There is an infrastructure shift happening and most businesses and executives still don’t realize it.
Matt Shumer, CEO of OthersideAI, wrote in an article recently something that me made stop and re-read it: “We’re in the February 2020 moment for AI.”
If you remember February 2020, you know exactly what he means. You remember the week when remote work shifted from “nice to have” to “figure it out by Monday or close your doors.” You remember watching businesses that hesitated get left behind — not because they were bad at what they did, but because the infrastructure underneath everything had fundamentally changed overnight.
When I read his quote, my honest reaction was: he nailed it.
And most people still don’t realize it’s happening.
Here’s the thing that gets me: I’m completely immersed in AI. I’m part of an AI cohort that works tirelessly to keep up with what’s happening. We meet weekly. We share discoveries. We test tools. We push capabilities.
And all of us, every single one, still feel behind.
That’s not hype. That’s the reality of how fast this is moving.
The learning curve is growing steeper every single day. We’re approaching the point where waiting has diminishing returns that compound into real operational debt. And yet most business owners are still thinking AI is something they should learn “one day.”
That one day already passed.
What Actually Changed (And Why Most People Missed It)
Eighteen months ago, I took my very first AI course.
It covered the basics of ChatGPT and LLMs and what AI could do, how to write prompts, basic use cases. Standard stuff for early 2024.
Within that year and a half, AI capabilities 10x’d. Then 10x’d again.
The things we’re building now — the workflows we’re creating, the systems we’re implementing — would have seemed like science fiction in that first course.
In my weekly AI cohort, someone discovers, learns, and expands on a new tool or capability every single week. Not because there’s one breakthrough tool that changed everything.
Because people are seeing things differently.
They’re realizing opportunities that exist when you stop thinking about AI as “another tool to learn” and start thinking about it as foundational infrastructure.
Let me show you what that actually looks like.
The Day I Stopped Researching and Started Building
Before:
I used to spend hours, sometimes entire days researching for client projects.
I’d scan websites looking at competitor designs. I’d scroll Instagram constantly, following dozens of accounts, tracking trending fonts and color palettes. I’d bookmark layouts, save inspiration, screenshot ideas. I’d spend hours just maintaining a knowledge base so I could give clients designs that felt current.
It was exhausting. And it never felt like enough.
After:
Now I approach projects completely differently.
I have AI research the actual brand I’m working with. The actual people who would buy from that brand. Their ideal clients’ pain points, their language, their frustrations, what they’re searching for at 2am when they can’t sleep.
We reverse-engineer a marketing plan based on what actually speaks to them — not based on what went viral last week, but based on patterns across hundreds of websites, thousands of social posts, dozens of competitor strategies.
AI can analyze all of that in the time it used to take me to scroll Instagram for an hour hoping to spot a trend.
That’s not a tool making my job faster.
That’s infrastructure changing how I think about the job entirely.
The difference is foundational. One is “AI helps me work faster.” The other is “AI changed what work actually means.”
And that shift? That’s what most people are missing while they wait to “learn AI one day.”
Your Business Systems Just Became Legacy Infrastructure
This isn’t about whether you use ChatGPT to draft emails.
This is about recognizing that your current business systems, the ones you built over years, the ones that “work fine” are now legacy infrastructure.
They were built for a world where humans did everything manually.
Where you spent hours researching competitors instead of having systems analyze them in seconds.
Where you created content from scratch every single week instead of having frameworks that compound.
Where you used Gmail search to find that email from six months ago instead of having organized systems that surface information when you need it.
Those aren’t personality flaws. They’re infrastructure problems.
And the businesses winning right now didn’t fix those problems by working harder.
They rebuilt the infrastructure entirely.
What You Give It Is What You Get Back
People say:
“I tried AI. It doesn’t get it right.”
“It’s only part way correct.”
“It gives me generic responses.”
“It doesn’t understand my business.”
When I hear that, I immediately think: What you give it is what you’re going to get back.
That includes:
- The information you provide
- The scope of the task
- The context you give
- The history you’ve built
- The knowledge you’ve shared
- The questions you ask
AI will tell you exactly what you want to hear via the quickest route… unless you push it.
Question it. Give it objections. Provide more context. Push past the first surface-level answer.
That’s when it gets powerful.
And here’s what people selling prompt libraries don’t tell you:
The best prompts mean nothing if you haven’t trained the AI on your business.
You need context. You need history. You need to teach it about your work, your clients, your problems, your specific operational reality.
A $97 prompt library is useless if the AI doesn’t understand what you’re actually trying to accomplish.
The Hallucinates Objection (And Why It Misses the Point)
“AI is dangerous. It hallucinates. It gives wrong information. It lies to you.”
Valid concerns. Absolutely.
But here’s the reframe:
You wouldn’t have your intern write something and use it as a legally binding document without reviewing it, right? Same with AI.
You use AI to expand your thinking. Generate options. Speed up research. Automate repetitive tasks.
Then you use your human knowledge to make it better, stronger, complete.
AI is a tool, not a replacement.
The businesses scared of AI “taking over” are missing the point entirely.
AI isn’t here to replace you. It’s here to eliminate the parts of your work that shouldn’t require you in the first place.
The Real Cost of Waiting (It’s Not Money)
If someone reads this article six months from now, everything will have changed.
Every platform will have new capabilities. Every AI program will be more complex, smarter, better, stronger.
And every person who’s learning it today; jumping in now, figuring it out as they go, will be that much closer to understanding how to leverage it.
But here’s the part that should terrify you: The gap isn’t just about technical skills.
It’s about complex thinking. Planning. Strategizing.
People who embrace AI infrastructure now will surpass those who wait, not just in work efficiency, but in how they think about problems, how they plan solutions, how they strategize growth.
This is the same pattern we saw with the internet.
We were all scared of it. We worried about what would happen if it got too big, too loud, too uncontrollable.
And here we are, doing the same thing with AI.
Except AI is already here. It’s already integrated into how business operates.
It’s up to us to learn how it works, take advantage of the opportunities, and position ourselves to evaluate, respond, react, and strategize around the problems that will inevitably come with it.
The Learning Curve Gets Steeper Every Day
I took my first AI course eighteen months ago.
People who start today are already behind where I was then — not because they’re slower learners, but because the baseline has moved.
And the baseline will move again next month. And the month after that.
In my AI cohort, people who joined six months ago are now helping people who joined last month. And the people who joined last month will be helping the people who join next month.
But everyone agrees on one thing:
Waiting makes it harder. Not easier.
There’s no “dust settling” moment coming. There’s no plateau where things slow down and you can casually catch up.
The capabilities are endless.
I hear wins from the cohort every week on things people are discovering, implementing, learning, building. The opportunity is endless.
It’s not about one specific tool or one capability.
It’s about seeing things differently.
It’s about realizing the opportunity that exists when you think differently about these tools.
Where to Actually Start (Not What You Think)
This isn’t written to scare you or make AI feel intimidating.
It’s to create awareness and curiosity about what it can do for you.
And to make clear that this isn’t some massive intimidating thing you need a computer science degree to understand.
You just need to jump in and start.
Here’s what NOT to do:
- Don’t start with free tools
- Don’t buy someone’s prompt library
- Don’t watch 47 YouTube tutorials before doing anything
Here’s what TO do:
Create a paid account on a platform of your choice — ChatGPT Plus, Claude Pro, Perplexity Pro. Pick one.
Then ask it hard questions. Not questions that you would google or that your previous “I asked ChatGPT” version of yourself would ask.
Give it hard tasks. Make complicated requests. See what it can do for you.
Don’t ask it to write your social media caption.
Ask it to analyze your last ten social posts, identify patterns in what performed well, reverse-engineer why, and suggest a content strategy based on actual data instead of guesses.
Don’t ask it to draft an email.
Ask it to analyze your client communication patterns, identify where you’re spending the most time, and suggest automation workflows that maintain your voice while eliminating repetitive work.
Push past the surface.
Because that’s where the real infrastructure shift happens. Not in using AI to do what you’re already doing faster.
In using AI to question whether you should be doing that task at all.
The Infrastructure Question You Actually Need to Ask
The real question isn’t “Should I learn AI?”
The real question is:
“Is my business built for the infrastructure that now exists, or am I running 2019 systems in a 2026 world?”
Because that’s what this moment actually is.
Not a tool adoption moment.
An infrastructure moment.
And infrastructure doesn’t get rebuilt with YouTube tutorials and free trials.
It gets rebuilt with strategic thinking about how your entire operation functions; and then systematically upgrading each piece.
What Happens Next
You’re in the February 2020 moment for AI infrastructure.
The businesses that recognize this and act will thrive.
The businesses that wait for the dust to settle will find themselves so far behind that catching up becomes nearly impossible.
I’ve rebuilt businesses with AI-native infrastructure. I’ve seen what’s possible when you stop bolting AI onto old systems and start building new systems from the ground up.
If you’re reading this and thinking “I need help with this”… I’m here. I work with established businesses to rebuild their operations with AI-native infrastructure, clear messaging, and systems that actually work together.
But honestly, whether you work with me or not doesn’t matter as much as you taking this seriously.
Start now. Jump in. Ask hard questions. Build context. Push past surface answers.
The infrastructure shift is happening whether you’re ready or not.
The only question is whether you’ll be running on it — or buried by it.
Tay Design Co helps service-based businesses build AI-native infrastructure that integrates clear messaging, operational systems, and visibility in AI search. If you’re ready to rebuild your business for the infrastructure that now exists, let’s talk.
FAQs
Q: What does “February 2020 moment for AI” mean?
Remember when businesses had 48 hours to figure out remote work or lose clients? That wasn’t about learning Zoom—it was about your entire operational infrastructure becoming obsolete overnight. Same thing’s happening now with AI. This isn’t “learn a new skill when you have time.” This is “the infrastructure you built your business on just became legacy tech.”
Q: Do I need a computer science degree to use AI for my business?
A: No. You need strategic thinking, not a computer science degree. Get a paid account (ChatGPT Plus, Claude Pro, or Perplexity Pro), then stop treating it like a search engine. Ask it to solve actual business problems. Have it analyze your client retention data, reverse-engineer competitor positioning, or build you a content framework. The gap between people who “use AI” and people who leverage it strategically is what they ask it to do.
Q: Why won’t free AI tools or prompt libraries work?
A: Free tools have capabilities capped at surface level, and prompt libraries are templates without context. The best prompt in the world means nothing if you haven’t given AI your business reality—your client language, your operational constraints, your strategic positioning. What you put in is what you get back. Generic inputs create generic outputs.
Q: Isn’t AI dangerous because it hallucinates and gives wrong information?
A: AI hallucinates the same way an intern might fill gaps with assumptions when they don’t have complete information. Use AI to eliminate repetitive thinking, generate strategic options, accelerate research, and automate documentation. Then apply your expertise to evaluate, refine, and finalize. AI doesn’t replace judgment—it removes the work that never needed your brain in the first place.
Q: What happens if I wait six months to start learning AI?
A: The baseline keeps moving up, which means the entry point keeps getting higher. Someone starting today is already behind where early adopters were 18 months ago—not because they’re less capable, but because capabilities have multiplied exponentially. The gap isn’t just technical skills. It’s how you conceptualize problems, architect solutions, and structure growth. Waiting doesn’t make this easier. It makes the learning curve steeper and the competitive disadvantage wider.
Q: What’s the difference between using AI as a tool vs. AI as infrastructure?
A: Using AI as a tool means “ChatGPT helps me write emails faster.” Using AI as infrastructure means fundamentally changing how you approach work. Example: Instead of spending hours researching competitor designs, you have AI analyze hundreds of websites, identify patterns, reverse-engineer marketing strategies, and surface insights in minutes. Infrastructure shifts change what you’re capable of accomplishing, not just how quickly you accomplish the same tasks. That’s not faster work—that’s different work entirely.
Q: Where should I actually start with AI implementation?
A: Don’t start with free tools, prompt libraries, or YouTube tutorials. Get a paid account on one platform. Then stop asking it to write Instagram captions or emails and start giving it real strategic work. Have it analyze your last quarter of client calls to identify patterns in objections. Ask it to reverse-engineer your most successful client’s buyer journey and create a replication framework. Request a content audit that identifies which topics drive engagement vs. which drive conversion. Push it beyond surface-level tasks to discover what infrastructure-level thinking actually looks like.
