Every time I think I'm beginning to understand what artificial intelligence can do, something new comes along.
A new tool. A new capability. A new way of doing something that, not long ago, seemed impossible—or at least considerably more difficult.
And I find myself wondering: How are we supposed to keep up?
Whether we're just beginning to explore AI or have considerable technical expertise, none of us can possibly know everything about technology that's changing this quickly.
I've been thinking about that question quite a bit lately, particularly because of a project I'm involved in at work.
When a Great Idea Isn't Enough
When I became involved in this project, I learned that an earlier AI hackathon had produced two winning solutions to the same business problem.
Both had merit. People had been enthusiastic about the possibilities. Teams had formed, ideas had taken shape, and some development had begun.
But the projects had stalled.
There were reasons for that, of course. Yet what interested me was the difference between demonstrating what technology could do and turning that possibility into something people could actually use.
The hackathon had shown what was possible. But getting from an exciting concept to a finished application still required something familiar: clear requirements, agreement about what we were building, investment, decisions, coordination, and people willing to carry the work through to completion.
AI hadn't eliminated any of that.
My role in the project is as the business owner, helping define what we need to accomplish, working through detailed requirements, and collaborating on the user experience so people can visualize and evaluate what we're building. We also have a project manager coordinating timelines, responsibilities, and the many moving pieces.
And this isn't just a conversation between me and an engineer. We're also working with Finance, a database architect, and data science teams, each bringing expertise to different parts of the problem. The technology may help us work faster, but it doesn't eliminate the need to understand how those pieces fit together.
Our engineer brings another kind of expertise. He understands the technical architecture, how different systems need to connect, and how to turn those requirements into a working application.
And AI is changing his work, too.
Coding that might once have taken weeks or even months could, in some cases, be completed in days or weeks. That's remarkable. But he still needs to determine how to use the technology appropriately, estimate the work, and ensure that what he builds accomplishes what we've agreed upon.
Meanwhile, we're examining requirements line by line, developing mockups so stakeholders can see what they're being asked to approve, and working through how information from different systems will come together.
We're also using technology to make some of that work easier. Meeting transcripts help us revisit discussions and catch details we might otherwise miss. AI can help us evaluate information, explore alternatives more quickly, and create visual presentations that make it easier for people to see what we're proposing and contribute their ideas.
But faster doesn't necessarily mean finished.
There are still decisions to make, people to bring along, changes to manage, and details to get right.
In some ways, I'm finding that the faster the technology moves, the more deliberate we need to be about what we're building.
How Much Do We Really Need to Know?
Something else happened during this project that made me think about what it means to be knowledgeable when technology changes so quickly.
People were discussing a knowledge graph—a way of organizing information by connecting related things and their relationships, which AI systems can use to understand context and make connections between information.
I noticed that people seemed to be using the term in different ways.
Some were describing what sounded like bringing information together in one place and displaying it visually. But our engineer appeared to be talking about something more technically specific.
I wasn't entirely sure what he meant.
So I did some research.
As I learned more, I began to understand the distinction between simply collecting information and representing the relationships within that information in ways AI systems can use.
I wasn't trying to become an engineer. I wanted to understand enough to participate in the conversation, ask better questions, and work more effectively with someone whose expertise is different from mine.
It reminded me of something simple.
I don't need to be an electrician to turn on a light switch.
But knowing how to turn on a light doesn't mean I understand electricity.
AI is making sophisticated technology increasingly accessible. That's exciting. But being able to use a tool, understanding what it's actually doing, and knowing how to apply it well aren't necessarily the same thing.
And that applies to experts as well as beginners.
Our engineer is learning how new capabilities change the way he develops software. I'm bringing my experience in business operations, processes, and organizational change, while using AI extensively to analyze information, explore possibilities, and accelerate parts of my work. Others involved in the project are learning what the technology might mean for the way they work.
We're not all learning the same things. Nor do we need to.
Finding Our Footing
I don't think the answer to rapidly changing technology is to try to become an expert in everything.
I do think curiosity matters. So does recognizing what we already know, what we need to understand, and when to draw on someone else's expertise.
The fundamentals of getting things done haven't disappeared simply because the tools have become more powerful.
We still need to know what we're trying to accomplish. We need to ask questions, evaluate what we're seeing, pay attention to details, and sometimes slow down enough to make sure we're heading in the right direction.
The hackathon demonstrated what was possible. The work that followed is helping turn that possibility into something real.
And perhaps that's an encouraging way to think about all this change.
We don't have to master every new tool before we can begin. We can start with something that matters to us, explore what the technology can do, and use our own experience and judgment to decide what comes next.
So I thought it might be interesting to give you something to try for yourself.
[Insert editorial illustration here]
Your Turn: Try Something New
You don't have to be a technology expert to experiment with AI. In fact, you might discover something interesting by starting with an activity you already know well.
Think about something you do regularly, enjoy, or have experience with. Perhaps it's gardening, cooking, volunteering, organizing family gatherings, writing, or working on a creative project.
What might AI bring to something you already know how to do?
Getting Started
If you already use an AI tool, you can use that for this experiment.
If you're new to AI, here are three free options:
ChatGPT: https://chatgpt.com
Microsoft Copilot: https://copilot.com
All three offer free AI conversations. You may be asked to register or sign in, depending on the service. Duck.ai is a free, privacy-focused option that you can use without registering.
Here's how to try the experiment:
Choose one of the tools above. Copy the first prompt below, replace the information in brackets with your own, paste it into the AI conversation, and press Enter.
Stay in the same conversation for all three steps so the AI can build on what you've shared.
Step 1: Start With What You Know
Copy and personalize this prompt:
I want to explore how AI might help me with something I already do.
The activity or project is: [describe your activity].
My experience with it is: [describe how long you've done it or what you know].
Three strengths or skills I bring are: [list three].
Ask me three questions, one at a time, to better understand how I approach this activity, what I enjoy about it, and what I find challenging or time-consuming.
Wait for my answer before asking the next question.
After the third answer, summarize what you've learned about me and how I approach this activity. Don't suggest solutions yet.
Step 2: Explore the Possibilities
Once you've read the summary, correct anything the AI misunderstood. Then copy and paste this next prompt into the same conversation:
Based on what you've learned about me and this activity, suggest five specific ways AI could help me do it more easily, effectively, or creatively.
Include at least one possibility I might not have considered.
For each suggestion, explain what AI could help with and what I would still need to contribute myself.
Step 3: Try Something
Choose the suggestion that interests you most. Then enter:
I'm interested in suggestion number [number].
Help me try it right now. What's one small first step we could take together?
Ask me for whatever information you need, and help me actually do that first step rather than simply giving me another plan.
Now I'd Love to Hear From You!
What did you discover? Did AI suggest something you hadn't thought about? Did it surprise you, get something wrong, or help you see a familiar activity differently?
And did you actually try that first step?
If you give this a try, come find me and tell me about your experience. I'd love to hear what worked, what didn't, and what you discovered along the way.
You might even give me a few new ideas to explore!
