Skip to content

Moat types

Is proprietary data a moat against AI?

Yes, and it may get stronger: data a model can't scrape, synthesize or license is one of the few inputs AI makes more valuable, not less.

Holds against AI

Updated November 19, 2026.

What is it?

Data you hold that nobody else can get: transaction records, outcomes, customer-specific context, sensor readings, expert judgments captured over years. It's proprietary only if a competitor can't buy, scrape or recreate it.

Does it hold against AI?

It holds. Models are only as good as their inputs, and public data is available to everyone. Data that isn't becomes the differentiator. Bustamante lists private data as holding, and possibly strengthening, under LLMs.

The trap is data that only feels proprietary. If your content is on public web pages, in a public API or in a dataset you licensed from a vendor, a model can get it too. Saker adds a test worth applying: is the data governed, discoverable, access-controlled and auditable, so agents add to its value rather than erode it?

How do you know if you have it?

Ask: could a funded competitor get equivalent data within a year? Is any of it publicly accessible, including through your own docs or API? Does it come from your customers' activity, or from sources others can use too?

If you can't answer the first question with a confident no, the data isn't a moat yet.

How do you deepen it?

Capture outcomes, not just activity: what happened after a decision is rarer than the decision. Close public leaks (open APIs, crawlable pages) that give the data away. Make the data more useful through your product than outside it, so customers keep feeding it.

Questions people ask

Is my customers' data a moat?
Only if you can use it to make the product better than a competitor could, and they can't easily take it elsewhere. Data you merely store is a switching cost at best.
Does publishing content make my data less proprietary?
Yes. Anything crawlable is available to models and to anyone building with them.

Sources

  1. Ten vertical-software moats under LLMs (Bustamante, summary): https://hq-raydata-co.pages.dev/vault/06-reference/2026-02-16-vertical-software-selloff/
  2. AI is eating enterprise SaaS (Saker): https://medium.com/@rsaker/ai-is-eating-enterprise-saas-1259d352f193
  3. How to tell if a startup has a moat (Lenny's Newsletter): https://www.lennysnewsletter.com/p/how-to-tell-if-a-startup-has-a-moat

Does your business have a moat?

Five AI attackers, live web research and a published rubric. Free, about five minutes, no account. Or talk it through with Matt on a strategy call.