What is the MOAT?

Software was first, the next disruption will be hardware. And that opens the question: what is a company MOAT in the future?

I’m quite confident to say software is a “solved” problem. It is absolutely incredible what we were able to accomplish over the past 2 months using ChatGPT, Claude, and DeepSeek. But during the same time period, I realized that it doesn’t stop at software. We have now AI powerful enough to build, hack, and operate any software we can dream of. But the same AI is now also powerful enough to design the PCBs, electronics, even new sensors on which AI-developed firmware and software is running on. It can also design 3D CAD designs for highly complex mechanical structures, it can simulate materials, and deliver the BOM and plans, to turn - soon - any hardware into reality. (By the way, it can also write, draw, and file for patents of any invention as well!)

What I expect soon are specialized on-demand manufacturing hubs who operate AI-first. They own 3D printers, they print PCBs, and they accept whatever crazy idea any human has via a direct ChatGPT or Claude connector. A clear vision turns into a dozen prompts, which results in software, firmware, and - with one click and and payment - is printed and delivered within hours.

The interesting question: Is traditional manufacturing and (expensive) stock, and distribution still a competitive advantage? At what scale does on-demand manufacturing become cheaper than mass-production in China (or wherever) and shipping it to your home? I only know that over the past 10 years, 3D printing became pretty cheap already. And with scale, it will become 10x cheaper. We’ll see similar on-demand technologies for clothing, wood, cosmetics, electronics, whatever you can imagine. Not near-shore, probably within a radius of 1h of your home. Delivered by a drone, within minutes it is manufactured.

Which means, that the MOAT behind the vast majority of products (and therefore businesses) turn to zero.

Companies that remain will produce either quality on mass (that no on-demand manufacturing can compete with) or they they produce the raw-materials required for on-demand manufacturing, or innovative components used for it.

And because I mentioned AI can patent your ideas: Even patented technology is not protected. If the future is an economy where consumers produce for a market of N=1, who do you want to sue? And how do you find out?

Where, therefore, is the MOAT that remains?

I think specialization is the wrong answer, because any nerd using an AI system 10x smart than experts themselves is more specialized (and determined) to outcompete any “specialized” firm. And soon they have the AI and the on-demand manufacturing to make any idea a reality.

The answer becomes also not quality, but “precision”. Quality does not equal precision, but precision equals quality. Companies that can produce certain things to a precision nobody else is capable of, will remain winners. Precision might be the result of unique or innovative manufacturing - too complex and costly, that it creates a big barrier to entry - or the result of novel and hard to copy data.

This leave us with two MOATs: infrastructure (mass produced products such as energy, food, etc.) and precision (products and machines too complex to easily replicate).

All other MOATs are temporary: brand, distribution, etc. Why? Brand becomes irrelevant unless it is backed by trust (infrastructure) or quality (precision). Distribution networks lose meaning, when the new status symbol is unique, personalized and manufactured upon demand.

Then there is one thing left: the human.

Quite honestly, the future looks as if the unpleasant work is finally overtaken by AI, robots, machines, and drones. This makes the human element more relevant than ever before. Unless we end up in a dystopian world where everyone lives in a virtual reality, human connections become more relevant than before. (What else do you want to spend your time on?)

I think that innovative hospitality concepts, community-based businesses, and travel companies that bring together groups of humans are going to be of high value. Basically any company that brings the human element back into an automated AI economy.

That is why I ultimately see three MOATs remaining: infrastructure, precision, and the human.