Are You Still Swinging an Axe?

Paul Bunyan didn’t lose to a chainsaw. He lost because he didn’t change.

I was in Quality Assurance (QA) at Adobe for years. For those unfamiliar, QA people are the ones who find bugs before software ships to customers. Over time, I found myself on the whitebox team—the more technical crew that builds tools and automation, instead of manually clicking through the software until your brain turns to mush.

I had this manager who was not unlike a caffeinated squirrel in a polo shirt. And he was a bit obsessed with "The Fable of Paul Bunyan vs. the Steam-Powered Chainsaw." He referenced it constantly but never actually explained it. He just assumed we all knew. When our automation found a pile of bugs, he’d grin and say, “Paul Bunyan just can’t keep up, can he?” Or when we'd flood the engineering team with so many bugs they'd avoid eye contact in the hallway, he'd cackle and start mocking poor old Paul Bunyan.

If you don’t know, the fable goes like this: Paul Bunyan, a legendary lumberjack, is challenged by someone with a steam-powered chainsaw in a tree-cutting competition. Despite his strength and skill, Bunyan can’t keep up. The fable ends with Paul fading into legend, as the world moves into the industrial age.

My manager’s Bunyan obsession has lived rent-free in my head for nearly 20 years, popping up at the strangest of times. And now, watching AI tools revolutionize everything from writing to coding to design, I swear I can hear his distinctive cackle echoing through time.

AI is the steam-powered chainsaw, and we're all Paul Bunyan now.

And I’d argue that Paul didn’t lose just because the chainsaw showed up. He lost because someone else learned to use it first. In today's AI landscape, we're all standing at the edge of the forest with chainsaws suddenly in our hands. The question isn't whether to use these powerful new tools, but how effectively can we wield them?

So this week, let’s talk about prompting—what it is, why it matters, and how to wield it like the chainsaw it is.

Prompting Is the New Interface

Let’s start with the basics: prompting is simply how we tell AI what we want.

But prompting isn't about getting AI to “understand” you. It’s about you understanding what you want well enough to express it with precision.

Think of it like ordering at a restaurant. Saying, “I want food” won’t get you very far. But saying, “I’d like a grilled chicken sandwich with no mayo (cause that’s gross), and a side of sweet potato fries” gets you exactly what you’re hungry for. Prompting works the same way. The clearer you are, the better the result.

What makes prompting so fascinating is that it’s the great equalizer. Everyone has access to the same AI models now. The CEO and the intern are using the same tools. The difference lies in how clearly each person can think and communicate.

A vague prompt leads to vague results. A clear one unlocks something powerful. A good prompt is structured thinking, creative direction, and system design wrapped in language. It’s a blueprint. And like any good blueprint, its value comes from the clarity of the decisions it captures as well as the possibilities it unlocks.

After months of working with these tools, I've settled on four truths that might help you navigate this new terrain:

• Prompting reveals your thinking. Clear prompts come from clear minds. If your thinking is fuzzy, your results will be too.

• Prompting is iterative. The first version is just a starting point. It improves through back-and-forth.

• Prompting is process. You’re not just asking questions—you’re designing a system, one instruction at a time.

• Prompting is partnership. The best outcomes come when you treat AI like a collaborator, not a machine that spits out answers.

If prompting is the interface, then your prompts are the tools. The next step is making sure you don’t start from scratch every time.

Building Prompt Capital—Your New Operational Literacy

There is, I believe, a quiet revolution happening in how we work. People aren't just using AI, they are building institutional knowledge around it. The individuals and teams that will pull ahead aren't necessarily those with the best access to AI models, but those who are systematically capturing their interactions with it.

And when people start organizing what works, not just sampling or experimenting, they are building a new kind of asset.

This is what I call prompt capital: a collection of proven, reusable prompts that consistently deliver value. It's the linguistic assets you build that take one-off experiments and turn them into repeatable processes. Like financial capital, it compounds over time and creates leverage beyond your individual capacity.

My son is actually the one who inspired me to start doing this more seriously. I had been using ChatGPT for fun—writing Slack messages in a Shakespearean tone or offloading math problems I didn’t feel like solving. But one evening, he came home with a lease agreement for an apartment he and his friends were moving into. He was transitioning out of the dorms, and while I instinctively started reading through the full 25-page document, he took a different route. He uploaded the PDF and fed ChatGPT a prompt that went something like this:

Almost out of spite, I read the entire document anyway. I didn’t need to; ChatGPT was more helpful than I could’ve hoped and gave us both peace of mind. After that, I was hooked. Anytime I found a prompt that worked especially well, I stopped recreating it from scratch and I started saving them in a simple note. Over time, that note turned into a structured collection—my personal prompt capital (which now lives in Notion).

Here’s what I’ve found especially valuable from my collection:

• The Research Extraction Prompt – This one is a workhorse. I feed it long-form content like articles, reports, transcripts, and even books, and it pulls out structured insights based on my instructions. It has saved me hours. Sometimes, after finishing a book chapter, I’ll drop it into ChatGPT and get a clear summary of the key points and highlights.

• My Personal Tutor Prompt – This prompt turns AI into a surprisingly patient teacher. Last year, I decided to move my video editing projects from iMovie to Final Cut Pro. The only problem was, I had no idea how to use Final Cut. Ten years ago, I probably would have picked up a “Final Cut for Dummies” book. Instead, I wrote a prompt that turned Claude, the AI from Anthropic, into my personal tutor.

  “How do I add transitions?”
  “How do you create a compound clip?”
  “Uh…what is a compound clip??”

  Claude guided me through each step at my pace. It felt like having a teacher who never gets tired or frustrated.

• The Socratic Clarifier Prompt – This one helps challenge my thinking in a useful way. It asks thoughtful questions that pressure-test my reasoning. I turn to it when I need to clarify assumptions or get unstuck. It’s like having a smart colleague who won’t let me settle for fuzzy logic. It’s also excellent for working through ideas using first principles, a mental model I’ll cover later this month.

• The Prompt Prompt – This one helps fix other prompts. When something isn’t working as expected, I use this to figure out what went wrong and how to improve it. It’s like having a prompt editor that helps fine-tune everything behind the scenes.

What makes these prompts valuable isn't just that they work. It's that they are saved, refined over time, and easy to reuse. I’m not starting from scratch every time. I’m building on what has already worked.

And I think there’s something in this approach that could completely change how teams work with AI. Instead of everyone making up prompts on their own, you could create a shared library of proven ones. You document them. You improve them together. You share them freely. It becomes part of your workflow.

The result is prompt leverage. Not from getting it right once, but from getting smarter over time. Your ability to work with AI improves continuously rather than staying flat or being dependent on individual talent. That’s the power of prompt capital. You go from doing the work yourself to building systems that do it with you.

But even with all this potential, a lot of people aren’t excited—they’re uneasy. For many, AI still brings more anxiety than clarity.

Turning Hesitation into Innovation

I am something of the AI evangelist at my work. Each time I wax poetic about its uses, however, I get a handful of people who mention Skynet or point out how AI can make up facts. Some share articles about AI-generated art stealing from human artists. Others worry about deepfakes undermining democracy. Or they're convinced that AI is coming for their job, that their creative skills will be devalued, or that we're one prompt away from unleashing digital chaos.

And I get it. I do. These are valid concerns, and like any new powerful technology we should proceed with caution. But beneath all this I believe is a deeper kind of fear. Not just of the technology itself, but of what it means for our identities.

It's easy to think, "If this can do in seconds what takes me hours... what value do I still bring?" But I'd like to offer a different set of questions: "If this can do in seconds what takes me hours... what impossible problems could I now solve? What creative heights could I reach if the mundane tasks were handled for me? What value can I create that was previously unimaginable?"

This brings us to an essential mindset shift: AI is not the threat—it's the upgrade.

I challenge you to approach AI with a growth mindset. See it as an extension of your capabilities rather than a replacement for them. This psychological orientation isn't just helpful for your personal comfort; it's becoming an important competitive advantage in today's landscape.

The real risk isn't that AI will make you irrelevant. It's that someone else will use it to become exponentially better at what you do. While you're deliberating about whether to incorporate these tools, there are those around you that are already iterating on their third or fourth implementation.

If you learn to work with these systems effectively, if you build your prompt capital and integrate it into your workflows, you become the person who can't be replaced. Not because you're doing the same work you've always done, but because you've evolved. You've grown into something new: an AI-enabled professional operating at a higher level of leverage than was previously possible.

The Road Ahead

A central goal my wife and I had in raising our children was NOT to teach them what to think, but instead how to think. It occurs to me as I write this, that this same principle applies to writing good prompts. The real skill isn't in knowing all the answers, but in knowing how to ask questions that lead to discovery.

When we talk about prompting as a new interface, building prompt capital, or overcoming implementation anxiety, we're really talking about a fundamental shift in how we approach problem-solving. We're developing a new form of literacy that extends our thinking rather than replacing it.

In the fable, Paul Bunyan didn't lose because the machine arrived. He lost because someone else figured out how to use it first. The steam-powered chainsaw wasn't inherently superior to human skill—it required human skill to wield it effectively.

The same is true with AI today. These tools aren't autonomous replacements for human intelligence. They're amplifiers of it. And like any amplifier, the quality of what comes out depends entirely on what goes in.

So as you navigate this new landscape, remember that your ability to think clearly, to articulate your thoughts precisely, and to iterate on your understanding has never been more valuable. The prompts you craft are reflections of your own thinking. What’s more, they are investments in your future capabilities.

The question isn't whether AI will change how we work. It's whether we'll seize the opportunity to work in ways that were previously impossible. Like never before, the distance between where you are and where you could go will be measured in the quality of the questions you ask.

The future of AI isn’t about who has access to the best model. It’s about who can talk to the model best.

In that world, your most valuable skill might not be code.

It might be language.
It might be clarity.
It might just be the prompt.

Until next week,
Derek
Oh and have something interesting you think I should write about? You can reply to this email (or any other Chief Rabbit email) to suggest it. 

Feeling Stuck? This Might Help

If all of this feels overwhelming or a bit foreign, and you’d like some help getting started, I’ve got you. I’m hosting a FREE, 45-minute live session on how to think about prompting, use it effectively, and start building your own prompt capital.

It’ll be hosted on Google Meet and is limited to the first 100 people, so if you want in, sign up here.

Before we meet, you’ll need to create a free account with Claude or ChatGPT (don’t worry, both have free plans) and spend a few minutes playing around. To make it easy, here’s a simple starter prompt to copy and paste:

That’s it; minimal prep, no pressure. Just show up curious. And again, this is free.