructure, massive data centers, compute clusters with energy grids and building more energy capacity into our system, water cooling, jobs, concrete, chips. The scale of investment we're seeing is like something either we've only seen during war or maybe during the New Deal.

It's happening so fast, trillions of dollars. But it's not just about the money here. These systems that we're investing in, these AI systems, what are they capable of doing?

We've already proven really well that AI can write code, and it's getting better at writing code every day. AI can analyze data. It can draft documents.

It can generate marketing contact. AI can also assist doctors and lawyers and engineers. What can't it do?

And that means something very important. Of course, intelligence itself is starting to become a utility. Just think about that for a second.

Intelligence itself is starting to become a utility. For most of human history, cognitive work, real cognitive work, was limited to human brains. We created tools to help us with physical work, but cognitive work was human.

We learned how to use levers and stuff like that, but we learned. We came up with the idea, then we used those tools. But now we're entering a world where intelligence can be scaled by machines.

And I'm not even talking at this point about super intelligence or AGI. I'm talking about just the intelligence that today's AI has. And when this happens, the structure of our economy changes.

Entire industries are being reshaped in AI time, in days and weeks, let alone months and years. Some jobs are going to disappear. I held out against that.

I thought if people embrace the AI, their job wouldn't disappear. There'd just be less of them, maybe. But no, some jobs are going to disappear.

A lot of jobs, a lot more that disappear, will evolve. I still believe that. And I think completely new categories of work will appear.

It's going to take time, like I said before about security. The patient may die on the table, but eventually we'll cure the disease. The last time we saw this kind of economic disruption at this scale was probably the Industrial Revolution.

Some people call it the Gilded Age. Industrialization transformed the economy, starting with the steam engine to electricity. The disruption that's happening now in the digital world, in knowledge work, decision-making, problem-solving, it's akin to those days.

I really believe that this week in the year 2026, we're more than a quarter of the way into this century, is the year that defines the rest of this century, as far as I can see, as being the AI century. Everything else we do from this century on, from now on, is going to be built on this AI. That's how pivotal it is.

So remember where you were when this happened. Because when we put these two stories together, on one side, AI breaking our whole security, our whole cyber way of doing business, forcing us to finally make really invulnerable code, putting pressure on the digital trust that our whole economy is based on today, so much of what we do beyond is based on. And then on the other sideAI transforming how work itself gets done, that combination tells me something.

It tells me we crossed the threshold into this age of AI, into the century of AI, the AI century. And if you think this was big, let me tell you something, it's just the beginning. The next wave is physical AI, robotics, machines that can see, reason, and interact with the physical world the same way our agentic AIs are working in the digital world.

Shortly on the heels of that, I mentioned it before, quantum computing, and what that's going to mean, not just in breaking our encryption, but what quantum computing is going to allow us to do. If quantum matures the way some researchers expect, it's going to create another massive shift, not just in cryptography, but advanced computation, modeling, figuring out how did we come into being? How did the universe start?

How did the galaxy start? What made Earth? By 2030, guys, it's all coming, and it's coming right now.

By 2030, all of this comes out. The world's going to be a very different place. It's going to be the AI century.

So what could you do about it? You want to go climb under your desk and wait for the bomb to go off? No.

Here's the question: What do you do? Because sitting around worrying about the future's never been a winning strategy. It never is the right thing to do.

Every technological revolution, ones we've been talking about, creates disruption, but it also creates opportunity. And the people who recognize the shift, that this was the week, and they start moving into it, they tend to do a lot better than those who laggard, who sit back. So here's a few things I would suggest.

First, if you have ever thought about starting a business, this is a good time, because times of disruption is when great new companies are built. And AI makes it easier than ever to build products, to build companies, to build a business. Small teams, two people and an AI can now accomplish what once required entire companies.

You can prototype faster. You can develop faster. You can reach customers faster.

You can do business faster. The next generation of startups is going to be built on top of these AI tools. Second, if you haven't already, embrace and learn how to work with AI.

Make it work for you. Don't fight it. Trying to compete against AI, saying, "I could edit faster, I could do this better, I could write better," it's a losing game, because even if you're right today, you're not going to be right tomorrow or the day after that.

The real advantage is knowing how to guide these systems, how to use and leverage these systems, verify their work, integrate them into real workflows, and be the master of your own. Be the master of your own workflow. Excuse me.

People who know how to do that will be incredibly valuable. Third, double down on the human stuff. What makes humans unique?

What brings the value? It's our judgment. It's our creativity.

It's our leadership. It's our trust. It's the spark of humanity.

In a world filled with machine-generated content, the ability to decide what matters becomes incredibly important. Fourth, stay adaptable. This is not a one-time disruption.

As I said, we're going to see wave after wave after wave. And it's coming in waves. I mentioned physical AI's coming, quantum's coming.

I can't see what comes beyond that, but there will be. And the people who succeed will be the ones who stay curious and keep learning. Closing.

" You'll have the answer. Because the real question is what that answer will be. Do you seize the moment?

Did you miss it and become a victim? The future's here, ready or not, and that's what Shimi says this week. Good luck.

We'll talk next week. Shimi says, Shimi says, Shimi, Shimi, Shimi says, ask me almost anything!