The AI Minute: When Strategy Gets Outdated After Lunch

Shimmy says, shimmy says, shimmy, shimmy, shimmy says, ask me almost anything. Hey, guys, it's Shimmy, and I'm back. And I got to tell you, it's good to be back.

For almost a month now, I've been out on the road. Mountains, airports, little towns, big cities, hotel Wi-Fi, cafe Wi-Fi, no Wi-Fi, almost Wi-Fi, some cell service, one bar, three bars, 3G, LTE. Sometimes I had five bars, but still nothing loaded.

You know I usually know what I'm talking about, right? It's crazy. But I'll tell you something I learned doing all of this.

That being away from my office and my comfortable surroundings where I'm always plugged in, it did something to me. It made me disconnect, not from the world, per se, because I was at some great places in Italy and Croatia and Austria and Prague, and just all over the place. So I didn't disconnect from the world.

In fact, I saw more of the world than I've ever seen. It was an interesting thing, too, because when I saw groups of people go by, whether they were talking Czech or German, or Croatian or Italian, or whatever, I didn't understand necessarily everything they were saying, but I'd always catch ChatGPT, AI, ChatGPT. No matter what language it was, people were still somehow talking about ChatGPT and AI.

I was checking in when I could, reading when I could, catching up when I could, watching the headlines. But trying to follow those stories when you're not plugged in almost constantly, quickly gaps have developed there. Right?

At Techstrong, when I looked in, yeah, we were still covering our stuff, but I wasn't plugged in, and I felt so isolated. It was really a different feeling for me than I've had in a long time. As I said earlier, when I'm plugged in at my desk, I might as well be jacked in like Neo in "The Matrix" or something.

I've got my feeds, my Slack, my email, my alerts, notifications, my sources, people calling me, texting me. There were a lot of times I didn't have any of that when I was on the road here. It's like, the car's driving, but you don't got your hands on the wheel, and this is not a Tesla either.

When you don't know what it's like till you're on some remote mountaintop with cows in Sicily, and you're trying to load an AI security story, and I only got one bar of service, no Wi-Fi, and you realize you're not plugged in. The world's moving on without you. I'm hanging on by my fingernails trying to figure out what's going on there.

But then when you do connect again, it's like a shot of drugs or something because, wow, you feel connected. When I finally got off the plane in New York on the way home, and we stopped in New York for a day or two to go see The Boss, Bruce Springsteen, I really almost felt like a Okie from Muskogee, as they used to say, getting off the bus at the Port Authority Bus Terminal in Times Square. Man, the lights were flashing, the horns were blowing, people moving in every direction.

" And to be honest, you know what? For a lot of you, this is what AI feels like right now. Every day, we're walking in, or every other day, it seems like we're walking into a new setting.

You blink, and the story changes. You go away for a few days, and Mythos goes from vulnerability apocalypse to zero-day factory to a defender's coalition. Now it's a policy question.

Tomorrow and today, it's national security, and maybe at the end of the day, it's going to just make our software get more secure, and all's well that ends well, right? You miss a week, agentic AI went from being a lab experiment to slideware, to workflows, identity problems, payment rails, governance issues, and companies trying to figure out what happens when the software just doesn't answer the questions. 0 and laying off 30% of their workforce.

It's mind-boggling how fast this stuff is going. You get on a plane, data centers are all about chips. You land, get off the plane, the data centers are about getting power or enough water.

You have dinner, and it's about permits. You go, Wall Street debt capacity, geopolitics, whether your neighborhood even wants a data center next to the soccer fields. And behind it all lurking is quantum, like some next band at CBGB's back in the day.

Wake me up when quantum comes. A little bit of a Green Day song in there if you're paying attention. But seriously, guys, this is the thing that really hit me while I was away this trip.

It's not just that AI is moving fast. Technologies always move fast. I have been in this business a long time, as you probably can tell looking at me.

By the way, you like my Martin Scorsese look today? How's that grab you? But I've been around long enough to know and know what fast was.

When PCs were coming out, we were moving from DOS to Windows, command line to graphical interfaces. It felt fast. When the internet came out, even that noise of the 12-board modem, that felt fast.

" Open source and how it kind of took over felt really fast. But what we're seeing today is much different. This is more like a car going down the highway.

It'sBut it's not just the car going down the highway. The highway's moving, too, it feels like. Literally, the earth under your feet is moving.

The road's moving. The signs are moving. Everything.

The map is updating while you're still reading it. The GPS just told you to make a left turn, and now there's a building in the middle of the road that wasn't there yesterday. I call all this, it's an AI minute.

This is what happens in an AI minute. And an AI minute is not 60 seconds, no more than a New York minute is. An AI minute is however long it takes for your strategy to become outdated, and that is crunching down.

It's however long it takes for a new model to make your product roadmap look like it was written on a stone tablet by Freddy Flintstone. It is however long it takes for a security assumption to become a liability. It's however long it takes for a competitive moat to become just nothing more than a mere puddle.

But that's where we are. And let's be honest, if I'm having trouble keeping up, and I try to stay on top of this every single day, it's what we do for a living here at Techstrong, what chance does the average business leader have? Forget the business leader.

What chance would the average security team? What about the average developer? The board member?

Just what about the average person? They're trying to plan their life. We're in work with budget cycles.

Who can do a cycle when things change day to day, minute to minute, hour to hour? You got to check you do something in the morning. You got to check in again after lunch to make sure it's still relevant.

That's the AI time crunch. That's the AI minute, and it's hitting us everywhere. Whether we're talking corporate strategy, security, infrastructure, power, chips, talent, governance, regulators, they're all crunched.

The human brain, definitely crunched. We used to talk about digital transformation. We have digital CXO, like it was a journey.

Remember that? A journey. Pack a bag, make a roadmap, have a two-year plan, get all the stakeholders aligned, everything lined up, hire a consultant or two, maybe bring in a service integrator.

Hey, we're going to make a maturity model. Everybody loves a maturity model. It's all gone.

Right? Even the beloved maturity model. Nothing says urgency like a five-stage maturity model and a 62-page PDF.

But AI doesn't give a rat's you know what about your maturity model or your PDFs. AI's not waiting for your steering committee. AI's not waiting outside the conference room with Danish and a name badge, right, until procurement signs off.

Those days are over. It's already been brought into the business. You want to call it shadow AI?

Fine. It's already in the code. It's in your supply chain.

It's in your closet. It's in your servers. It's in your data center.

It's in the SOC. It's in the developer's IDE, the intern's browser. It's probably in your wife's phone as well, or your husband's phone, or your significant other's phone.

It's definitely in your competitor's products. It's in your kids' homeworks, too. Homework.

It's in the workflow that the CI blessed, or maybe he didn't, but it's still in there. And that's what makes this so wild, folks. The adoption curve is not top-down, it's not bottom-up, it's not middle-out.

It's leaking in from everywhere. It's 360 degrees coming in from the floors, the ceilings, the walls, internally, externally, and everywhere in between. It's invasive.

We're moving now, not just from using AI, but for many of us, we're becoming AI supervisors. AI's doing the job, we're supervising it. And that's a big jump.

It's a jump a lot of people don't realize and aren't equipped for. Asking AI-- Excuse me. Something like ChatGPT, generative AI, to write a summary is one thing.

Letting an agent book meetings for you, pull data, write code, open tickets, make payments, or trigger infrastructure changes, that's a horse of a different color. That's not a chatbot. That's a worker.

And that's what we're dealing with. We have digital workers. I was up in Orlando yesterday for KnowBefore's conference, KB4.

That's what they spoke about. We've got to be managing our digital workforce, and they don't mean people. They may not be a full replacement for you, but they're doing a lot of work.

They're coworkers. They have access, permissions, and context. They have intent, or at least it seems like they have intent.

And that means that every old problem is coming back and it's wearing a new hat. Identity access, auditability, policy, security, accountability. Who told the agent to do that?

What data did it use? What system did it touch? What did it change?

Who approved it? Can we roll it back? Can we even explain it?

This isn't some bad science fiction play. This is our lives right now. This is what you're living through, I'm living through, we're living through.

And that's just Tuesday. Then there's infrastructure. Everyone wants AI.

Every board wants AI. Every CEO wants AI. Every vendor says they have AI.

Every investor wants an AI investment. Every customer wants AI features. Great.

But where are you going to run it? No one wants it in their backyard. No one wants it near them.

I'm not paying extra electricity. I don't want to give any tax breaks. And guess what?

Those AI data centers don't make no jobs. Where are we getting the cooling, the power, the network? Where's the budgets coming for all this?

It's not a magic trick, guys. This all has to be figured out. AI isn't a software story.

If it ever was, it certainly isn't anymore. It's a power story. It's a semiconductor story.

It's a real estate story. It's a supply chain story, a geopolitics story. It's a Wall Street story.

It's your story. It's my story. And you want to understand AI?

Well, you got to follow the GPUs, you got to follow electrons, follow water, follow the debt, because the truth's hiding in between and under and above all of that. And layer security on top. Don't forget about the security.

Because now we have AI finding vulnerabilities faster than humans can triage them. We have AI writing code faster than humans can review them. AI generating attacks faster than defenders can model them.

AI helping defenders, yes, absolutely. But also helping attackers, yes, absolutely. So let's not kid ourselves.

You see where this is going. Acceleration is all around it, all around us. That's why this Mythos story and with Glasswing and everything, it all matters.

Security by design, it matters. Software supply chain matters. Platform teams, they matter.

Because if AI can compress the time in all of these things it takes to find a flaw, exploit it, and make you pay, then the only sane response is to compress the time it takes to fix one. You cannot fix machine-speed discovery with human-speed bureaucracy. Human-speed bureaucracy is gone.

You cannot fight agentic AI attacks with a weekly change advisory board meeting. Like those things ever worked. You can't fight AI native threats with PDF policies nobody reads.

No offense to PDF policies, but when did you read the last one? I think some of our best friends are PDF policies. But here's the thing, and let's get real.

It's like an Einstein theory of relativity thing. I wrote about it before I went away, the theory of time dilation, that if you're going closer to the speed of light, time may feel normal to you, but someone who's not traveling that fast, years, decades might have gone by in what you perceive as minutes. I know it feels freaky, and I'm not a physicist, and I can't give you the math behind it.

Look, I went to law school. I've been in tech 30 years. I'm a media guy as a profession.

But I'm just a guy from New York. But I feel this in my bones. I feel it in my muscles.

I see it in my eyes. If you're looking for a formal explanation of how time can be compressing like this, go ask Einstein or some other dude at a chalkboard with hair that looks like they stuck a fork in an outlet. But conceptually, I'm telling you, this is what I'm feeling.

And as my wife always tells me, "Don't tell me, or don't deny how I'm feeling. " Companies using AI well are operating in one time zone. Companies not using AI well are operating in a whole different time zone.

And that gap keeps widening. The people who have embedded AI into how they research, write code, test, analyze, sell, support, and decide are experiencing time differently. They are doing more cycles, more drafts, more tests, more experiments, more decisions, more learning.

The company's still treating AI like a side project or watching from the sidewalk. And pretty soon they're going to be watching from the peanut gallery. Because that gap, as I said, is widening every day.

And if that makes you uncomfortable, do something about it. The answer to AI moving too fast is not to ask AI to slow down. It's not.

The answer is not to romanticize your old pace or how good things the good old days used to be. Those days are gone, too. " Settle down?

My friends, this ain't some snow globe. This isn't the beach down here in Boca in the summer and when it heats up and the hurricanes are starting. Nothing's settling down.

So what do you got to do? What do I have to do? What do we have to do?

We got to learn to move faster. I'm not saying be reckless. I'm not saying be stupid.

But give the keys to the robot and hope for the best, that's probably not an answer either. It's not a strategy. That's a lawsuit with a name waiting to happen.

But we need to move faster. We build AI how we keep up with AI. I know what it sounds like.

It sounds like you've been out on a bender last night and someone just gave you a shot and a beer, calling it the hair of the dog. And maybe it is, but we need AI to do AI or keep up with AI. Sometimes the things that get you in the mess are the very same things that let you survive the mess.

You can't manually monitor the AI landscape anymore. I know. " We can't even put the human in the loop anymore.

We need the human at the helm. You need AI to help you summarize these things just to keep that human at the helm. We need AI to help us compare, help us detect patterns, help us test assumptions, analyze risk, review code, watch infrastructure, find the signal in the noise.

Because that noise isn't going away. It's getting louder. The feeds are not slowing down.

They're getting faster. The breakthroughs out there are not waiting for your return flight. The models are not asking if you had a good night's sleep.

I mentioned quantum. Quantum's somewhere just in the green room behind us warming up. As I said before, wake me up when quantum comes because that's going to just jolt this whole thing even faster.

So here's Shimmy's take. The next advantage is not going to belong to the company with the biggest AI press release. We've seen enough of those.

It's not going to belong to the company with the most pilots or the most agents or that can say agentic the most times in an earning call. The advantages are going to belong to the people and orgs that learn how to operate inside this AI minute time dilation bubble. Shorter loops, better judgment, cleaner governments, more automation, more human oversight when it matters, less theater, more muscle.

It means admitting that no one can help keep up alone anymore. Not me, not you, not any of us. Not even the smartest person in your company, not your analyst firm certainly, not your board, not your vendor.

Nobody. The only way to keep up with AI is to use AI, manage AI, govern AI, and keep your head screwed on straight as best you can while you're doing all this. Because it's not slowing down, as I said.

And if you're standing in the middle of Times Square looking around like you just got off the bus from nowhere, welcome to the club. Just don't stand there too long and don't look up at the tall buildings. That's a dead giveaway.

In an AI minute, the light changes fast. And as I said, I was in New York. If you don't move when the light changes, somebody behind you is going to let you know and just walk over you.

I'm Shimmy. That's what Shimmy says this week. I'll see you next time, assuming the world doesn't change three more times before I finish uploading this.

Have a great week. Shimmy says, Shimmy says. Shimmy, Shimmy, Shimmy says.

Ask me almost anything.