The Meeting That Changed How I Think About AI Readiness
Today I sat in a room with my teammates, and we did something we've never done before.
We didn't install a new tool. We didn't watch a demo. We didn't sit through a vendor pitch about how AI will revolutionize our workflow.
We picked one process—just one—and we took it apart.
Here's the backstory. We have a document library. SOPs, wikis, reference guides—the kind of thing every team accumulates over years. And if I'm being honest? It's a gigantic mess. Multiple versions of the same document. Outdated procedures nobody's retired. Files that are technically findable if you already know exactly where to look and what they're called.
Everyone knows this. Everyone's known it for a while.
But today, we finally did something about it. Not because someone sent a memo about "document management best practices." Because we're getting ready to work with AI—and AI forced us to look at what we'd been avoiding.
What AI Actually Exposed
Here's the thing nobody tells you about AI readiness: the first barrier isn't learning prompts. It isn't picking the right tool. It isn't even getting access behind your corporate firewall.
The first barrier is that most of us don't actually understand our own processes well enough to explain them.
We do our work. We do it well. We've been doing it for years. But ask us to describe, step by step, what we actually do, why we do it that way, what the constraints are, what success looks like? That's where it gets uncomfortable.
AI didn't create this gap. It just made it impossible to ignore.
Because here's the reality: AI can't help you improve a process you can't describe. It can't automate a workflow you haven't documented. It can't standardize something that exists in five different versions across three people's desktops and someone's memory from 2019.
The Day We Became Engineers
So today, for the first time, we approached our work like engineers.
We chose one process. We looked at the smaller pieces of the workflow. We started asking questions we'd never formally asked before: What are the actual steps? What are the constraints? What does a good outcome look like? Where does this break down? What are the specs?
We didn't touch AI once. And yet this was the most important AI preparation work we've ever done.
Because before you can work *with* AI, you—the human—need to understand your own processes deeply enough to explain them. You need documentation. You need standardization so everyone follows the same guidelines. You need clarity about what success actually means.
That's the prerequisite. And most teams haven't done it.
This Isn't a Skill. It's a Shift.
I need to say something that might sound uncomfortable: if you're not interacting with AI today—in any shape or form—you will have a hard time adapting to what's coming.
Not in five years. Not in some theoretical future. The change is happening right now, and within a year or two, working with AI will be standard procedure. AI assistants will be embedded in our daily work. This won't be optional.
But here's what makes this different from every other technology shift you've lived through: this isn't just about learning a new tool. It's not like learning to use the internet, or picking up a new software platform, or figuring out yet another project management app.
Working with AI requires changing the way you think.
AI exposed our deficiencies—in how we work, how we organize, how we communicate about what we do. And to work with AI effectively, we need to become something we've never formally been asked to be: builders. Engineers of our own workflows.
We need to be able to take our processes apart, examine each piece, understand the requirements and expectations, and put them back together in a way that's better. That's not a skill you learn from a tutorial. That's a fundamental shift in how you approach your work.
The Part That Scares People (And Why It Shouldn't)
I know what you might be thinking: *I didn't sign up to be an engineer. I'm an analyst. A coordinator. A manager. A specialist.*
But think about what my team did today. We didn't write code. We didn't use any technical tools. We sat in a room and had a structured conversation about a process we all touch.
That's engineering. That's what it actually looks like for knowledge workers.
Taking a messy document library and asking: *What do we actually have? What's current? What's redundant? How should someone navigate this?* That's systems thinking. And it's exactly the muscle you'll need when AI becomes part of your daily toolkit.
The people who will thrive aren't the ones who memorize the best prompts. They're the ones who can look at their own work with fresh eyes, break it into pieces, and clearly articulate what each piece does and why.
Two Things You Can Do This Week
So here's what I'm asking of you—two things, and neither requires a subscription or an IT ticket.
First: *start interacting with AI.* Today. Ask ChatGPT for a recipe. Use Claude to help you draft an email. Generate an image with Gemini. Plan a trip. Ask a question you've been curious about. It doesn't matter what—just start building the muscle of working *with* an AI, having that back-and-forth, learning how to communicate with something that's smart but not a mind reader.
This matters because working with AI is a conversation, and like any conversation, you get better at it with practice. The people who start now will have months of instinct built up by the time AI lands on their desk at work.
Second: *pick one workflow and take it apart.* Just one. Something you do regularly—weekly, daily, whatever. And ask yourself: What are the actual steps? Where does the information come from? What do I do to it? What does the final result look like? Could someone else follow these steps without calling me?
You don't need to automate it. You don't need to optimize it. Just understand it. Describe it. Write it down.
Because that clarity—that ability to see your own work as a system with inputs, processes, and outputs—is the real foundation for everything AI will make possible.
The Upgrade Isn't Given
Every job description will include some version of "works with AI" within a couple of years. That's not a prediction—that's the trajectory we're already on.
But the people who will be ready aren't the ones who wait for a training module to appear in their LMS. They're the ones who start now. Who interact with AI out of curiosity, even when it's imperfect. Who look at their own work with an engineer's eye, even when nobody's asking them to.
Today my team took one messy process and started making sense of it. No AI involved. Just humans doing the hard, honest work of understanding what they actually do.
That's where it starts.
Still taking processes apart, still finding documents from 2019 that nobody remembers creating, still convinced that the real AI revolution starts with a honest look at your own file system.
P.S. If you're wondering what happened in our meeting—we're not done. We chose one process, mapped the pieces, and now we have homework. But for the first time, we have a shared picture of what "organized" actually looks like. That alone was worth the meeting.
P.P.S. To the colleague who said "wait, we have THREE versions of this SOP?"—yes. Yes we do. And now we're going to do something about it.