A flat surface dissolves into particles from one corner, revealing a slender isometric framework behind it

When the interface disappears

Berger+Team · Oct 18, 2026

Topic: Machine-readable website

There is a kind of software you use every day and never see. The anti-lock braking system in a car, for instance. It has no interface, no menu, no settings dialogue. It works, and you notice it only because something did not happen.

Part of what still exists today as a website with menus and forms is moving in the same direction. Not immediately and not everywhere — but far enough that it is worth asking what it means for a business.

What "the disappearing interface" means

The technical term is Zero UI, sometimes generative interface. Two different developments sit behind it, and they are often lumped together.

The first: the interface is created at the moment of the request. Instead of a fixed form with twelve fields, a system builds exactly the three fields needed for this case. That exists in early form already — language models can generate controls nobody designed in advance.

The second: there is no interface at all. Someone asks an assistant whether your business is open on Saturday. The assistant answers. Your website was involved, but nobody saw it.

The second development is the practically significant one, because it is already happening.

What it does to your business

If nobody sees your page any more, everything built to be looked at loses weight: the image at the top, the arrangement, the colour choices, the text that appears as you scroll. And everything built to be queried gains weight: are the opening hours stored so they can be read out? Does it say anywhere which location they apply to? Is there an unambiguous statement of what you do — and what you don't?

That is not a design question but one of data management. And it is why we have been writing for a while that a website has two audiences: people and machines, with different demands on the same content.

Important: the interface does not disappear everywhere. Someone choosing a craftsman usually still wants to see what they have built. The shift affects looking things up first — opening hours, availability, prices, responsibilities — and the decision itself only much later.

A typical example

A hospitality business in the Vinschgau valley had its opening hours clearly on the site — as a nicely set graphic in the header, part of the layout. Perfectly readable for guests.

We asked three assistants whether it was open on Wednesdays. Two said the information was not available. One gave hours taken from a three-year-old directory entry that were no longer correct.

The business had done nothing wrong. The information was there, current and well presented — just in a form a machine cannot open. Storing the same information additionally as structured data was an afternoon's work. The graphic stayed where it was.

That is the whole point: it is rarely about doing something differently. It is almost always about providing the same thing additionally in a second form.

Why this is not new

The shift looks more dramatic than it is if you take it for a break. It is more like the next step in a sequence.

There was a time when a website was a brochure put online. Then the search engine arrived, and suddenly the same content had to be readable by a program that classifies it. Whoever thought only about design back then disappeared from the results.

Then the phone became the primary device, and the same content had to work in the palm of a hand. Same content again, new form again.

Now a third audience joins, one that neither scrolls nor clicks but queries. Anyone who knows the sequence sees it: it is not the website that dies. What died, every time, was the assumption that there is exactly one way to use it.

What we know for certain — and what we don't

This is a topic where much is asserted, so let's separate.

It is proven that assistant systems answer questions without a click on the source. Several studies of zero-click searches show this — with different figures, but pointing the same way.

Plausible but unproven is how fast this continues and which sectors are affected when. Anyone giving you a year is estimating.

False is the claim that websites become superfluous. The opposite follows from the thing itself: if a machine answers, it needs a source. Whoever has no well-kept source of their own does not appear in the answer — or appears wrongly.

What follows from this, practically

Nothing dramatic. Preparing for a world with less interface looks exactly like a properly maintained website today:

  • The hard facts are there machine-readably, not only as running text or an image.
  • They do not contradict each other across website, map service and directories.
  • Every language version contains the same facts.
  • It states explicitly what you do — and where your service ends.

That is the same list as for AI optimization, and that is no coincidence: both are answers to the same shift. Whoever ensures today that their details are queryable is prepared for an interface that may no longer exist.

The rest is observation. We keep an eye on this because we measure it — not because we know how fast it goes.

Frequently asked questions.

What does my business gain if AI assistants can talk directly to my website?

Your details are ready exactly where people increasingly ask. Instead of an AI scraping your page and guessing, it reads exactly the content you've provided through an open interface — MCP (Model Context Protocol): services, opening hours, contact details. The aim: fewer incorrect statements about your business and correctly sourced citations — whether an AI uses this route is up to its provider. You stay in control — you decide which content is accessible, and every access is logged.

Do I need a chatbot on my website?

Usually no. A chatbot only serves visitors already on your site — it doesn't solve the real AI problem: that agents retrieve your facts elsewhere and run into outdated or contradictory data there. “Agent-ready” doesn't mean a chatbot, but a machine-readable foundation with clear, current data from a reliable source. Only once that's in place can a chatbot be a sensible extra — never the first step.

What is the difference between SEO, GEO and LLMO?

The three terms build on each other rather than replacing one another. SEO (Search Engine Optimisation) makes sure a page ranks well in classic search results. GEO (Generative Engine Optimisation) extends that goal to being cited as a source in AI-generated answers — in a ChatGPT or Perplexity answer, for example.

LLMO (Large Language Model Optimisation) goes a step further and gets more technical: it means specifically understanding and influencing how individual language models process, weigh and reproduce content — essentially reverse-engineering the model in question. In practice: SEO remains the entry ticket, GEO secures broad citability, and LLMO fine-tunes the details for the systems that matter most to a given business.

Whether you need this distinction day to day is a separate question — we answered it at length: Do you really need five new acronyms?

Next step

Want to look at it together?

No pitch, no standard package — an honest take on your project. A reply, usually within 24h.

Start a conversation →