TL;DR — Not every AI interaction should be a chatbot. Real work sorts into three interface shapes: Direct GUI (point and click), Headless Agent (runs in the background), and Generative UI (builds a bespoke screen per question). The skill is matching the shape to the task. This note includes an interactive demo and a 3×3 decision matrix.

Field Note · Visual Interface

The app that builds itself for one question.

Chat gave every problem the same little text box. Generative UI does something stranger and more useful — it assembles a bespoke, throwaway interface around the exact thing you asked, then forgets it when you're done.

For three years the answer to "how should people use AI?" has been a single, stubborn shape: a chat box. You type, it types back — a wall of markdown you then have to read, parse, and act on yourself. It was a huge leap over a blank prompt. It is also, for a surprising amount of real work, the wrong tool.

Because not every question wants a paragraph. Ask a model to help you dress for a winter wedding and a wall of text is a downgrade from a page you could actually look at. Ask it to explain fractals to a curious ten-year-old and the right answer is something they can drag a slider on. Ask it to compare database options for your workload and you want a dynamic simulator, not prose.

That is the bet behind generative UI: instead of returning content, the model generates the interface — a small, custom, interactive application built for one prompt and thrown away after. Google's research team put it plainly: not just content, but "immersive visual experiences and interactive interfaces… automatically designed and fully customized in response to any question, instruction, or prompt."

Interface fit map
Three Interface Shapes
Each mode has a sweet spot — and a point where it breaks.

Direct GUI

The user points at and manipulates known objects. Best when the action is small and the context is clear.

Strong at

Editing, approving, selecting, scanning.

Breaks when

Stitching together multiple systems/policies.

Executive test

If they can say "change this line," give them the line.

Headless Agent

The work runs in the background. Best when tasks are long, multi-step, and procedural.

Strong at

Running tool chains, tracing data, retrying steps.

Breaks when

Comparing tradeoffs or steering paths visually.

Executive test

If the job is "do the analysis, show errors," hide the screen.

Generative UI

The layout adapts to the question. Best when teams need to see planning constraints dynamically.

Strong at

Temporary surfaces for messy planning/tradeoffs.

Breaks when

Rebuilding layouts adds reading time instead of saving it.

Executive test

If each team needs a different lens on the same plan, generate it.

Watch it happen

Here is the same prompt bar, three different questions. A chatbot would answer all three with text. Generative UI answers each with a different tool. Pick a question and watch the screen assemble itself.

>_ Pick a prompt below… generating
Composing interface…

↑ Choose a prompt. The same box will build a different interface for each one.

A chatbot would have returned three walls of text. Generative UI returns three different tools — and discards each when you move on.
Demo · a bespoke interface generated per prompt (mocked; real systems generate the markup live)

Why this is different, not just prettier

It is tempting to file generative UI under "nicer formatting." It is more than that. A chat answer is static — the model's one shot at being useful, frozen as text. A generated interface is a place you can keep working: adjust the fractal's depth, swap a jacket, get the next problem. The output stops being a message and starts being a small machine.

When Google ran the comparison, humans strongly preferred generated interfaces over the standard markdown reply — a "substantial gap" — landing second only to sites hand-built by human experts. The honest catch is speed: generating a real interface can take a minute or more, and that minute is not free. Which is exactly why generative UI is a choice, not a default.

It's one of three moves, not the only one

The mistake would be to make everything generative, the same way we made everything a chatbot. Most real work sorts into three interface shapes — and the skill is matching the shape to the task in front of the team.

Chat was never the destination. It was the placeholder we all agreed to while we figured out what the interface should have been.

We must distinguish between three distinct product interfaces, choosing the correct shape for the cognitive overhead of the target workload.

The 3×3 Journey Matrix
Journey Comparison
How the winner changes based on workload context.
Workload 01
Edit the launch brief (Known Object)
Direct GUI WINNER
Low overhead, high control. User touches the line directly.
Headless Agent
Turning visible edits into conversations adds reading work.
Generative UI
Overbuilt for simple text changes.
Workload 02
Run the market analysis (Analysis Loop)
Direct GUI
Too much process on screen. Exposes database columns but leaves analysis to the user.
Headless Agent WINNER
Human reviews assumptions, not clicks
Generative UI
Good for review, not for the whole run.
Workload 03
Align roadmap and marketing (Team Decision)
Direct GUI
The SaaS grid objects are too rigid for cross-department dependencies.
Headless Agent
Black-box recommendations compress critical tradeoffs into simple text.
Generative UI WINNER
The interface adapts to the planning question

What it means if you build things

The teams who win the next wave won't be the ones who bolt a chat box onto every product. They'll be the ones who ask, task by task, whether this moment wants a screen you point at, a daemon you never see, or an interface generated on the spot — and who treat tokens, latency, and cognitive load as the real product costs they are.

Generative UI is the most futuristic of the three and the easiest to overuse. Reach for it when the question genuinely deserves its own tool. The rest of the time, the kindest thing you can build is the boring screen that was always the right answer.