See the Gap, Supply It: What Survives When AI Commoditizes Everything
Everyone asks 'what moat do I build against AI?' Wrong question. The market doesn't reward entrenchment. It rewards perception. See the gap before anyone else and fill it while they're still arguing about frameworks.
Everyone is asking the wrong question.
"What moat do I build against AI?" implies defense. Walls. Entrenchment. Accumulate assets. Dig in. Wait.
Wrong frame.
The market doesn't reward entrenchment. It rewards perception. The person who wins isn't the one with the thickest walls. It's the one who sees the gap before anyone else and fills it while they're still arguing about frameworks.
The Levi Strauss Play
Levi Strauss didn't mine gold. He watched miners rip their pants and sold them durable jeans.
The gold rushers competed with each other. Strauss competed with nobody because he was looking at a different problem entirely.
That's the pattern. The crowd runs somewhere. They forget things. They create needs they don't even know they have yet. You supply those needs.
What Doesn't Work
1. Implementation Speed
Someone with Claude + n8n + a YouTube tutorial builds a speed-to-lead AI agent in a weekend. Today that's a $5K consulting engagement. In 18 months, it's a commodity.
Boris Cherny (Head of Claude Code): "Coding is largely solved." GIC practitioners report $4K/month in tokens generating 20 PRs/day. AI Economic Restructuring data: rate compression to $10-15/hour for pure implementation within 2 years.
Implementation speed is a cash engine. Ride it while it pays. Never mistake it for identity.
2. Domain Knowledge as a Shield
Someone uses Claude to learn Indian markets in a week. Reads 100 PDFs about e-commerce operations in a day. Knowledge compression is real. "I know this industry" is a 6-month head start, not a 5-year advantage.
Domain knowledge helps you see gaps faster. It's the telescope, not the fortress.
3. Walls, Period
Proprietary data, audience, brand. These are assets. But the strategy of "accumulate and defend" is slow. By the time you've built the wall, the landscape shifted. The wall protects territory that's already being commoditized.
What Works: Speed + Perception
Not three assets to accumulate. Three capabilities to develop:
1. Pattern Recognition Across Domains
2,500+ analyzed trading days. Conversion data from e-commerce. Healthcare AI architecture experience. These are valuable not because they're walls but because they sharpen your ability to see what's missing in a market.
Data is a lens, not a shield.
2. Speed of Response
When you spot the gap, can you fill it before the crowd notices? This is the muscle. Every fast consulting turnaround, every "built in a weekend" delivery, every rapid prototype trained this reflex.
The gap-thinker doesn't plan for six months. They ship in one.
3. Willingness to Drop What's Commoditizing
Most people cling. They spent 3 months building something, so they defend it even as the commodity line swallows it. Gap-thinking requires letting go fast. If it's below the line now, walk away. Find the next gap.
The test: Not "does this get smarter with use?" but "am I the first person in the room who noticed this need?"
The Two-Track Model
TRACK 1: CASH ENGINE (Below the line)
Purpose: Fund the lab
What: Consulting, workshops, automations
Why: The gold rushers need picks and shovels TODAY
Reality: Commoditizes. You know this. Ride it while it pays.
TRACK 2: IP ENGINE (Above the line)
Purpose: Own the gap
What: Products, platforms, content
Why: Each product exists because you saw a gap nobody filled
Reality: Only compounds if you keep reading the wind.
Services = rent money. Pay bills. Don't fall in love. Products = ownership. But only if they're pointed at live gaps. Content = the scanner. It forces you to articulate what you're seeing, which sharpens the seeing.
"Build products. Rent out services while the lab is running."
The Crowd Creates the Gap
Every gold rush creates demand for things the rushers forgot:
In 1849:
- Tools the miners needed (picks, shovels, jeans)
- Infrastructure they assumed (assay offices, banks, supply routes)
- Recovery from their mistakes (medical care, legal help, debt)
In 2026:
- People who can evaluate AI vendors (not build, evaluate)
- Vertical integration that actually works (not demos, not POCs)
- Trust in a geography where "AI" is still a buzzword
- Products built on real usage data, not pitch decks
These are gaps. Not walls to build. Gaps to fill. When they close, find the next ones.
The Commodity Line
Sidu Ponnappa framed it perfectly: there's a line between inventory (commoditizes) and assets (persists). The line only moves up.
Below the line (and sinking): chatbots, basic automations, speed-to-lead agents, generic AI integration.
Above the line (for now): real-time trading systems with proprietary data, domain knowledge encoded in production systems, vertical AI with accumulated usage data.
The gap-thinker doesn't fight the line. They read it, anticipate it, and relocate before it arrives.
How to See Gaps
Three questions to ask about any market:
1. "Can someone with Claude + n8n + a YouTube tutorial replicate this in a weekend?" If yes, you haven't found a gap. You've found a feature.
2. "What are the rushers forgetting to pack?" When 49.7% of AI agents are in software engineering and 1% are in healthcare, the gap isn't in software. It's in healthcare.
3. "What would switching require rebuilding a human relationship?" If your client would need to rebuild trust, re-explain context, re-establish working chemistry to switch to a competitor—that's a gap. Not a technology gap. A human gap. Those are the most durable.
The people building walls are playing defense in a game where the field keeps changing shape. The people reading gaps don't need walls. They're already gone by the time the crowd arrives.
Rewritten March 14, 2026. Core strategic conviction. The old framework was about defense. This one is about perception.