Why the AI Moat Conversation Is Missing Its Most Important Layer

We are currently witnessing the Great Leveling. In the age of AI, "hard to do" is a dying competitive advantage. If your value is built on execution, you are building on a sandbank during a rising tide. Michael Bloch recently published "The Only Moats That Matter“ a brilliant autopsy of what actually survives when intelligence becomes a commodity. He's right, but for the leaders I work with, his framework is missing the most human layer of all.

His core thesis: AI is making "hard to do" worthless as a competitive advantage. Building software, maintaining integrations, embedding your product so deep that switching costs protect you. All of that required intelligence and engineering time. AI compresses both. What still matters, Bloch argues, is "hard to get." Things that require real-world elapsed time that AI can't speed up. Network effects. Regulatory permission. Physical infrastructure. Capital at scale. Proprietary compounding data. Time that can't be parallelized. You get the gist.

He's right. And I'd encourage every founder and executive to internalize that filter. And I think his framework is incomplete.

Bloch writes for investors evaluating billion-dollar positions. His moats are chip fabs and government contracts and exchange funds with liquidity advantages. If you're a VC deploying capital into deep tech, his five moats are pretty spot on.

But most of the leaders I work with aren't building chip fabs. They're running companies. Leading teams. Building brands. Making decisions about where to invest their time, energy, and reputation in a world that's shifting underneath them faster than any of us expected.

For that audience, I think there are three categories that matter. Not just two.

Hard to do. Hard to get. And hard to believe.

Hard to Do: Execution

This is the one AI eats first.

I run a workshop called Rewired where we show business leaders what AI can actually do right now. Not in theory but in real time. We create agents to write copy, build landing pages, connect payment systems, and generate designs. Tasks that used to require at least some skill, time, and sometimes specialized knowledge happen in minutes.

The room always goes quiet at the same moment. Not when people see the speed. When they realize the thing they spent years getting good at just became a commodity. That's over.

If your competitive advantage is that you can execute a task faster or more skillfully than someone else, you are competing against a force that doubles in capability every few months.

What to do about it: Stop competing on execution. If the thing you're known for can be described as a task, it's being automated. The question is not "how do I do this better?" It's "what do I do that can't be described as a task at all?" The leaders who see this shift early and reposition accordingly will have an enormous head start. The ones who keep optimizing their execution speed are perfecting a skill that's about to be worth zero.

Hard to Get: Access

This is Bloch's territory, and he maps it well. But I want to translate it for the leaders and founders in this audience.

Hard to get isn't just chip fabs and government clearances. It's the dinner you got invited to because someone trusts you. It's the introduction that happened because you showed up for ten years and people know your name. It's the deal flow that comes to you because you've built a reputation in a specific space.

Network effects work for people, not just platforms. Every relationship you build makes the next one easier. Every room you're in leads to another room. This compounds over years in ways that can't be shortcut. Access creates its own momentum. It's a force that builds behind you, and the longer you've been building it, the harder it is for anyone to catch up.

Access is the key. Trust is the weight. Access is binary. You're either in the room or you're not. It's the "hard to get" infrastructure of your career. But once you're in that room, access alone loses its power. That's where the third moat takes over.

What to do about it: Invest in access deliberately. Map your network. Show up in rooms where decisions get made. Don't outsource your relationships to automation. The leaders who will thrive in the AI era aren't the ones with the best tools. They're the ones with the best rolodex and the earned right to use it.

In short, proximity is power.

Hard to Believe: Trust

This is the moat I believe has been overlooked. And it might be the most important one.

Bloch actually hints at it. At the end of his piece, he writes: "Does trust become its own moat when AI is doing more of the work? Someone has to be accountable when things go wrong." He flags it as an open question. I think it is the answer.

When someone raised this exact point in the replies, Bloch pushed back. "Trust is hard to get," he wrote. "It takes years to build and seconds to lose. But I think that fragility is exactly why it didn't make my list. The five moats I named compound over time and are hard to lose once you have them. Trust can evaporate overnight."

He's right that trust is fragile. But I'd argue that's what makes it valuable, not what disqualifies it. A chip fab doesn't require courage. A government contract doesn't require you to have a point of view. Trust does. It demands that you show up, take positions, and risk being wrong in public. The fact that it can be lost is precisely why people respect the people who've kept it. Fragility is the price of the most powerful moat there is.

Here's what's happening. AI has made it possible for anyone to produce anything. Any piece of content, any website, any proposal, any pitch deck, any product demo. The cost of production has collapsed to zero. In the age of AI, mediocrity is the floor.

So what's scarce?

Definitely not output. Not execution. Not even access, though that still matters.

What's scarce is the thing that makes someone stop scrolling and actually listen. The thing that makes a CEO hire you instead of the other ten firms who sent identical AI-generated proposals. The thing that turns information into influence.

Belief.

Trust is the new bottleneck. When everyone can produce everything, the only question that matters is: do I believe this person? Have they earned the right to say what they're saying? Is there a track record behind this, or is it just well-packaged noise?

I believe we will see the end of thought leadership and the beginning of thought "doership." Receipts, not resumes.

I've spent 20 years building a brand. 500+ keynotes in 26 countries. Two million followers. 180,000 newsletter subscribers. A company I built and led for 17 years. None of that happened because I had better tools than anyone else. It happened because I showed up consistently, said things that turned out to be right, and did the work long enough for people to notice.

That's not a humble brag. That's the point. Those 20 years can't be compressed. AI can write a thought leadership post in 30 seconds. It cannot manufacture the reputation that makes people actually care what the post says.

Trust shares every property of Bloch's moats. It compounds over time. It benefits from network effects: the more people who trust you, the more new people are inclined to. It requires real-world elapsed time to build. You can't parallelize credibility. And it generates its own momentum. Each time you make the right call in public, the next call carries more weight. Each time you show up and deliver, showing up gets easier and the audience gets larger. Trust isn't static. It's a force that builds behind you. Once it's moving, it's very hard to stop. And nearly impossible to replicate from a standing start.

And unlike chip fabs and government contracts, trust is a moat available to every single leader reading this. You don't need a billion dollars. You need a point of view, consistency, and time.

What to do about it: Start building trust now. Not by producing more content. By standing behind what you produce. Have a point of view and be willing to be wrong in public. Show your work over years, not weeks.

Here's what most leaders get wrong: they see the signal but they wait to act on it. They know trust matters. They know personal brand matters. But they treat it as something they'll get to later, after the "real" work is done. That's the same mistake companies make when they see a market shift and add it to a six-month strategic roadmap. By the time they move, the window has closed.

The window to establish yourself as a trusted voice in your space is closing. Not because it's too late. Because it's about to get dramatically harder. Once AI floods every channel with competent-sounding content, the bar for breaking through will be almost impossibly high. The leaders who started early will have a compounding advantage that late entrants can't touch. The decision to build has to happen now. The building itself will take years. Both of those things are true at the same time.

The Full Framework

Here's how I'd reframe the conversation for anyone building a business, a brand, or a career in the age of AI:

Hard to Do = Execution. AI eats this. Stop competing here.

Hard to Get = Access. Still real. Invest in your network, your relationships, your presence in rooms that matter.

Hard to Believe = Trust. The moat that 99% of leaders actually have access to. Build it now. Build it consistently. Build it in public.

Bloch is right that the moats which survive are the ones bottlenecked by time, not intelligence. I'd just add: the most democratic version of that moat isn't infrastructure or regulation or capital. It's the trust you build by being a real person who shows up, takes positions, and earns belief over years.

AI is about to make every company look competent. The ones that win will be the ones people actually believe in.

PS: I mentioned Rewired earlier in this piece. It’s where I work with leaders who are ready to stop dabbling and actually build. You’ll be in a great cohort with CEOs, Fortune 10 CMOs, and some world-class thought leaders. The next Mastermind cohort starts April 15: shamahyder.com/rewired-mastermind