AI Is Splitting Interior Design Into Two Careers. Which Side Are You On?

Last spring, a lead designer at a mid-sized agency sat through a presentation about AI tools for "asset variations and quick mockups." She went quiet. Her manager noticed immediately. Two weeks later, she resigned — joining a boutique firm that had made a point of advertising itself as "all human craftsmanship, no AI." Her old agency started cutting design headcount shortly after.

Around the same time, Dixie Willard — a veteran designer with stints at Williams-Sonoma, Ethan Allen, and multiple design-tech platforms — sat down with ChatGPT and rebuilt her entire business in a week. New name (Poised & Plumb), new positioning, new sales materials, new operations playbook. "From the get-go, ChatGPT has been my best friend," she said afterward. "It's all in my words. It sounds like me. It feels like me."

Same profession. Same tools. Completely different outcomes. The difference wasn't talent, seniority, or market conditions. It was whether AI arrived as something being done to each of them — or something they decided to pick up first.

If you're reading this because you're worried about your own position, you're asking a reasonable question. But it might be the slightly wrong one. The more useful question is: which version of this story are you currently set up to live?

Before that can be answered honestly, you need to know what AI is actually doing inside design work right now — not in theory, but task by task.

What AI Is Already Replacing (And What It Cannot Touch)

AI has absorbed the research-heavy, documentation-heavy, and iteration-heavy layers of design work. It has not touched the interpretation, lifestyle-reading, and real-time problem-solving that constitute a designer's irreplaceable value. The line between those two categories is sharper than most practitioners realize.

Julia Begbie, a London-based designer with more than 25 years of practice, documented the automation shift in concrete terms. Using NotebookLM, she transcribed a 30-minute client meeting and produced detailed minutes plus a fresh webinar plan "within moments of saying goodbye." Sourcing starting lists, drafting client emails, assembling mood boards, generating preliminary layout variations — all of this is now compressible to minutes. One designer put the sourcing timeline bluntly: ChatGPT returns a starting list in approximately 45 seconds.

The economic pressure behind that speed matters. AI design subscriptions run $9 to $20 per month. The average interior designer charges $150 per hour, according to a May 2025 Wall Street Journal analysis. When the price gap is this wide, even imperfect AI output displaces entry-level billable work — not because clients prefer it, but because they can rationalize it.

AI is fast but lacks intuition: it generates multiple layouts but cannot sense emotional or practical needs.
— Lisa and Marc Welch, Welch Design Studio

Lisa and Marc Welch of Welch Design Studio in Los Angeles published a clear framework for what sits on the other side of that line. "AI is fast but lacks intuition," they wrote. "It generates multiple layouts but cannot sense emotional or practical needs." Their taxonomy of the gap covers three dimensions: no lifestyle interpretation (daily habits, personal history, cultural context), no real-time adaptation (when a remodel opens a wall and reveals something unexpected), and no creative authority over the final judgment call.

The tasks at risk are real — and they are the tasks that have historically funded early-career design work. But the tasks that remain protected are also real, and they are the ones most directly tied to client trust and project fee.

This pattern holds across any profession where junior work is documentation-heavy and senior work is judgment-heavy — marketing, HR, architecture, project management. The automation hits the entry ramp hardest. The question worth asking about your own work: what do you do that a well-prompted AI could reproduce in under an hour? That's the inventory that needs rebuilding.

The Job Numbers Look Calm. The Inside Doesn't.

The Bureau of Labor Statistics projects 3% job growth for interior designers between 2024 and 2034 — roughly 87,100 jobs in 2024, median pay $63,490. The headline looks stable. But a closer look at what's happening beneath that number tells a different story.

A 2025 Houzz survey of 722 U.S. construction and design firms found that 34% are actively using AI — but that number jumps to 50% in firms with 10 or more employees. Larger firms are gaining productivity advantages that solo practitioners and small studios don't have access to yet. The gap between what an AI-equipped firm can deliver per designer and what an unequipped competitor can is already measurable. Houzz puts the average annual productivity gain from AI adoption at more than $108,000 per firm.

Architects — a close professional proxy for interior designers — provide a leading indicator of where this goes. Eighty-six percent of architects using AI report significant time savings, with 59% saying those savings span most of their work, according to a 2026 Chaos/Architizer survey. Interior design is running roughly 12 to 18 months behind that adoption curve.

Business consultant Katie McFarlan of Dakota Design Company identified the most exposed tier precisely in an April 2026 analysis. Designers who rely on "non-custom retail, hourly research, and commodity layout" are being squeezed out within 12 to 24 months. The generalist middle is the most vulnerable position — not the high-end specialist, not the scale operator, but the designer caught in between without a clear differentiating answer to the question: what do you do that AI can't approximate cheaply?

The 3% BLS growth figure isn't reassuring — it's a warning dressed as stability. The profession isn't shrinking, but the productive work is concentrating. Firms and practitioners that have adopted AI are generating more output per designer; those that haven't are losing ground without necessarily knowing it yet.

The Variable Isn't the Tool. It's How You Frame Your Work.

Here's the part most displacement coverage skips.

If your stomach has tightened reading this, that's rational. The displacement stories are real and the data is specific. But Harvard Business School researchers found something that changes the frame entirely. Tracking AI's effect across hundreds of firms, they found that jobs in the top quartile of automation exposure saw a 24% decline in AI-related skills demanded. Jobs structured around interpretation and judgment, however, saw a 15% increase in those same skills. The same technology. The same time period. Two opposite career trajectories.

The variable wasn't industry or job title. It was how the work was framed. Jobs described as output production competed on cost with AI and lost ground. Jobs described as interpretation and judgment became more valuable as AI raised the baseline for what clients expected.

This is where Dixie Willard's story stops being an interesting anecdote and becomes something structurally useful. She didn't wait to discover which category her work fell into. She actively moved it. By using ChatGPT as a language mirror — reflecting her own voice, her own positioning, her own client relationships back to her in compressed form — she reframed her practice from a service that could be partially automated to something AI cannot perform: the translation between designer, contractor, and homeowner that depends entirely on her specific accumulated judgment. "It sounds like me. It feels like me." That's not AI replacing her expertise. It's AI amplifying it.

The designers who end up on the wrong side of this shift won't necessarily be the ones who moved too slowly on tools. They'll be the ones who never made a deliberate choice about which kind of work they were doing — and let the automation decide for them by default.

What Successful Integration Actually Looks Like

Designers who are successfully integrating AI share a common structural move: they use AI to absorb documentation and iteration while investing the recovered time into judgment-layer work clients pay premiums for. The most forward-positioned among them have also shifted how clients find them in the first place.

Stacy Thorwart spent a decade leading global AI strategy at Steelcase before launching The Intelligent Designer, a training platform that has now reached more than 8,500 designers worldwide. Her framing of the underlying skill is precise: "AI literacy isn't just about writing better prompts. It's about understanding how designers collaborate with intelligent tools." The distinction matters. Designers who treat AI as a prompt machine are still doing the same work faster. Designers who treat it as a collaborative layer are doing fundamentally different work — and billing for the judgment that shapes it, not the hours that produce it.

AI literacy isn't just about writing better prompts. It's about understanding how designers collaborate with intelligent tools.
— Stacy Thorwart, Founder, The Intelligent Designer

On the client acquisition side, something unexpected is already happening. San Diego designer Rachel Moriarty was hired for a Padres-themed condo project after the client discovered her through ChatGPT — not Google, not Instagram. She called it "a wild, modern marketing moment." A parallel case emerged in Houston around the same period. The referral channel is shifting: clients are asking AI assistants to recommend designers, and the designers who have structured their online presence for that context are appearing in those answers while SEO-only studios are not.

The integration playbook runs on two fronts. Internal: use AI to reclaim time from documentation and iteration so that judgment-layer work can expand. External: ensure your firm's positioning is legible to AI-mediated client discovery, not just search engines. The second front is the one most working designers haven't addressed. If a potential client asks ChatGPT to recommend a designer specializing in your niche, does your name appear? That's now a concrete competitive question — not a theoretical one.

This same dynamic applies across any knowledge-based professional service. Therapists, accountants, consultants, architects — if your clients are using AI assistants to research and shortlist providers, and your positioning isn't structured for that context, you're invisible to a growing segment of the market regardless of your actual quality.

The Choice the Data Keeps Pointing Back To

The anonymous designer who went quiet in that pitch meeting wasn't wrong to have concerns. Her instinct — that AI was being used to devalue her craftsmanship — may have been exactly right about her specific employer's intentions. But walking into a "no AI" boutique only delays the question. It doesn't answer it.

Willard faced the same tools, the same disruption, and asked a different question. Not "what is this doing to my work?" but "what can I make it do?" The week she spent rebuilding Poised & Plumb wasn't a productivity hack. It was a clarification exercise — using AI as a mirror to find out what only she could say.

The BIID, the British Institute of Interior Design, identified a risk that cuts the other way too: replacing junior designers with AI wholesale could produce "a smaller and less skilled workforce," because the apprenticeship pipeline that builds senior judgment depends on junior work existing. The designers most protected from displacement are the ones who have accumulated enough judgment to direct AI rather than be directed by it. That judgment takes time to build. The window to start is now.

Here is a self-audit you can run this week. Pull the last five client projects you completed and list every major task you performed. For each task, ask one question: could a well-prompted AI have produced a serviceable version of this in under an hour? Tasks that answer yes are your automation exposure. Tasks that answer no are your protected core. If your exposure list is longer than your protected list, you have a specific problem with a specific solution — restructure your service offerings so that the judgment-layer tasks are what clients are explicitly paying for, not bundled invisibly inside hourly rates.

The designers who end up on the wrong side of this shift won't be the ones who moved too slowly on tools. They'll be the ones who never decided which side they were on — and let the algorithm make the call for them.

Recommended Tools & Resources