State of AI in Architecture 2026: 63 Stats from 11,000 Architects
Industry Report · April 2026

The 2026 State of AI in Architecture

A synthesized industry report aggregating findings from the leading surveys, studies, and benchmarks on artificial intelligence adoption in the architectural profession — covering 11,000+ respondents across 19 institutional sources.

  • 19 sources aggregated
  • ~11,000 respondents
  • Published 17 April 2026
  • 18 min read
  • By
46%
Global AI adoption among architects
Chaos/Architizer 2024, n=1,200+
59%
UK practices using AI
RIBA 2025, n≈500
6%
US architects using AI regularly
AIA 2025, n=541
74%
Plan to expand AI use in 12 months
Chaos/Architizer 2024
$20K
Revenue premium per FTE at AI-adopting firms
Monograph 2026, n=16,000+

Executive Summary: AI adoption in architecture at a glance

Between 2024 and early 2026, artificial intelligence moved from novelty to embedded practice across the global architectural profession — though the pace and depth of adoption vary dramatically by geography, firm size, and source of measurement. Aggregating data from Chaos/Architizer, RIBA, AIA, Autodesk, NBS, Monograph, and the Architects' Journal reveals a profession mid-transition.

The headline numbers: 46% of architects globally report currently using AI tools Chaos/Architizer 2024, n=1,200+, rising to 59% of UK practices RIBA 2025 and as low as 6% regular usage among US AIA members AIA 2025 — a gap that is itself one of the most revealing data points in this report. 74% of AI-using architects plan to expand usage over the next 12 months Chaos/Architizer. 44% use AI for concept imagery and 35% for rapid design variation Chaos 2024–25. 85% of AI users report measurable time savings, with more than half saving at least 5 hours per week Chaos/Architizer 2026. 86% believe AI will play a significant role in the future of architecture practice, yet 60% of current users have received no formal training.

The profession is adopting — thoughtfully, unevenly, and with considerably more skepticism than headlines suggest.

How this AI architecture report was assembled

This report is a secondary aggregation, not primary research. creativetoolsai.com did not conduct its own survey; instead, we synthesised publicly available findings from 19 institutional sources published between January 2024 and March 2026, prioritising surveys with disclosed sample sizes and methodologies. Total sample coverage: ~11,000 respondents.

1. What percentage of architects use AI in 2026?

Between 40% and 60% of architects globally have tried AI tools at least once professionally, 10–20% have embedded AI into routine workflow, and fewer than 10% use it as a core daily tool. The headline figure depends on who is asking, who is answering, and what counts as "using AI": global surveys cluster around 46%, UK practices at 59%, US architects at just 6% regular use.

UK practicesRIBA 2025 · n≈500
59%
GlobalChaos/Architizer 2024 · n=1,200+
46%
US (regular use)AIA 2025 · n=541
6%
Figure 1 — Architect AI adoption by region. UK practices lead at 59% (RIBA 2025); the global figure is 46% (Chaos/Architizer 2024); only 6% of US architects use AI regularly (AIA 2025). The apparent gap is partly methodological — see explanation below.

Global and multi-national survey results

  1. Chaos + Architizer, The State of AI in Architecture (2024, n=1,200+, 70+ countries): 46% of architects currently use AI tools; an additional 24% plan to adopt soon. Chaos/Architizer 2024
  2. Chaos + Architizer, 2024–25 State of Architectural Visualization (n=1,000+, 75 countries): 44% use AI for concept imagery; 11% have fully embedded AI in their visualization workflow. Chaos ArchViz 2024–25
  3. Chaos + Architizer, 2026 AI in Architecture Report (n≈800, surveyed Nov 2025): 64% have experimented with AI tools; 20% have fully embraced them; 74% intend to increase AI usage in the next 12 months. Chaos/Architizer 2026
  4. Architizer webinar update, May 2025: 56% of respondents were "actively using AI tools" in their workflows — framed as a significant jump year-over-year. Architizer 2025

UK-specific survey results

  1. RIBA AI Report 2024 (n≈500): 41% of UK architectural practices used AI for at least the occasional project. RIBA 2024
  2. RIBA AI Report 2025 (n≈500): 59% of UK practices use AI — an 18-point rise year-over-year. The share of practices that "never use AI" fell from 59% to 41%. The share using AI for "most projects" more than doubled, from 4% to 9%. RIBA 2025
  3. Architects' Journal Changes in Practice survey (Oct 2025, n≈200, mainly UK): 64% use text-based AI tools weekly; 40% use them daily. AJ 2025
  4. NBS Digital Construction Report 2025 (n≈550, including 200+ architects): more than two in five industry professionals have integrated AI into daily work — a fivefold increase from under 10% in 2020. NBS 2025

US-specific survey results

  1. AIA Artificial Intelligence Adoption in Architecture Firms (n=541, June–July 2024, published March 2025): only 6% of individual architects regularly use AI. 53% have experimented with AI in some capacity. 8% of firms have formally implemented AI; 39% of non-adopters are not interested in doing so. AIA 2025
"Somewhere between 40% and 60% of architects globally have tried AI tools at least once in a professional context; 10–20% have embedded AI into routine workflow; and fewer than 10% use it as a core daily tool." — Synthesis finding, this report

Which firm sizes adopt AI fastest?

83%
64%
48%
Large
50+ staff
Medium
10–50 staff
Small
<10 staff
Figure 2 — UK AI adoption by firm size. RIBA 2025 (n≈500). Large firms lead on formal implementation, but note the counter-pattern below.

Firm size is the single most consistent predictor of adoption across every source:

  • RIBA 2025: 83% adoption at large practices (50+ staff), 64% at medium (10–50), 48% at small (<10). RIBA 2025
  • AIA 2025: 8% of firms have implemented AI, "driven significantly more by large firms (50+ employees), the early adopters in this space." AIA 2025
  • Chaos/Architizer 2024–25 ArchViz: 60% of respondents came from practices of up to 19 employees; larger firms lead adoption but smaller firms and freelancers are experimenting actively. Chaos 2024–25
  • D5 Render survey, 2025 (n=665, 100+ countries): reports a counter-pattern — small studios adopting AI faster than large firms, attributing the reversal to compliance, IT control, and workflow complexity at bigger firms. D5 Render 2025

These two patterns are not contradictory. Large firms are more likely to have formally implemented AI with IT policies and paid subscriptions; small firms and solo practitioners are more likely to be personally experimenting with consumer tools like ChatGPT and Midjourney without institutional oversight.

[INTERNAL LINK: firm-size deep dive → /ai-adoption-by-firm-size]

2. Which AI tools do architects actually use?

ChatGPT dominates architecture-AI usage at 77% among UK architects, more than three times Midjourney's 22% (Architects' Journal 2025, n≈200). Text-based LLMs are rising while image generators are in decline — a "flight to utility" as architects prioritise reliable daily wins over creative experimentation. BIM-integrated AI tools trail far behind despite strong vendor momentum.

Most-used AI tools in architecture (UK ranking)

ChatGPTand comparable LLMs
77% ▲ from ~60%
Midjourney
22% ▼ from 42%
Stable Diffusion
8% ▼ from 21%
Figure 3 — "Flight to utility." Text tools rise while image generators decline. Source: Architects' Journal Changes in Practice survey, October 2025, n≈200 UK architects.
UK architect AI-tool usage, October 2025
ToolShare of UK architects usingYoY change
ChatGPT (and comparable LLMs)77%▲ up from ~60%
Midjourney22%▼ from 42% (2024)
Stable Diffusion8%▼ from 21% (2024)

Source: Architects' Journal Changes in Practice survey, October 2025.

This pattern — text-based LLMs rising, image generators declining in dominance — is one of the most important 2024-to-2025 shifts in the data. The AJ's analysis calls it "a flight to utility": architects are finding that text tools (specification writing, report drafting, code research) produce reliable daily wins, while image tools require significant skill investment to integrate into professional deliverables.

BIM-integrated AI tools (Veras, Maket, TestFit, Arko) do not yet appear at the top of usage rankings in any aggregated survey — but Chaos's own product data shows rapid growth of Veras, which as of Q1 2026 integrates with seven major BIM/CAD platforms (Revit, SketchUp, Rhino, Vectorworks, Archicad, Forma, Allplan). Chaos 2026

[INTERNAL LINK: AI render tools benchmark → /ai-render-tools-benchmark]

What do architects actually use AI for?

Concept images & early design ideas
44%
Quick variations of design options
35%
Photorealism enhancement
32%
Image quality optimization
26%
Figure 4 — Top AI use cases among architects using AI. Chaos/Architizer State of ArchViz 2024–25, n=1,000+.
AI use cases, Chaos/Architizer ArchViz 2024–25 (n=1,000+)
Use case% of AI-using architects
Concept images and early design ideas44%
Quick variations of design options35%
Photorealism enhancement32%
Image quality optimization26%

From the Chaos/Architizer 2024 State of AI in Architecture report, AI users reported these specific activities: imagery from text prompts 74%; image editing 61%; images from model inputs 50%. Chaos/Architizer 2024

And 68% of designers identified conceptualization and pre-design as the project stages with the highest potential for AI improvement, compared with only 34% for design development, 29% for construction documentation, and low single digits for planning/review. Chaos/Architizer 2024

The RIBA 2025 report confirms the same pattern in the UK: AI use is "most prevalent in tasks such as early design visualisations and specification writing, where efficiency gains are immediate. More advanced applications, such as performance simulation or environmental modelling, are emerging but less established." RIBA 2025

The Architects' Journal found roughly 39% of architects use AI at the RFP/bid stage — the same share that uses it for concept design — and more than 40% use it for practice administration. AJ 2025

3. How much time and money does AI save architecture firms?

85% of AI-using architects report measurable time savings, more than half save at least 5 hours per week, and AI-adopting firms generate $20,000 more net revenue per FTE than baseline — a roughly 4:1 return. The most credible firm-level financial data comes from Monograph's 2026 Benchmarks Report (n=16,000+), drawn from operational data rather than self-reports.

Time savings

  • Chaos/Architizer 2026: 85% of AI-using architects report measurable efficiency gains. More than half save at least 5 hours per week through AI-assisted workflows. Chaos 2026
  • Chaos/Architizer 2024: 60% of AI users cite improved efficiency as a top benefit; 57% cite enhanced creativity; 53% cite unlocking new creative workflows. Only 12% report no significant impact. Chaos 2024
  • RIBA 2024: Of UK architects using AI, 43% said it had made the design process more efficient. RIBA 2024

Firm-level financial impact

AI-adopting firms
$210K
Net revenue per FTE
Baseline firms
$190K
Net revenue per FTE
+$20,000 premium per FTE Roughly 4:1 return on a $5,000 cost increase ($143K vs. $138K)
Figure 5 — Monograph 2026 Architecture & Engineering Business Benchmarks Report, operational data from 16,000+ A&E professionals. US-dominated sample, 90% of firms have 1–50 employees.
Monograph 2026 firm-level benchmarks (n=16,000+)
MetricAI-adopting firmsBaseline firmsDelta
Net revenue per FTE$210,000$190,000+$20K
Cost per FTE$143,000$138,000+$5K
Top-quartile net revenue per FTE$269,000$228,000+$41K
Operations-role utilization+14%baseline+14 pts
Back-office salary costs−11%baseline−11%

These Monograph figures come from operational data (not survey self-reports) and are the most credible firm-level financial numbers in this report. They are US-dominated and skewed toward small-to-midsize firms, since 90% of firms in the Monograph dataset have 1–50 employees. Monograph 2026

Broader AEC productivity context

McKinsey's widely cited figures on AI productivity in construction and E&C — up to 20% productivity gain, 15% cost reduction, 30% delivery-time improvement — are asset-owner and contractor figures, not architect-specific. They apply to projects where AI is integrated across the full construction lifecycle, not to design-stage AI use alone. McKinsey, via multiple 2024–25 analyses

Spending on AI tools

There is no reliable public benchmark for average monthly or annual AI-tool spending at architecture firms. What exists:

  • Bluebeam AEC Outlook 2025: 70% of AEC firms allocate some IT budget to AI; roughly a quarter dedicate 20–25% of tech spend to AI. Bluebeam 2025
  • Architects' Journal 2025: 60% of UK architectural practices have no formal AI budget — despite 64% of their architects using text-based AI weekly. AJ 2025
"The gap between widespread personal AI use and the absence of formal firm-level budgets is one of the clearest operational findings in the 2025 data." — Synthesis finding, this report

[INTERNAL LINK: cost per render analysis → /cost-per-render-analysis]

4. What are architects most concerned about with AI?

Near-universal concerns dominate: 94% of US architects cite AI inaccuracies and unintended consequences, 93% privacy and security, 90% authenticity and transparency (AIA 2025, n=541). Practically, 48% of architects globally name inconsistent output quality as their biggest barrier, and 60% have received no formal training. Trust in AI actually fell 11 points year-over-year at leadership level (Autodesk 2025).

Quality and reliability

  • 48% of architects cite inconsistent or poor output quality as their biggest AI challenge. Chaos 2026
  • 76.8% of architects struggle with inconsistent results when using general-purpose image generators for multi-view architectural visualization (CGarchitect survey, cited by Chaos 2026). Chaos 2026
  • 69% of AI users report being only "somewhat" satisfied with what AI produces; relatively few describe their experience as outright negative. Chaos 2026

Training and skills

  • 60% of AI-using architects have received no formal training and are "learning on the fly." Chaos 2024
  • Fewer than 1 in 5 UK practices has invested in AI R&D; only 15% have a formal AI policy. 53% expect to adopt an AI policy within two years. RIBA 2025
  • Insufficient worker skills is the biggest barrier to AI integration overall, per Deloitte's State of AI in the Enterprise 2026 (n=3,235 leaders across 24 countries). Deloitte 2026

Trust and accuracy (AIA 2025)

Inaccuracies of AI outputs
94%
Unintended consequences
94%
Privacy and security
93%
Authenticity of AI-generated content
90%
Lack of transparency
90%
Figure 6 — Near-universal concern. AIA Artificial Intelligence Adoption in Architecture Firms 2025, n=541 US architects.
US architect concerns about AI (AIA 2025, n=541)
Concern% of architects citing
Inaccuracies of AI outputs94%
Unintended consequences94%
Privacy and security93%
Authenticity of AI-generated content90%
Lack of transparency90%

Cost, integration, and institutional friction

  • Chaos/Architizer 2024: top non-quality barriers included software integration issues, time constraints for testing, and limited training resources. Chaos 2024
  • Autodesk 2025 State of Design & Make: Trust in AI has fallen — only 65% of design-and-make leaders now express trust in AI in their field, an 11-point decrease year-over-year. 48% believe AI will destabilize their industry, up from 41% in 2024. Just 69% believe AI will enhance their industry, down 12 points from 2024. Autodesk 2025

Client perception

No major survey has isolated client perception of AI-generated deliverables as a distinct barrier with publishable numbers. Chaos's 2026 analysis notes that "clients now arrive with their own AI-generated visuals" — suggesting a shifting dynamic more than a simple barrier. This is a meaningful data gap (see Section 8).

5. How does AI adoption vary by region?

Every major architect AI survey is North-America– and UK-dominated. The UK leads at 59% of practices, global surveys cluster at 46%, and Asia-Pacific AEC AI use grew from 26% to 37% between 2023 and 2024. MENA, Latin America, and Sub-Saharan Africa remain essentially unmeasured — a structural gap in the evidence base.

North America
Well-measured
Europe / UK
Well-measured
Asia-Pacific
Partly measured
MENA
Unmeasured
Latin America
Under-measured
Sub-Saharan Africa
Unmeasured
Well-measured Partly measured Under- / unmeasured
Figure 7 — Global coverage of architect AI-survey data. The Chaos/Architizer 2026 sample breaks down as 50% North America, 22% Europe, 10% Asia, 18% other. Oceania and MENA have no meaningful representation in any architect-specific survey.
Architect AI adoption by region — headline figures
RegionHeadline AI adoption figureSource
United Kingdom59% of practices use AI for at least the occasional project (2025); 77% use ChatGPTRIBA 2025; AJ 2025
United States6% of architects regularly use AI; 53% have experimented; 8% of firms have formally implementedAIA 2025
Global / multi-country46% currently use AI; 64% have experimentedChaos/Architizer 2024 and 2026
Asia-Pacific37% of AEC firms use AI/ML (2024), up from 26% (2023); 6.2 digital technologies adopted on average per firmDeloitte/Autodesk APAC 2024
North America (AEC broad)70% of firms allocate some IT budget to AI; 55% report active AI use in at least one project phaseBluebeam AEC Outlook 2025

Geographic coverage gap

The Chaos/Architizer 2026 report — the single most comprehensive recent architect-specific survey — breaks down respondents as follows: 50% North America, 22% Europe, 10% Asia, remaining 18% across Africa, South America, and Oceania. Chaos 2026

No major architect-specific survey has meaningful MENA representation. No major survey breaks out Latin America, Sub-Saharan Africa, or Oceania as stand-alone regional slices. Asia is typically treated as a single aggregate block.

Market forecasts

The generative AI in architecture market is projected to grow from $1.47 billion (2025) to $8 billion (2030), a 40.2% CAGR, with North America the largest region in 2025 and Asia-Pacific the fastest-growing through 2030. The Business Research Company, 2026

6. How do architects feel about AI — excited, cautious, or resistant?

The emotional profile of the profession is a cautious-curious majority, with committed adopters (~20%) and committed refusers (~20%) flanking on either side. 86% believe AI will play a significant role in architecture's future, yet only 20% have "fully embraced" it. 39% of non-adopting US firms are actively not interested.

~20%
Fully embraced
~60%
Cautious / experimenting
~20%
Resistant
Chaos 2026: 20% "fully embraced"
Chaos 2026: 64% experimented · AIA 2025: 78% want to learn more
AIA 2025: 39% not interested
Figure 8 — Approximate architect attitudes toward AI, synthesised across Chaos/Architizer 2026, AIA 2025, and RIBA 2025.

The neat three-part split does not exist as such in any single source but can be approximated by combining answers across surveys:

  • Excited / highly engaged: 20% of Chaos/Architizer 2026 respondents have "fully embraced" AI in their workflow; 86% believe AI will play a significant role in the future of architecture Chaos 2024.
  • Cautious / experimenting: 64% of Chaos/Architizer 2026 respondents have experimented with AI but are not fully committed. Among AIA 2025 respondents, 78% want to learn more about AI — but the same 78% also express concerns. 35% of non-adopters in Chaos 2026 will only adopt AI "when it clearly adds value."
  • Resistant: 39% of non-adopting US firms in AIA's 2025 survey reported being not interested in exploring AI. 41% of UK practices still "never use AI" RIBA 2025. Under 5% of UK architects have tried AI and then stopped.

7. What is the future of AI in architecture (2026–2027)?

74% of architects plan to increase AI use in the next 12 months, the architecture-AI market is projected to grow from $1.47B to $8B by 2030, and 46% of design-and-make leaders list AI proficiency as a top hiring priority. UK architects expect Stage 2 (concept design) to be the stage most likely to become "fully automated" in the next decade.

UK adoption trajectory

<10%
2020 · NBS
41%
2024 · RIBA
59%
2025 · RIBA
↑ projected
2026
Figure 9 — UK architectural AI adoption trajectory. Points: NBS 2020 (baseline); RIBA 2024 (41%); RIBA 2025 (59%). 2026 projection based on stated 12-month expansion intentions.

Intended 12-month adoption

  • 74% of architects plan to increase AI use over the next 12 months. Chaos 2024 & 2026
  • 24% of non-adopters plan to adopt soon. Chaos 2024
  • 53% of UK practices expect to have an AI policy within two years. RIBA 2025
  • 47% of UK practices anticipate investing in AI R&D within two years. RIBA 2025
  • 78% of UK architects expect RIBA Plan of Work Stages 2–4 to be transformationally or significantly affected by AI in the next decade; Stage 2 (concept design) is most expected to become "fully automated." RIBA Future Business of Architecture 2025

Investment trajectory

Hiring and skills

  • 46% of Design & Make leaders list AI proficiency as a top hiring priority over the next three years — up 5 points from 2024. Autodesk 2025
  • 61% of leaders say candidates with the right technical skills are difficult to find, up 16 points year-over-year. Autodesk 2025

Licensed vs. student vs. enthusiast breakdown

Reliable data does not exist for the licensed/student/enthusiast split specifically for AI in architecture. The closest proxies: of AIA 2025 respondents, 51% were firm partners or principals; 36% were licensed architects; the remaining 13% were other roles. Technology decision-makers under 35 were significantly more likely to have tried image generators (66% vs. 41% for those over 50) and chatbots (87% vs. under-half for those over 50). AIA 2025this is a significant data gap (see Section 8).

8. Data gaps: where primary research would add real value

Five meaningful data gaps emerged through this aggregation. Each represents an opportunity for genuine new primary research: client perception, MENA/LATAM/Sub-Saharan Africa adoption, per-firm AI spending, student-vs-professional usage patterns, and AI use in construction documentation. These five gaps are the white space where a primary 500-architect survey would deliver findings not already available.

9. Ten standalone quotable statistics

Ten pull-quote-ready statistics, each under 25 words with inline source, designed to be extracted by Google AI Overviews, ChatGPT, Perplexity, and Claude — or screenshotted and shared on social.

1. 46% of architects globally currently use AI tools, with 74% planning to expand usage in the next 12 months. Chaos/Architizer, The State of AI in Architecture, 2024 — n=1,200+
2. UK architectural practices using AI rose from 41% in 2024 to 59% in 2025 — an 18-point jump in a single year. RIBA AI Report 2025 — n≈500
3. Only 6% of US architects regularly use AI, and just 8% of US firms have formally implemented it — the lowest rate measured. AIA Journey to Specification, 2025 — n=541
4. 44% of architects use AI for concept imagery; 35% for rapid design variation; 32% for photorealism enhancement. Chaos/Architizer State of ArchViz 2024–25 — n=1,000+
5. ChatGPT use (77%) is more than three times Midjourney (22%) among UK architects; Midjourney has fallen from 42% a year earlier. Architects' Journal Changes in Practice, October 2025 — n≈200
6. 85% of AI-using architects report time savings; more than half save at least 5 hours per week. Chaos/Architizer 2026 AI in Architecture Report — n≈800
7. Architecture firms that adopt AI generate $20,000 more net revenue per employee than non-adopting peers — a 4:1 return. Monograph 2026 A&E Benchmarks Report — n=16,000+
8. 60% of AI-using architects have received no formal training; 60% of UK practices have no formal AI budget. Chaos/Architizer 2024; Architects' Journal 2025
9. 94% of US architects cite AI inaccuracies and unintended consequences as top concerns — the highest across five risk categories. AIA Journey to Specification, 2025
10. 86% of architects believe AI will play a significant role in architecture's future — yet only 20% have fully embraced AI today. Chaos/Architizer 2024; Chaos/Architizer 2026

Frequently asked questions about AI in architecture

What percentage of architects use AI in 2026?

46% of architects globally currently use AI tools, according to the Chaos/Architizer 2024 survey of 1,200+ architects. UK adoption is higher at 59% of practices (RIBA 2025), while only 6% of US architects use AI regularly (AIA 2025, n=541). Between 40% and 60% have tried AI at least once professionally.

Which AI tools do architects use most?

ChatGPT dominates: 77% of UK architects use it, versus 22% for Midjourney and 8% for Stable Diffusion (Architects' Journal 2025, n=200). Text-based LLMs are rising while image generators are declining — a "flight to utility" as architects prioritise reliable daily wins over creative experimentation.

How much time do architects save with AI?

85% of AI-using architects report measurable time savings, and more than half save at least 5 hours per week (Chaos/Architizer 2026 AI in Architecture Report, n=800). 60% cite improved efficiency as a top benefit; 57% cite enhanced creativity; only 12% report no significant impact.

Why is US AI adoption so low compared to the UK?

RIBA's 59% UK figure and AIA's 6% US figure measure different things: RIBA asked about "at least occasional" practice use, AIA asked about "regular" individual use. The AIA sample also skews older (55% over 50). Genuine UK-vs-US variation exists, but methodology explains most of the gap.

Do architecture firms make more money using AI?

Yes. Monograph's 2026 Benchmarks Report (n=16,000+) shows AI-adopting firms generate $210,000 net revenue per FTE versus $190,000 at baseline — a $20,000 premium on a $5,000 cost increase, roughly a 4:1 return. Top-quartile AI firms reach $269K per FTE.

What are architects most concerned about with AI?

Near-universal: 94% of US architects cite AI inaccuracies and unintended consequences, 93% privacy and security, 90% authenticity of outputs, and 90% lack of transparency (AIA 2025, n=541). 48% of architects globally name inconsistent output quality as their biggest practical barrier.

Which firm sizes adopt AI fastest?

Large UK practices lead at 83%, medium 64%, small 48% (RIBA 2025). But D5 Render's 2025 survey found small studios experimenting faster than large firms, citing compliance and IT friction at scale. Large firms implement formally; small firms adopt personally through consumer tools.

Are clients accepting AI-generated architecture?

No major survey has isolated client perception with publishable numbers — a genuine data gap. Chaos's 2026 report notes clients "now arrive with their own AI-generated visuals," suggesting adoption on the client side too, but quantitative baselines do not yet exist for this question.

Is Midjourney still popular with architects?

Midjourney usage among UK architects has nearly halved — from 42% in 2024 to 22% in 2025 (Architects' Journal). Stable Diffusion fell from 21% to 8%. ChatGPT rose to 77%. Image generators are being displaced by text tools for specification, drafting, and research tasks.

Will AI replace architects?

Not fully. 86% of architects believe AI will play a significant role in the profession's future (Chaos/Architizer 2024), and 78% of UK architects expect RIBA Plan of Work Stages 2–4 to be significantly affected in the next decade. Stage 2 (concept design) is most expected to become "fully automated."

How much do architects spend on AI tools?

No reliable per-firm benchmark exists — a documented data gap. 60% of UK practices have no formal AI budget (Architects' Journal 2025) despite 64% using text-based AI weekly. 70% of AEC firms allocate some IT budget to AI; roughly a quarter dedicate 20–25% of tech spend to AI (Bluebeam 2025).

What is the future of AI in architecture?

74% of architects plan to increase AI use in the next 12 months (Chaos/Architizer). The generative-AI architecture market is projected to grow from $1.47B (2025) to $8B (2030) at 40.2% CAGR (The Business Research Company). 46% of Design & Make leaders list AI proficiency as a top hiring priority (Autodesk 2025).

Dataset appendix: 63 statistics with sources

Every figure referenced in this report, with source and date. Sample sizes given where disclosed. This appendix is machine-readable and marked up as schema:Dataset in the page JSON-LD.

Full aggregated dataset, 63 rows
#StatisticSourceYearSample
146% of architects currently use AI toolsChaos/Architizer State of AI in Architecture20241,200+
224% of non-users plan to adopt AI soonChaos/Architizer20241,200+
374% will likely increase AI use in next 12 monthsChaos/Architizer20241,200+
486% believe AI will significantly impact architecture's futureChaos/Architizer20241,200+
560% of AI users have had no formal trainingChaos/Architizer20241,200+
668% identify conceptualization as highest-potential AI stageChaos/Architizer20241,200+
744% use AI for concept imageryChaos/Architizer State of ArchViz2024–251,000+
835% use AI for rapid design variationChaos/Architizer ArchViz2024–251,000+
932% use AI for photorealismChaos/Architizer ArchViz2024–251,000+
1026% use AI for image qualityChaos/Architizer ArchViz2024–251,000+
1111% of firms have embedded AI in visualization workflowChaos/Architizer ArchViz2024–251,000+
1220% YoY rise in AI experimentation excitementChaos/Architizer ArchViz2024–251,000+
1364% have experimented with AI; 20% fully embracedChaos/Architizer 2026 report2026~800
1485% report measurable time savingsChaos/Architizer 20262026~800
15Over 50% of users save 5+ hours per weekChaos/Architizer 20262026~800
1669% "somewhat satisfied" with AI outputChaos/Architizer 20262026~800
1748% cite inconsistent output quality as biggest challengeChaos/Architizer 20262026~800
1843% identify conceptual/pre-design as highest AI-value stageChaos/Architizer 20262026~800
1941% UK AI adoptionRIBA AI Report 20242024~500
2059% UK AI adoptionRIBA AI Report 20252025~500
2183% large-firm UK adoptionRIBA 20252025~500
2264% medium-firm UK adoptionRIBA 20252025~500
2348% small-firm UK adoptionRIBA 20252025~500
249% use AI on most projects (up from 4%)RIBA 20252025~500
2515% of UK practices have formal AI policyRIBA 20252025~500
2653% expect AI policy within two yearsRIBA 20252025~500
2747% anticipate AI R&D investment within two yearsRIBA 20252025~500
2888% of UK architects think AI will be more important in next 10 yearsRIBA Future Business of Architecture2025N/D
296% of US architects regularly use AIAIA Journey to Specification2025541
308% of US firms have implemented AIAIA 20252025541
3153% of US architects have experimented with AIAIA 20252025541
3239% of non-adopting US firms not interested in AIAIA 20252025541
3378% want to learn more / 78% have concernsAIA 20252025541
3494% cite inaccuracy concernAIA 20252025541
3594% cite unintended-consequences concernAIA 20252025541
3693% cite privacy/security concernAIA 20252025541
3790% cite authenticity concernAIA 20252025541
3890% cite lack-of-transparency concernAIA 20252025541
3948% feel AI will be necessary to their careerAIA 20252025541
40Under-35 architects 66% tried image generators (vs. 41% over-50)AIA 20252025541
4177% of UK architects use ChatGPTAJ Changes in Practice2025~200
4264% use text-based AI weeklyAJ 20252025~200
4340% use text-based AI dailyAJ 20252025~200
4422% use Midjourney (down from 42%)AJ 20252025~200
458% use Stable Diffusion (down from 21%)AJ 20252025~200
4660% of practices have no formal AI budgetAJ 20252025~200
475× increase in AI adoption among NBS respondents 2020–2025NBS Digital Construction Report2025~550
4869% of leaders say AI will enhance their industry (down 12 pts)Autodesk State of Design & Make20255,594
4965% trust in AI (down 11 pts)Autodesk 202520255,594
5048% expect AI to destabilize their industry (up from 41%)Autodesk 202520255,594
5146% prioritize AI skills in hiringAutodesk 202520255,594
5261% struggle to find technical talentAutodesk 202520255,594
5339% use AI for sustainabilityAutodesk 202520255,594
54AI firms earn $210K per FTE vs $190K baselineMonograph 2026 A&E Benchmarks202616,000+
55AI firms cost $143K per FTE vs $138K baselineMonograph 2026202616,000+
5614% higher utilization on operations roles at AI firmsMonograph 2026202616,000+
5711% lower back-office salary cost at AI firmsMonograph 2026202616,000+
58Top-quartile AI firms hit $269K net revenue per FTEMonograph 2026202616,000+
59Gen-AI in architecture market: $1.47B (2025) → $8B (2030), 40.2% CAGRThe Business Research Company2026Forecast
6037% of APAC AEC firms use AI/ML (up from 26%)Deloitte/Autodesk APAC, via Buildcheck2024N/D
6170% of AEC firms allocate some IT budget to AIBluebeam AEC Outlook2025N/D
6255% active AI use in at least one project phaseBluebeam 20252025N/D
6346% of Q1 2025 contech investment went to AINymbl Ventures / Construction Dive2025Industry

Full source bibliography

Nineteen sources cited, each marked up as schema:Report or schema:Article via isBasedOn in the page JSON-LD. Click any card to view source details; open the full URL to read the original study.

  1. 1. Chaos + Architizer — The State of AI in Architecture 2024 · 1,200+ respondents · 70+ countries blog.chaos.com/the-state-of-ai-in-architecture…
  2. 2. Chaos + Architizer — State of Architectural Visualization 2024–25 2025 · 1,000+ respondents blog.chaos.com/2025-state-of-archviz-report-insights
  3. 3. Chaos + Architizer — How AI Is Reshaping Architectural Design & Visualization in 2026 Survey Nov 2025, released 2026 · ~800 respondents chaos.com/ai-in-architecture-report-2026
  4. 4. RIBA — Artificial Intelligence Report 2024 Feb 2024 · ~500 members riba.org/…/riba-artificial-intelligence-report-2024
  5. 5. RIBA — AI Report 2025 2025 · ~500 members riba.org/…/riba-ai-report-2025
  6. 6. RIBA — Future Business of Architecture Survey / White Paper on AI 2025 riba.org/…/future-business-of-architecture/artificial-intelligence
  7. 7. AIA — Artificial Intelligence Adoption in Architecture Firms March 2025 · 541 respondents aia.org/…/artificial-intelligence
  8. 8. Autodesk — 2025 State of Design & Make April 2025 · 5,594 leaders, AECO / D&M / M&E globally autodesk.com/design-make/research/state-of-design-and-make-2025
  9. 9. Architects' Journal (with Deltek) — Changes in Practice Survey 2025 Oct 2025 · ~200 UK respondents architectsjournal.co.uk/…/ai-in-practice
  10. 10. NBS (Hubexo) — Digital Construction Report 2025 2025 · 550+ professionals, 200+ architects ribaj.com/intelligence/…/digital-construction-report-2025
  11. 11. Monograph — 2026 Architecture & Engineering Business Benchmarks Report 2026 · 16,000+ A&E professionals (operational data) monograph.com/benchmark
  12. 12. Deloitte — State of AI in the Enterprise 2026 Surveyed Aug–Sept 2025 · 3,235 leaders, 24 countries deloitte.com/…/state-of-ai-in-enterprise
  13. 13. McKinsey & Company — AEC and construction productivity analysis Various, 2023–2025 mckinsey.com/…/artificial-intelligence-construction
  14. 14. D5 Render — The State of AI in Architecture & Design 2025 2025 · 665 respondents, 100+ countries d5render.com/posts/ai-in-architecture-2025-report
  15. 15. The Business Research Company — Generative AI in Architecture Market Report 2026 2026 · Market forecast researchandmarkets.com/…/generative-ai-in-architecture-market-report
  16. 16. Bluebeam — AEC Outlook 2025 2025 · Sample not publicly disclosed bluebeam.com
  17. 17. Fast Company — "Only 6% of architects are using AI regularly" March 2025 fastcompany.com/…/aia-2025-ai-report
  18. 18. Dezeen — "Only six per cent of architects regularly using AI says AIA study" March 2025 dezeen.com/2025/03/14/ai-architecture-study-american-architect
  19. 19. Designboom — "41% of architects are already using AI, RIBA report shows" March 2024 designboom.com/…/riba-latest-report-shows-41-percent-architects-using-ai
Built on Unicorn Platform