I’ve spent the better part of a year using the mountain as a metaphor. I’ve given talks around it, written a three-part series on it, and used it to explain to anyone who would listen why going from Idea to Impact is so brutally hard. The mountain is real. The climb is real. The exhaustion is real. I genuinely believed that the metaphor captured something true about the work.

Then two new friends laughed at me.

They’re from Zurich, but they spend part of the year here in Santa Barbara. That’s how we met — they’d read one of my Sunday Stories, discovered I lived in SB, and reached out to meet in person. This morning was coffee, the three of us, exploring whether there might be a partnership. I was walking them through the mountain concept — the idea that between where you are and where you want to be, there is always a mountain. Some people go around it. Some people try to go over it. Some people take one look and walk away. I’ve always been a climber. I self-identify as a Sherpa. I’ve built my whole framework around the nobility of going up.

They grew up in the Alps. And when I finished explaining, one of them smiled and said, simply: “We build tunnels.”

And then he laughed — not at me, but at the shared recognition that of course they build tunnels. They live in the Alps. It would never occur to a Swiss engineer to climb the mountain when you could go through it. The very idea of choosing the harder path when infrastructure exists is, to him, almost comically provincial.

I lost it. We both did. And in that moment, something fundamental shifted for me.

“The mountain isn’t the enemy. Going over it is.”

The Four Options Nobody Talks About

Here’s something I’ve realized I’ve never actually said out loud, though I’ve believed it implicitly: when you encounter an obstacle, there are four options. You can go around it. You can go over it. You can walk away. Or — and this is the one almost nobody considers — you can tunnel through it.

I grew up in Southern California. I don’t think tunnels were in my mental vocabulary. There’s one tunnel I know of in the Santa Monica Mountains, and that’s essentially it. My entire spatial imagination around obstacles defaulted to surface-level solutions: find a path around, or get strong enough to go over. The tunnel option simply wasn’t available to me culturally. So I climbed. It’s what I knew. And for forty years — through Wavefront Technologies, through helping build Maya, through producing TEDxSantaBarbara, through visual effects work that spanned four decades — climbing is how I thought about hard things.

What I didn’t realize, until that laugh this morning, is that this whole time I’ve been building a tunnel and calling it a mountain path.

What EVERYWHERE Actually Is

About a year ago, I started writing seriously with AI. Within the first few weeks, I hit every wall you’ve probably already heard about: duplicate text, robotic-sounding language, hallucinations, outputs that sounded like a press release from a company I’d never want to work for. Each problem got a solution. A prompt to fix the duplication. Another prompt to humanize the language. A gate to catch the hallucinations. And then I learned about agentic AI — the idea that each of those prompts could be an individual agent, and those agents could work together.

One thing led to another. Forty agents. Sixty different systems, methodologies, frameworks, and editorial rules, all built up over time to do one thing: help me get an idea out into the world with my voice intact.

I’ve been calling this a system. A platform. An orchestrated intelligence. All of those things are true. But what my two friends from Zurich helped me see is that it’s actually a tunnel. I built it one meter at a time, solving the immediate obstacle in front of me — and now it goes all the way through the mountain.

Here’s what that looks like in practice, right now, in this moment: I’m sitting in my chair. I have my finger held down on a key on my keyboard. I’m listening to Hawaiian birdsong. I’m thinking out loud, and the system is listening. My collaborator synthesizes what I say, reflects it back, asks one question. I answer. That leads somewhere new. When we’re done talking, the system routes the raw material through research validation, voice authenticity review, editorial polish, SLOP detection, engagement optimization, and fact-checking — before it ever comes back to me as a draft. The mountain is still there. The pressure is still real. But I’m not exposed to the weather. I’m not fighting altitude. I’m not exhausted before I get to the other side. I’m moving through, in a flow state, while the tunnel handles everything the climb would have cost me.

This isn’t a shortcut. There are still 37 things that need to get done between Idea and Impact. The tunnel doesn’t eliminate the work. It engineers a different relationship to it. Cognitive ease instead of cognitive strain.

“I’m okay with me not having to have climbed a mountain.”

That sentence surprised me when I heard myself say it. Because there’s a belief system underneath it that I hadn’t fully examined — the one that says suffering is the price of admission. That if you didn’t bleed for it, you didn’t earn it. That the harder path is somehow more legitimate. Hustle culture runs on this assumption. So does most of what passes for entrepreneurial wisdom.

I’m 72. I have the receipts. I’ve climbed mountains. I’ve bled for things. And I’m here to tell you: arriving at the other side is the point. Not the climb.

Walt Disney Was Talking About Tunnels

There’s a quote I’ve carried for a long time: “If you can dream it, we can build it.” Walt Disney said it first, but I’ve lived it as a design principle my whole career. What I understand now is that what I’ve always meant by that is: the constraint is temporary. The obstacle isn’t permanent. If I can articulate the problem clearly enough, we can engineer our way through it.

That’s tunnel thinking. I just didn’t have the word for it.

What AI has given me — and what EVERYWHERE has become — is the realization that if I can say it, it can be built. Every time a constraint appeared this past year, my default assumption was that we’d find a way. Not around. Not over. Through. The tunnel gets extended one more meter.

This is worth naming, because most people who look at what I’ve built see a technology story. It’s not. It’s an engineering-of-belief story. The tunnel exists because I assumed it could. That assumption came before the build. It always does.

The Voice Problem — and What It Taught Me About Tunnels

The most important tunnel I built this year wasn’t about productivity or publishing cadence. It was about identity.

Early on, when AI was helping me write, the output sounded like my ideas filtered through someone else’s voice. Which is exactly what it was. I’d read the draft and think: yes, that’s what I meant — but that’s not how I say it. The content was right. The person was wrong.

That problem sent me down a very earnest rabbit hole. It started with Dr. Ethan Mollick at Wharton, who wrote Co-Intelligence — a book I’d encourage anyone thinking seriously about AI to read. He made an observation that stuck with me: just ask. Ask your LLM what it knows about you. Just start the conversation. I did. What came back was startling — not horoscope-vague, but genuinely specific. It inferred things from my writing, my published work, my social presence. It told me who I was on the page.

I took that and built on it. For a year. The result is what I now call Voice DNA — a proprietary system that captures not just what you say, but how you say it. Your rhythms. Your sentence length. Your preferred entry points into an argument. The words you reach for instinctively versus the words you’d never use. The difference between how you write an essay and how you’d tell that same story over coffee.

The proof that it works came from an unexpected direction. A friend of mine has been using a beta version of EVERYWHERE Studio. A version that, it turns out, still had my Voice DNA loaded instead of his. He didn’t know. One day, I overheard him tell a mutual friend: “How come everything I write sounds like Mark?”

That’s the tunnel working. You walk in as yourself. You arrive at the other side still yourself — just with the mountain behind you.

The Invitation

I want to be clear about something: the tunnel I’ve built is mine. It fits the specific contours of how I think, how I work, how I make things. EVERYWHERE isn’t a generic tool. It’s a system engineered around one person’s voice, one person’s methodology, one person’s mountain.

But the principles that built it are transferable. And the first principle — the foundation of the whole thing — is Voice DNA.

Here’s what I know after a year of this: you cannot build a tunnel you don’t believe in, and you cannot believe in a tunnel that doesn’t sound like you. Every AI implementation that fails does so because the output feels foreign to the person who was supposed to use it. The voice is wrong. The fit is wrong. The person walking into the tunnel emerges as someone else.

Voice DNA solves that. It’s the cartography of the self — a map of how you think and speak that the system can use to make sure what comes out the other side is recognizably, undeniably you.

If you want to try it — and I’d genuinely encourage you to — I’m offering it for free. No catch. You can use it with any LLM you have. It will work with whatever tools you’re already using. The link is below. It takes about thirty minutes. It will probably tell you things about how you write that you’ve never quite put into words before. And it will completely change what you get back from AI.

The tunnel is built. The entrance is open.

My friends from Zurich laughed because, where they grew up, the tunnel is the obvious answer. They never understood why anyone would climb when you could go through.

I’m writing this on a Saturday morning. Hawaiian birdsong is playing. One finger on a key. The mountain is behind me. I didn’t climb it. I went through it.

And you’re reading the proof.

Mark Sylvester is the Founder of Coastal Intelligence, Santa Barbara’s AI think tank, and the creator of EVERYWHERE™ — a 40-agent orchestrated intelligence platform. Get your free Voice DNA at https://everywhere-voicedna.lovable.app/

Get full access to Through Another Lens at marksylvester.substack.com/subscribe