Why AI Can’t Recommend What It Doesn’t Understand

Lately, I keep having a very familiar feeling.

It feels like the early 2000s again.

Back then, I was working as a Novell engineer. I was doing desktop and server support, rolling out backups, managing upgrades, and helping businesses move from Novell 3.11 to 4.x and then 5.x. I installed GroupWise when most companies had never experienced shared email or calendars before.

I also worked at an MSP called Entex, based out of Westport, Connecticut. And I remember very clearly telling business owners, “You should really build a website.”

They would look at me like I was speaking another language.

“Why would we need that?”

At the time, it sounded optional. Abstract. A nice-to-have.

We all know how that turned out.

And right now, AI visibility feels exactly like that moment.

We’ve Been Here Before, Even If It Doesn’t Feel Like It

In the early days of networks, email, and websites, the resistance wasn’t technical.

It was conceptual.

People didn’t yet understand:

• why shared systems mattered
• why centralization mattered
• why visibility mattered before demand existed

They didn’t see the value until after the shift was already underway.

AI is creating the same kind of shift.

Not because it’s flashy.

Not because it’s trendy.

But because it changes how information is consumed and surfaced.

AI Doesn’t Browse Websites. It Extracts Understanding.

This is the part most people miss.

AI systems don’t visit websites the way humans do. They don’t scroll, admire design, or “get the gist.”

They extract meaning.

They look for:

• clearly defined entities
• consistent facts
• structured relationships
• content that can be summarized with confidence

If a site is unclear, inconsistent, or structurally messy, AI doesn’t argue with it.

It just moves on.

As I often say to clients when this clicks:

“AI can’t recommend what it doesn’t understand.”

This Isn’t a New SEO Thing. It’s the Next Consumption Layer.

This is where my background makes this moment feel so familiar.

When we rolled out GroupWise, people didn’t yet understand group email. When we implemented backups, they didn’t feel urgent until the first server failed. When we talked about websites, businesses couldn’t imagine being discovered online.

The people who listened early didn’t win because they were smarter.

They won because they adapted sooner.

AI is not replacing search overnight.

It’s layering on top of it.

And it’s inheriting the same signals search engines already rely on.

Google has been very clear that its systems aim to understand content and context across sites, not just individual pages. That same understanding feeds how content is summarized, surfaced, and reused in AI-driven experiences.

What We’re Seeing with Clients Right Now

This is already playing out in real ways.

We see businesses with:

• strong websites
• good content
• decent SEO performance

But when we evaluate them through an AI lens, gaps appear.

The business is described differently across pages.

Core services aren’t clearly prioritized.

Facts are buried inside long-form content.

Search engines can still rank them occasionally.

AI systems can’t confidently summarize them.

That’s the difference.

As I often explain:

“Search can tolerate some ambiguity. AI won’t.”

The Pattern Is Always the Same

The businesses that will win here are not chasing tools.

They are:

• clarifying who they are
• tightening structure
• reinforcing meaning
• making their websites easier to understand, not louder

This is exactly what happened with networks.

With email.

With websites.

The technology changed.

The winners prepared early.

Why This Moment Matters

We are at another inflection point.

Not everyone will act.

Not everyone will see it.

And that’s okay.

But the businesses that listen now, adjust calmly, and build clarity into their websites will have an advantage that compounds over time.

Just like the businesses that said “yes” to networks, email, and websites back then.

As I often say now, with a bit of a smile:

“I’ve seen this movie before.”

And the ending is always the same.