Answers long enough to be useful

There is already a great deal of generic writing about AI and interiors. Very little of it is based on running the work, and almost none of it tells you when the tool is the wrong choice.

Each guide here states its conclusion first, shows the evidence behind it, and names the conditions under which it does not hold. It is written against the workflows Studio publishes and whatever examples exist on the day it is written.

They are organised by the decision you are trying to make rather than by date, and the three sections below are those decisions. Five guides are published on this hub so far, at least one under each of the three; for the questions they do not answer, this page publishes the question and the standard rather than the answer.

This section answers three decisions a design practice has to make: whether visual agents can take repeatable production work without giving up design control, what a drawing can carry into a room view and what has to be inferred from it, and which decision AI visualisation is genuinely good at compared with traditional rendering.

Each of those three has a guide below. The Design Visual Production and Client Review and Approval workflow pages state what Studio does at each step behind them, and each agent-team page states what it accepts, what it returns and what your staff still have to check.

This section answers the showroom decisions: how a product-in-room visual is produced with a member of your staff operating it, where category fidelity stops, what has to be disclosed to the customer looking at it, and how to check it on one category before committing to more.

All four of those are answered by the guide below. The Human-led Visual Sales Consultation workflow page states what Studio does at each step behind it, and the surface specialists listed on the Agent Teams page state what each one accepts and returns for a given product category.

This section will answer the questions an owner asks: how to separate governed repeatable work from the expert judgment that has to stay with a person, what to measure before and after so the difference is visible rather than anecdotal, and what adoption asks of the staff who have to run it.

The first of those has a guide below. No guide in this section is published for either of the other two questions yet: the Business Value page publishes the value argument and the measures it depends on, and the team trial and process pilot page explains how an evaluation is scoped and what it does not commit you to.

Every guide here is written against the workflows Studio publishes and whatever examples exist on the day it is written. Where a claim rests on a controlled demonstration rather than a customer result, the guide says so on the spot.

Where we hold a professional opinion the evidence does not settle, it is marked as an opinion. Where a statement is a product claim, it points at the page that describes the capability, so the two are never read as the same kind of thing.

We do not publish comparisons against named competitors, because we cannot run their tools under conditions we would accept as fair.

Each guide carries the date it was last revised, and we revise rather than republish.

Five guides have been published against this standard so far. That is why this section comes before the three above it rather than after them: it is the part of this page that stays true whatever is or is not written yet.