Field Note · Governance
Your Design Firm Already Has an AI Policy. It's Just Unwritten.
The principal experiments with LLMs. The intern runs everything through a personal account. Nobody has talked about it as a firm. That silence is the policy.
DefinitionAn unwritten AI policy is what you have when individual team members use AI tools independently, without firm-level guidance on what the tools are for, what client data can be entered, or how the firm responds when clients bring their own AI-generated work. It is a policy by default rather than by design. Most firms have one. None of them chose it.
The principal experiments with LLMs now. The intern runs everything through a personal Claude account. The procurement manager thinks nobody uses AI but worries it will replace their job.
Business of Home called it the "private dabbling" phase. It's the condition most interior design and architecture firms are in right now: adoption real, policy absent, exposure accumulating.
Fred Nicolaus reported in May 2026 that AI adoption in interior design has tripled since 2023. It took less than three years to go from a niche experiment to a normalized part of how many studios operate. The adoption story is over. The policy story hasn't started.
AI adoption in interior design, tripled since 2023. Source: Business of Home, May 2026.
An unwritten AI policy is what you have when individual team members use AI tools independently, without firm-level guidance on what the tools are for, what client data can be entered, or how the firm responds when clients bring their own AI-generated work. It is a policy by default rather than by design.
Most firms have one. None of them chose it.
What Business of Home actually found
The diagnosis in the BOH piece is specific. The principal uses one tool. The senior designer uses another. The intern uses three. Nobody has talked about it as a firm. Nobody knows whether the others are entering client data. Nobody has a stance on what to say when a client shows up with AI-generated furniture arrangements and asks why they should pay for yours.
This is not a technology adoption problem. It is an operational coherence problem. The same kind that appears when onboarding is informal, when project handoffs are verbal, when proposal language varies by who's writing that week.
Firms running on one expert's judgment, which describes most design studios, are particularly exposed. When the decision-making authority is concentrated, informal practices spread from the top down. Whatever the principal does with AI becomes the implicit standard, without ever being stated.
You don't have an AI problem. You have an unwritten policy, and it is being enforced by accident.
The five questions that cannot wait
BOH's framing identifies five questions every design firm now needs to answer. These are not theoretical. They are surfacing in client conversations, in hiring, and in competitive differentiation right now.
What do you actually want AI to do? Not what could AI do. What specific jobs in your operation would genuinely benefit from AI, and which jobs require your judgment in ways AI cannot approximate? The firms getting real value have answered this question. The firms getting noise haven't.
What is your policy on tools? Do you provide a firm-licensed platform, or do you offer a stipend for personal subscriptions? The tool question is also a data question. Personal accounts mean client information lives in systems the firm doesn't control.
How will you handle client data and privacy? Interior design involves confidential details about how clients live. Floor plans, purchase records, family composition, personal habits. What can be entered into an AI system and what cannot is a question that needs a firm-level answer before an incident requires one.
How will you work with AI-armed clients? BOH flagged this directly. Clients now arrive with AI-generated renderings, furniture arrangements, and material palettes. Some are genuinely useful; many are not. Having a considered, consistent response to this is both a client management skill and a positioning question.
How does your team feel about it? Not a soft question. Teams that feel threatened by AI without a clear answer from leadership will either avoid the tools or use them covertly. Either creates operational fragmentation.
| The question | Private dabbling (policy by default) | Deliberate policy (by design) |
|---|---|---|
| Purpose | Each person decides for themselves what AI is for. | The firm has named what AI is for, and what stays human judgment. |
| Tools & stipend | Mixed tools, personal accounts, no shared standard. | A firm-licensed platform or a defined stipend, one standard. |
| Client data & privacy | Confidential details enter systems nobody reviewed. | A firm-level rule on what can and cannot be entered. |
| AI-armed clients | No agreed response when a client brings AI work. | A consistent, considered stance on client-generated work. |
| Team alignment | Fear and covert use, never spoken aloud. | Leadership has answered, so the team is aligned. |
The same five dimensions, answered by accident or answered on purpose.
Why this starts with an audit
These five questions don't have universal answers. What makes sense for a three-person residential studio is not what makes sense for a twelve-person firm doing commercial hospitality work. The right policy is a function of how work actually moves through the specific business.
That is why the questions are better answered after an honest audit of current operations rather than before. An audit maps how decisions actually get made, where client data actually flows, and what the team is actually doing versus what leadership assumes they're doing.
The gap between those two pictures is where the risk lives. And it is almost always wider than expected.
Business of Home is the trade press of the creative studio audience Radiant Work works with. When it identifies five unavoidable questions and frames them as the defining operational challenge of the year, that is a signal worth taking seriously. The firms that answer them deliberately will have a structural advantage over the firms that continue to answer them by default.
The Operations Audit is how firms answer them deliberately.
Related Questions
What does an unwritten AI policy look like in a design firm?
An unwritten AI policy is the informal, inconsistent pattern of AI tool use that develops when there's no firm-level guidance. Individual team members use different tools with different assumptions about what client data can be shared. Business of Home documented this as the private dabbling phase common across the interior design industry in 2026.
How should design firms handle clients using AI to generate their own designs?
Business of Home identified AI-armed clients as one of five unavoidable AI questions for design firms. A considered response requires two things: a clear articulation of what design judgment AI cannot replicate, and a process for addressing client-generated work in project conversations without either dismissing or unconditionally validating it.
What is the risk of using AI with no firm policy in place?
Without a firm-level AI policy, client data may be entered into tools the firm doesn't control, team members operate with different assumptions about what's allowed, and the firm lacks a consistent response when clients or regulators ask how AI is being used. The exposure is both operational and reputational.
When did AI adoption in interior design take off?
Business of Home reported in May 2026 that AI adoption in interior design has tripled since 2023, moving the industry from early-adopter experimentation to widespread informal use within three years.
What should a design firm's first step be on AI policy?
Audit before you prescribe. Understanding how AI is currently being used, what data is being entered, and what the team's experience and concerns are gives you the foundation for a policy that will actually be followed. A policy built without that foundation is a document, not a practice.
The Work Behind the Work
Answer the five questions on purpose, not by accident.
Take the first step toward a business that runs with clarity and momentum.
For Deeper Context
- Fred Nicolaus, "The 5 AI questions every designer needs to answer," Business of Home (May 27, 2026)
- "AI Rules Nobody Follows: The Creative Governance Gap" (2026), on the distance between having an AI policy and having one the team follows
- "Data Privacy Day 2026: Privacy as the Foundation of Responsible AI Governance," The National Law Review (2026)