Architects, Except Landscape and Naval

17-1011.00
Median wage $99,280/yr106,770 employed (US)Rank #368 of 923 scored · top 40% by substitution

Plan and design structures, such as private residences, office buildings, theaters, factories, and other structural property.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution31
Exposure30
Augmentation67

Substitution — the headline: capability discounted by cost, barriers and adoption.

Exposure — technical capability alone, regardless of whether anyone deploys it.

Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.

Tasks on the substitution scale

24 rated tasks, binned by substitution score.

Position among all scored occupations

Distribution of 923 occupation scores; the marker is this occupation.

Tasks with substitution ≥ 70

0%

Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.

Why this score

The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.

Task automatabilityw 35%31

panel mean rating 2.2/5 → substitution pressure 31/100

Technical feasibility todayw 20%30

panel mean rating 2.2/5 → substitution pressure 30/100

Cost vs. human wagew 15%30

panel mean rating 2.2/5 → substitution pressure 30/100

Adoption barriersw 20%inverted — strong barriers lower the score34

panel mean rating 3.7/5 (barrier strength) → substitution pressure 34/100

Sector adoption velocityw 10%30

panel mean rating 2.2/5 → substitution pressure 30/100

Task breakdown (24 tasks)

Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.

Develop marketing materials, proposals, or presentations to generate new work opportunities.

66

CI 55–77 · exposure 62 · augmentation 88 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Architecture and professional services are moderately digitized and early-stage in generative AI adoption; some firms use AI for draft copy and renderings, but production deployment of end-to-end marketing automation remains uncommon and pilot-heavy.
Sector adoption velocityclaude-sonnet-53/5Architecture firms are small-to-midsize and generally slower AI adopters than pure information/professional services sectors, though marketing/business development functions are adopting generative AI faster than core design work.
Augmentation potentialclaude-haiku-4-5-202510014/5AI is already transforming productivity by rapidly generating draft text, visual concepts, and presentation structures that architects refine. This assistive use is widespread and materially speeds up proposal cycles while keeping human judgment and client relationships central.
Augmentation potentialclaude-sonnet-55/5AI substantially speeds up drafting, brainstorming, and refining proposals and presentations while architects retain control over strategy, firm voice, and client relationships.
Task automatabilityclaude-haiku-4-5-202510013/5AI can generate draft presentations, proposal text, and marketing copy at scale, but architectural work requires domain expertise, client-specific customization, and visual design that demands human oversight. The task is roughly half-automatable with significant setup for templates and brand guidelines.
Task automatabilityclaude-sonnet-54/5AI can draft marketing copy, proposal narratives, and presentation content from firm materials and project data, requiring mainly human review and customization, meeting the time-saving threshold for large portions of the work.
Adoption barriersclaude-haiku-4-5-202510012/5Marketing and proposal development face minimal legal or regulatory barriers. Client preference for human-vetted materials and organizational attachment to existing processes provide some friction, but nothing structurally prevents AI-assisted or automated substitution.
Adoption barriersclaude-sonnet-51/5No licensing, regulatory, or liability barriers exist for marketing materials since they are not architectural instruments of service requiring stamped professional sign-off.
Cost vs. human wageclaude-haiku-4-5-202510013/5AI inference and integration costs are low, but oversight by an architect (to ensure quality, accuracy, and brand fit) remains necessary, making total cost comparable to partial human effort rather than dramatically cheaper than full outsourcing.
Cost vs. human wageclaude-sonnet-54/5AI-assisted drafting of marketing content costs a small fraction of billable marketing staff or principal time, especially for iterative editing tasks.
Technical feasibility todayclaude-haiku-4-5-202510013/5Tools like GPT-4, design platforms, and presentation software can produce serviceable marketing materials, but material customization, visual coherence, and strategic positioning typically require human refinement. Products exist but fall short of fully autonomous, production-ready output without architect review.
Technical feasibility todayclaude-sonnet-54/5Generative AI tools (e.g., ChatGPT, Copilot, Canva AI) are already used in professional services firms including architecture practices to draft proposals, RFP responses, and pitch decks reliably, though final polish and firm-specific branding still need human input.

Calculate potential energy savings by comparing estimated energy consumption of proposed design to baseline standards.

65

CI 55–75 · exposure 62 · augmentation 75 · importance 2.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Architectural firms in the information/professional-services sector have rapidly adopted automated energy analysis tools and BIM-integrated workflows; major practices now routinely deploy such systems in design iteration.
Sector adoption velocityclaude-sonnet-53/5Architecture and AEC firms are adopting energy modeling and sustainability software at a moderate pace, with pilots and specialized consultants common but not yet universal integration into daily design workflows.
Augmentation potentialclaude-haiku-4-5-202510014/5AI augments architects significantly by running rapid parametric analyses, comparing multiple design variants, and flagging energy inefficiencies—enabling architects to explore designs faster and make better-informed decisions while retaining overall design authority.
Augmentation potentialclaude-sonnet-54/5AI-assisted energy modeling tools significantly speed up baseline comparisons and iteration on design options, letting architects explore more scenarios while retaining decision-making control.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI systems can reliably calculate energy consumption estimates, compare them to baseline standards, and quantify savings using standard methodologies and building codes. While the task requires translating design specifications into inputs for energy modeling, this is largely routine data mapping that AI can handle consistently.
Task automatabilityclaude-sonnet-53/5AI-driven energy modeling tools can calculate consumption and compare to baseline codes (e.g., ASHRAE 90.1) automatically, but require accurate building geometry, material, and system inputs plus interpretation, so full automation is partial.
Adoption barriersclaude-haiku-4-5-202510012/5Energy calculations do not require legal licensure or regulatory sign-off in most jurisdictions; however, professional judgment and liability considerations mean architects typically retain responsibility for validating results and interpreting findings in context.
Adoption barriersclaude-sonnet-52/5No licensing requirement mandates a human perform this specific calculation, though building code compliance documentation may need architect/engineer sign-off, creating moderate liability-driven oversight.
Cost vs. human wageclaude-haiku-4-5-202510014/5The computational cost of energy modeling and comparison is low relative to the loaded hourly wage of an architect; a single run may cost dollars in computation and integration overhead, while a human architect would spend hours on the same analysis.
Cost vs. human wageclaude-sonnet-53/5Software licenses and cloud computation are relatively cheap, but architect/engineer time to input accurate models and validate outputs still represents significant cost, keeping ratio near parity rather than order-of-magnitude savings.
Technical feasibility todayclaude-haiku-4-5-202510014/5Mature energy analysis software (e.g., EnergyPlus, DOE-2, specialized tools integrated into BIM platforms) demonstrably performs this calculation in production at architectural firms. AI can automate the comparison and savings calculation, though integration with existing design workflows varies by firm.
Technical feasibility todayclaude-sonnet-53/5Energy modeling software (eQuest, EnergyPlus, cove.tool, Autodesk Insight) is widely used in production and can automate comparisons, but often needs expert setup and validation, limiting reliability for complex designs.

Gather information related to projects' environmental sustainability or operational efficiency.

63

CI 56–70 · exposure 62 · augmentation 88 · importance 3.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architecture remains relatively traditional and project-based, with slower digital adoption compared to finance or software. Firms are piloting AI research tools, but few have integrated agents into standard workflows; adoption is in early stages despite obvious feasibility.
Sector adoption velocityclaude-sonnet-52/5Architecture firms are generally slower AI adopters compared to finance or software due to project-based, less digitized workflows, though sustainability data tools are gaining traction.
Augmentation potentialclaude-haiku-4-5-202510015/5AI significantly augments architects by delivering curated, organized research on sustainability standards, material comparisons, and efficiency benchmarks in seconds rather than hours, allowing architects to focus on design trade-offs and creative problem-solving while staying fully in control of project decisions.
Augmentation potentialclaude-sonnet-54/5AI tools significantly speed up gathering of environmental data, precedent studies, and energy performance benchmarks, letting architects focus on design integration and decision-making.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI systems can efficiently gather, organize, and synthesize public data on environmental standards, building codes, material specifications, and operational efficiency metrics from multiple sources (databases, PDFs, websites) with minimal human oversight. This would easily achieve >50% time savings over manual research, though final validation of project-specific requirements typically requires human judgment.
Task automatabilityclaude-sonnet-53/5AI can retrieve, summarize, and compile environmental codes, climate data, energy benchmarks, and site conditions quickly, but synthesizing this into project-specific sustainability strategy still needs human judgment and verification.
Adoption barriersclaude-haiku-4-5-202510012/5Few regulatory or legal barriers prevent automated research; architects typically remain responsible for interpreting and applying findings, but no licensing requirement mandates human-only data gathering. Organizational friction (preference for familiar manual processes, quality review procedures) is the main friction, not legal prohibition.
Adoption barriersclaude-sonnet-52/5No licensing requirement specifically governs this information-gathering sub-task, though downstream sustainability certifications and code compliance may require licensed sign-off, creating light institutional friction.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-driven information gathering (via API-based research tools or enterprise search systems) costs a fraction of a human architect's hourly rate; inference and integration together run to cents or low single-dollar amounts per project research phase, vastly cheaper than 2-3 days of architect time at loaded rates.
Cost vs. human wageclaude-sonnet-54/5Automated data gathering (climate data, code lookups, precedent research) is far cheaper than paying an architect's hourly rate for equivalent manual research, though final validation still requires expert time.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed AI tools (web search agents, document extraction, data aggregation platforms) already perform similar research tasks reliably in professional contexts. Products like specialized legal/compliance research platforms and construction databases with AI indexing are in production use, though integration into architectural workflows remains partially manual.
Technical feasibility todayclaude-sonnet-53/5Products like AI-assisted research tools, energy modeling software, and building performance databases exist and are used, but integration into full architectural workflows is uneven and often requires manual curation.

Create three-dimensional or interactive representations of designs, using computer-assisted design software.

56

CI 45–66 · exposure 58 · augmentation 88 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Architecture and design firms are actively adopting AI-assisted CAD tools and generative design workflows; major CAD vendors have integrated AI features and production pilots are common in large firms. Adoption is faster in information-rich design contexts and slower in small practices, but momentum is strong.
Sector adoption velocityclaude-sonnet-53/5Architecture firms are piloting AI visualization and generative design tools actively, but full production integration into BIM workflows remains uneven compared to faster-adopting sectors like finance or software.
Augmentation potentialclaude-haiku-4-5-202510015/5AI design assistants substantially boost architect productivity by automating geometry generation, constraint exploration, and variant creation, allowing architects to focus on intent and refinement. This is a canonical use case for human-AI teaming where AI augmentation is already transforming output.
Augmentation potentialclaude-sonnet-54/5AI significantly speeds up early-stage visualization, iteration, and client presentation renderings, meaningfully boosting architect productivity while they retain design control.
Task automatabilityclaude-haiku-4-5-202510014/5AI can automate substantial portions of 3D model generation and parametric design tasks using existing CAD APIs and generative 3D tools (e.g., constraint-based generation, procedural modeling), achieving significant time savings. However, the creative direction, design intent capture, and refinement typically require human judgment, preventing full end-to-end automation at production quality.
Task automatabilityclaude-sonnet-53/5AI tools can generate 3D massing models and interactive renderings from prompts or sketches, but producing construction-accurate, code-compliant BIM models still requires significant human modeling and adjustment.
Adoption barriersclaude-haiku-4-5-202510014/5Architects bear professional liability for designs and must sign off on deliverables; many jurisdictions require a licensed architect to review and certify final designs. These licensing and liability requirements create meaningful legal barriers to full automation, even if the technical capability exists.
Adoption barriersclaude-sonnet-52/5No licensing requirement mandates a human perform 3D modeling itself, though final design responsibility and stamped drawings remain with licensed architects, creating light oversight friction.
Cost vs. human wageclaude-haiku-4-5-202510013/5AI inference and integration costs are modest, but the task still requires architect oversight, iteration, and quality assurance. The all-in cost approaches parity with a junior architect's time for many routine modeling tasks, but is not dramatically cheaper.
Cost vs. human wageclaude-sonnet-53/5AI-assisted rendering and massing tools can cut modeling time substantially, but licensing, integration, and the need for architect oversight keep the all-in cost only moderately below traditional CAD/BIM labor.
Technical feasibility todayclaude-haiku-4-5-202510013/5Products like generative design plugins (Autodesk, Fusion 360) and AI-assisted 3D modeling exist and are deployed, but they remain narrow in scope and require substantial human oversight and correction. Reliability is good for routine geometry but falters on complex design constraints and aesthetic refinement.
Technical feasibility todayclaude-sonnet-53/5Products like Midjourney, Veras, TestFit, and generative plugins for Revit/SketchUp exist and are used in practice, but they mostly assist ideation/visualization rather than reliably producing final deliverable 3D models end-to-end.

Prepare operating and maintenance manuals, studies, or reports.

49

CI 30–67 · exposure 50 · augmentation 88 · importance 2.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architectural and engineering firms have been slow to adopt AI for documentation generation at scale; most adoption is limited to pilots or supplementary drafting. The profession remains human-centric for high-stakes deliverables, and digital-first firms show only moderate uptake of automated manual generation.
Sector adoption velocityclaude-sonnet-53/5Architecture firms are moderately adopting AI for documentation and reporting tasks, but the profession overall lags software/finance in production deployment of AI drafting tools.
Augmentation potentialclaude-haiku-4-5-202510014/5AI excels at assisting architects by auto-drafting template sections, formatting, organizing information, and suggesting standard maintenance procedures, significantly accelerating the manual preparation process while the architect retains oversight and customization control.
Augmentation potentialclaude-sonnet-55/5AI substantially speeds up drafting of manuals, reports, and studies by generating first drafts from technical inputs, letting architects focus on review, customization, and accuracy checks.
Task automatabilityclaude-haiku-4-5-202510012/5AI can draft portions of manuals (e.g., standard maintenance procedures, formatting, basic documentation) but cannot independently produce complete, context-specific operating manuals that require deep domain knowledge of specific building systems, compliance verification, and architectural judgment. The task requires integration of custom design details and regulatory compliance that AI struggles to fully automate without substantial human oversight.
Task automatabilityclaude-sonnet-54/5Drafting operating and maintenance manuals is largely template-driven synthesis of technical specifications, which LLMs can produce with substantial time savings when given source materials like building specs and equipment lists.
Adoption barriersclaude-haiku-4-5-202510014/5Architects often face liability and professional responsibility requirements to certify that operations and maintenance documentation is accurate and complete for building systems they designed. Professional licensing, building code compliance, and client accountability create regulatory and contractual friction against full automation without architect sign-off.
Adoption barriersclaude-sonnet-52/5No licensing requirement mandates a human architect personally author O&M manuals, though professional liability and accuracy expectations mean a licensed professional typically reviews and stamps final documents.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI writing and document generation tools are inexpensive per document, but the overhead of human review, correction, and validation of technical accuracy often approaches or exceeds the cost of human-authored documentation, especially for complex custom projects requiring specialized knowledge.
Cost vs. human wageclaude-sonnet-54/5Generating draft manuals via AI is far cheaper per page than billable architect hours, though some cost remains for review, verification against project specifics, and formatting to firm/client standards.
Technical feasibility todayclaude-haiku-4-5-202510013/5AI writing tools and documentation generators exist and are used in professional practice, but they typically require significant human editing, verification, and integration of project-specific details. Current products handle templated sections well but produce material errors when handling complex technical specifications or novel architectural systems.
Technical feasibility todayclaude-sonnet-53/5AI writing tools and document generation products exist and are used for technical documentation, but architecture-specific O&M manuals require domain accuracy and integration with project-specific data that current products handle inconsistently without human review.

Prepare scale drawings or architectural designs, using computer-aided design or other tools.

45

CI 28–62 · exposure 45 · augmentation 88 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Architecture and engineering firms are in pilot and early-adoption phases with generative design tools; uptake is visible but not yet deep production displacement comparable to technical/finance sectors.
Sector adoption velocityclaude-sonnet-53/5Architecture firms are adopting CAD/BIM-integrated AI features (e.g., generative layout tools, rendering assistants) at a moderate pace, with pilots more common than full production reliance.
Augmentation potentialclaude-haiku-4-5-202510015/5AI-driven design generation and rapid iteration are already transforming architect productivity by enabling fast exploration of alternatives, site-responsive variations, and code-compliant baseline designs that architects then refine—a strong human-in-the-loop productivity multiplier.
Augmentation potentialclaude-sonnet-54/5AI significantly speeds up early-stage design exploration, drafting repetitive elements, and rendering, meaningfully boosting architect productivity while they retain design authority.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI can generate detailed architectural designs, floor plans, and scale drawings from specifications or sketches with substantial time savings (50%+ feasible), though human architects typically need to review, refine, and validate outputs for compliance and aesthetic judgment.
Task automatabilityclaude-sonnet-52/5AI tools can generate concept sketches or assist with drafting, but producing compliant, buildable scale drawings requires professional judgment, code compliance, and client-specific iteration that current systems cannot fully replace end-to-end.
Adoption barriersclaude-haiku-4-5-202510013/5While no explicit legal licensing block prevents AI-generated CAD output, professional liability, code compliance responsibility, and architectural licensure requirements (final designs must be sealed by a licensed architect) create meaningful friction to full substitution.
Adoption barriersclaude-sonnet-54/5Architectural drawings for construction typically require a licensed architect's stamp/sign-off, creating strong legal and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI inference and integration costs for generating design alternatives are now a small fraction of senior architect hourly rates ($75–150+ loaded), especially when amortized across multiple design iterations.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce some drafting time but licensed architects still must review, refine, and stamp drawings, so overall cost savings versus human labor are modest rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510013/5Multiple CAD-integrated AI tools and generative design systems exist in production (e.g., Autodesk Revit plugins, Midjourney-to-CAD workflows), but they have material limitations in handling complex site constraints, code compliance, and seamless integration into professional pipelines.
Technical feasibility todayclaude-sonnet-52/5AI-assisted design generation (e.g., generative design plugins, text-to-render tools) exists but is used narrowly for early concepting, not for producing final construction-grade drawings reliably in production workflows.

Prepare information regarding design, structure specifications, materials, color, equipment, estimated costs, or construction time.

37

CI 25–50 · exposure 38 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architectural firms are experimenting with AI-assisted design tools and generative systems, but adoption remains pilot-stage; most firms continue traditional preparation workflows, and professional liability concerns slow production deployment.
Sector adoption velocityclaude-sonnet-53/5Architecture firms are adopting AI-assisted design and specification tools at a moderate pace, with pilots and partial integration common but full-scale replacement of this task still limited.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist architects by auto-generating preliminary cost estimates, material databases, or code-compliance checklists, reducing manual data work and allowing architects to focus on design judgment and integration.
Augmentation potentialclaude-sonnet-54/5AI tools significantly speed up specification writing, material selection suggestions, and cost estimation drafts, meaningfully boosting architect productivity while keeping them in the decision loop.
Task automatabilityclaude-haiku-4-5-202510012/5AI can generate partial outputs (material lists, cost estimates, basic specifications) but cannot reliably produce integrated design information that meets professional standards or handles context-dependent decisions about structure, aesthetics, and feasibility without substantial human review and revision.
Task automatabilityclaude-sonnet-53/5AI can draft specifications, cost estimates, and material descriptions from structured inputs, but architects must verify accuracy, integrate design intent, and finalize technical judgment calls, limiting full end-to-end automation.
Adoption barriersclaude-haiku-4-5-202510014/5Architectural practice is heavily regulated; in most jurisdictions, design documents and specifications must be prepared or signed by a licensed architect, creating a legal requirement for human professional involvement that blocks full automation.
Adoption barriersclaude-sonnet-53/5Architectural documents often require licensed architect stamps/sign-off for liability and code compliance, creating moderate regulatory and liability barriers even though drafting itself isn't restricted.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted preparation still requires substantial architect oversight to validate specifications, ensure design coherence, and correct errors; the all-in cost of AI + architect review likely approaches or exceeds having an architect prepare the information directly.
Cost vs. human wageclaude-sonnet-53/5AI-assisted drafting and estimation tools reduce time substantially, but licensing, integration with CAD/BIM systems, and mandatory human oversight keep total costs only moderately below traditional labor costs.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI tools exist for generating specifications and cost estimates (e.g., through parametric design or data lookup), no deployed product reliably produces comprehensive, accurate design preparation documents that satisfy architectural practice standards without significant human curation and domain expertise.
Technical feasibility todayclaude-sonnet-53/5BIM-integrated AI tools and generative design software exist and are used in practice for spec drafting and cost estimation, but they require significant human review and are not fully autonomous in production workflows.

Design structures that incorporate environmentally friendly building practices or concepts, such as Leadership in Energy and Environmental Design (LEED) standards.

29

CI 25–32 · exposure 30 · augmentation 75 · importance 3.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Architecture firms are experimenting with generative design and sustainability analysis tools, but adoption remains pilot-stage rather than production-scale displacement. Large and digitally mature practices lead; small and traditional firms lag significantly.
Sector adoption velocityclaude-sonnet-52/5Architecture firms are adopting AI tools slowly compared to software or finance, given the physical, project-based, highly regulated nature of the industry.
Augmentation potentialclaude-haiku-4-5-202510014/5AI excels at rapidly generating design alternatives, checking LEED compliance matrices, and optimizing energy performance simulations—all tasks that meaningfully accelerate an architect's workflow while the architect retains creative and regulatory judgment.
Augmentation potentialclaude-sonnet-54/5AI tools significantly aid energy simulation, material selection, and generative massing studies for sustainable design, meaningfully boosting architect productivity while the architect retains final design authority.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can retrieve LEED standards and generate design suggestions, architecture requires spatial judgment, structural integrity validation, cost optimization, and client-specific trade-offs that demand human expertise. AI tools lack the end-to-end autonomous capability to design complex structures meeting regulatory and sustainability criteria without substantial human oversight.
Task automatabilityclaude-sonnet-52/5Sustainable design requires integrated judgment across code compliance, site conditions, client goals, and LEED credit tradeoffs that current AI cannot autonomously reconcile end-to-end; AI can assist pieces but not replace the holistic design process.
Adoption barriersclaude-haiku-4-5-202510014/5Architects must be licensed professionals; structural and environmental compliance signatures are legally required in most jurisdictions. Building codes, liability for design failures, and professional responsibility create hard barriers to full automation, though AI can assist the licensed architect.
Adoption barriersclaude-sonnet-54/5Licensed architect sign-off is legally required for building designs, and LEED certification involves formal third-party review, creating strong regulatory and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI systems require significant integration with existing CAD workflows, LEED databases, and human review cycles, making the total cost comparable to or higher than portions of traditional architectural work. The need for expert human sign-off limits cost savings.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce time on analysis and documentation but licensed architects and engineers still must review, stamp, and integrate designs, so overall cost savings are moderate rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510013/5AI products (e.g., generative design tools, parametric software, LEED checklist assistants) exist and are deployed, but they typically handle component optimization or compliance documentation rather than full architectural design. Real-world deployment remains narrow and requires human architects to validate and integrate outputs.
Technical feasibility todayclaude-sonnet-52/5Some AI-assisted energy modeling and sustainability analysis tools are deployed (e.g., generative design plugins, LEED credit checklists), but no product autonomously produces a certifiable sustainable building design without heavy architect involvement.

Consult with clients to determine functional or spatial requirements of structures.

28

CI 25–30 · exposure 25 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architecture remains a relationship-driven, low-automation sector. Adoption of AI consultation aids is slow; most firms still rely on in-person or synchronous meetings, and client-facing workflows change slowly.
Sector adoption velocityclaude-sonnet-52/5Architecture as a sector has been slower to adopt AI agents for client-facing tasks compared to information/finance sectors; adoption is mostly in design visualization, not consultation.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by drafting requirement summaries, suggesting spatial configurations, or flagging missing constraints after consultation, moderately raising the architect's efficiency in synthesizing and organizing client input.
Augmentation potentialclaude-sonnet-54/5AI tools can help architects prepare questionnaires, summarize client input, generate space-planning drafts, and visualize options quickly, meaningfully boosting productivity during the consultation process.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in gathering and organizing spatial requirements, the core task requires real-time negotiation, interpretation of nuanced client needs, and contextual judgment that demands human presence. Current AI systems cannot reliably conduct the full client consultation process end-to-end with equal quality outcomes.
Task automatabilityclaude-sonnet-52/5This task requires live client interaction, reading unstated needs, and building trust/rapport, which current AI cannot fully replicate end-to-end; some prep and note synthesis can be automated but the core consultation cannot meet the 50% time-saving bar.
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: clients expect direct engagement with a licensed architect; professional liability and duty-of-care requirements mean the architect must personally sign off on interpreted requirements; many jurisdictions require documented architect–client consultation as part of project approval.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement mandates a human for the consultation itself, but clients strongly prefer human interaction for trust, liability, and design judgment reasons, creating moderate friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted requirement gathering tools exist but cannot eliminate the architect's billable consultation hours. The loaded cost of human architect time ($150–300+/hour) far exceeds the marginal cost of AI assistance, leaving the human cost dominant.
Cost vs. human wageclaude-sonnet-52/5While AI-assisted intake tools are cheap, the actual consultation still needs a human architect's time and judgment, so overall cost savings versus the human-led process are modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed products reliably perform full client consultation independently. Some AI tools can help draft requirement summaries or suggest spatial solutions post-consultation, but the interactive discovery process itself—understanding unstated needs, building trust, clarifying priorities—remains a task where deployed systems have narrow, unreliable scope.
Technical feasibility todayclaude-sonnet-52/5AI chatbots and intake forms exist to gather requirements, but no deployed product reliably conducts full client consultations for architectural spatial programming in production at scale.

Plan layouts of structural architectural projects.

26

CI 25–28 · exposure 25 · augmentation 75 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Architecture firms are actively piloting generative design and AI-assisted tools, but adoption remains primarily in exploratory phases. Most production workflows still rely on traditional CAD and human-led design; AI is accelerating specific tasks like iteration and analysis rather than replacing full planning responsibility.
Sector adoption velocityclaude-sonnet-52/5Architecture/construction is a design-heavy but historically slower-adopting sector for AI in core technical work; AI use is more common in visualization and early concept stages than in structural layout finalization.
Augmentation potentialclaude-haiku-4-5-202510014/5Generative design and AI-assisted layout tools significantly augment architect productivity by rapidly exploring spatial options, flag code conflicts, and automating routine adjustments. Architects remain in the loop making final decisions, but AI materially accelerates the planning phase and improves design exploration.
Augmentation potentialclaude-sonnet-54/5AI-assisted design tools meaningfully speed up early-stage layout exploration, generate variations, and help visualize options, significantly boosting architect productivity while the architect retains final control.
Task automatabilityclaude-haiku-4-5-202510012/5AI can generate initial layout suggestions and coordinate spatial relationships, but architectural planning requires human judgment on aesthetics, building codes, client needs, and site-specific constraints. Current systems cannot reliably produce end-to-end designs meeting 50% time savings at equal quality without extensive human revision.
Task automatabilityclaude-sonnet-52/5AI can generate draft layout options and floor plans from prompts or constraints, but integrating structural, code, site, and client requirements into a coherent full layout still requires substantial human judgment and iteration. Current tools save time on ideation but do not replace the full planning task end-to-end at equal quality.
Adoption barriersclaude-haiku-4-5-202510014/5Architectural projects require licensed professionals to approve designs for code compliance, safety, and legal liability. Building codes, zoning laws, and professional licensure create hard barriers—a licensed architect must sign off on plans regardless of how much AI assists the process.
Adoption barriersclaude-sonnet-54/5Architectural plans typically require a licensed architect's stamp/sign-off for code compliance and liability reasons, creating a strong regulatory and legal barrier to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted design tools reduce some drafting labor, but architects must oversee, validate, and often substantially rework outputs. The combined cost of software, integration, and required architect oversight remains comparable to or exceeds direct human planning labor.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce some drafting/iteration time but licensed architects must still validate and finalize layouts, so overall cost savings are moderate rather than order-of-magnitude, given required professional oversight.
Technical feasibility todayclaude-haiku-4-5-202510012/5While CAD tools and generative design products exist, they function primarily as assistants rather than autonomous planners. No deployed product reliably produces production-ready layouts without significant human expertise guiding and validating every major decision.
Technical feasibility todayclaude-sonnet-52/5Generative design and AI-assisted CAD tools (e.g., Autodesk's AI features, Midjourney for massing studies) exist but are used as ideation aids rather than reliable production-grade layout planners; architects still perform the core structural layout work manually.

Integrate engineering elements into unified architectural designs.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While architectural firms are exploring generative design and BIM tools, adoption of AI for integrated design synthesis remains in pilot phases; most firms continue to rely on licensed architects for critical integration decisions rather than deploying autonomous systems.
Sector adoption velocityclaude-sonnet-52/5AEC (architecture, engineering, construction) is a traditionally slow-adopting, project-based sector with limited AI production deployment beyond visualization and clash-detection pilots.
Augmentation potentialclaude-haiku-4-5-202510013/5AI tools can meaningfully assist architects by generating design alternatives, checking code compliance, and coordinating engineering systems, but the architect typically retains decision-making authority and refines outputs rather than being fully transformed in productivity.
Augmentation potentialclaude-sonnet-54/5AI-assisted BIM tools, generative design software, and clash-detection systems meaningfully speed up coordination and drafting work while architects retain final design authority.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in analyzing engineering constraints and generating design options, integrating engineering elements into a unified architectural design requires complex judgment calls, trade-offs between aesthetics and function, and synthesis across multiple domains that current systems cannot reliably perform end-to-end with 50% time savings at equal quality.
Task automatabilityclaude-sonnet-52/5Integrating structural, mechanical, and electrical engineering elements into a coherent design requires holistic judgment, tradeoff resolution, and creative synthesis that current AI cannot perform end-to-end without heavy human direction.dz Only sub-components like clash detection or layout suggestions are automatable.
Adoption barriersclaude-haiku-4-5-202510014/5Architectural licensure requirements mandate that a licensed architect take professional responsibility for designs; liability and building code compliance create legal barriers that prevent full substitution of human judgment, even where AI could theoretically assist.
Adoption barriersclaude-sonnet-54/5Architectural drawings and engineering integration typically require licensed professional sign-off (stamped drawings) for liability and code-compliance reasons, creating a significant regulatory barrier to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI design tools (generative design, coordination software) require significant human oversight, additional computing infrastructure, and data preparation, making total cost comparable to or higher than paying an architect to perform the integration directly.
Cost vs. human wageclaude-sonnet-52/5Current tools require significant licensed software, setup, and expert oversight, so while some efficiency gains exist, the all-in cost is not dramatically cheaper than skilled architect/engineer time for this complex task.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs full architectural integration of engineering elements in production; tools exist for narrow tasks (structural analysis, MEP coordination) but lack the holistic synthesis and design intent judgment that this task requires.
Technical feasibility todayclaude-sonnet-52/5Some BIM-integrated tools (e.g., clash detection, generative layout software) exist and are used in practice, but they assist rather than reliably perform the full integration task autonomously.

Prepare contract documents for building contractors.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption remains low in architectural practices; most firms rely on templates and legal counsel rather than AI systems. The high-touch nature and liability concerns of contract work slow adoption compared to information-intensive sectors.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction remain relatively slow adopters of AI compared to finance or software, with pilots for AI-assisted specification writing still uncommon in mainstream practice.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by generating initial drafts, suggesting standard clauses, and flagging inconsistencies, meaningfully reducing the time architects spend on document preparation while they retain responsibility for final review and approval.
Augmentation potentialclaude-sonnet-54/5AI writing and BIM-integrated tools meaningfully speed up drafting of specifications and standard contract language, letting architects focus review time on project-specific and liability-sensitive details.
Task automatabilityclaude-haiku-4-5-202510012/5Current AI can draft standard contract sections and templates with reasonable accuracy, but architectural contracts require nuanced legal language, site-specific conditions, and compliance with local building codes. A human architect or legal expert must review, customize, and sign off on final documents, preventing the ≥50% time-saving threshold.
Task automatabilityclaude-sonnet-52/5AI can draft portions of contract documents (boilerplate specifications, standard clauses) but assembling coordinated, project-specific contract documents with drawings, specs, and legal terms still requires substantial licensed professional judgment and coordination that current tools cannot fully replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Architectural contracts often involve licensed architects who bear professional and legal liability for contract terms. Regulatory frameworks, professional licensing requirements, and insurance liability create strong legal and organizational barriers to full automation.
Adoption barriersclaude-sonnet-54/5Contract documents typically require a licensed architect's stamp/seal and carry legal and liability implications, creating strong professional and regulatory barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted drafting may reduce initial writing time, but the required human review by architects or legal professionals consumes most of the labor cost. The all-in cost of AI generation plus mandatory expert review remains comparable to or higher than direct human drafting.
Cost vs. human wageclaude-sonnet-52/5AI tools can reduce drafting time for specification sections, but the human oversight, liability review, and coordination required keep the effective all-in cost close to or only modestly below traditional professional labor.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can generate contract boilerplate and assist with document structure, no deployed product reliably handles the legal complexity, liability implications, and jurisdiction-specific requirements of architectural contracts at scale. Existing legal AI tools are nascent and typically require substantial human oversight.
Technical feasibility todayclaude-sonnet-52/5Some AI-assisted specification writing and document generation tools exist (e.g., BIM-integrated spec generators), but no deployed product reliably produces complete, legally sound contract documents for construction without heavy architect review.

Administer construction contracts.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While architecture and construction firms increasingly pilot contract management software, meaningful AI-driven automation of contract administration is still uncommon in production; most firms rely on human project managers and legal review.
Sector adoption velocityclaude-sonnet-52/5placeholder
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by organizing contract data, highlighting non-standard clauses, tracking deadlines, and summarizing key terms, allowing architects to focus on negotiation and enforcement decisions rather than document sifting.
Augmentation potentialclaude-sonnet-54/5placeholder
Task automatabilityclaude-haiku-4-5-202510012/5While AI can draft contract templates, organize documents, and flag risk language, administering contracts requires negotiating terms, interpreting ambiguous clauses, managing disputes, and making discretionary enforcement decisions that currently demand human legal judgment and client relationship management.
Task automatabilityclaude-sonnet-52/5Contract administration requires site judgment, negotiation, change-order evaluation, and legal accountability that current AI cannot fully replicate end-to-end.},
Adoption barriersclaude-haiku-4-5-202510014/5Contract administration often requires the responsible architect or a licensed professional to legally sign off on changes, certifications, and compliance attestations; regulatory frameworks and professional liability requirements create strong barriers to full automation.
Adoption barriersclaude-sonnet-54/5placeholder
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI contract administration tools remain expensive relative to junior staff for the parts they automate, and architects still require oversight and human sign-off, so the total cost advantage is marginal and site-specific.
Cost vs. human wageclaude-sonnet-52/5placeholder
Technical feasibility todayclaude-haiku-4-5-202510012/5Products exist for contract review and clause extraction (e.g., AI-powered contract analysis tools), but deployed systems show material error rates in legal interpretation and lack the end-to-end capability to independently administer a contract through its lifecycle, including amendment negotiation and dispute resolution.
Technical feasibility todayclaude-sonnet-52/5placeholder

Design or plan construction of green building projects to minimize adverse environmental impact or conserve energy.

25

CI 25–25 · exposure 25 · augmentation 75 · importance 3.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI in architecture is moderate and primarily assistive rather than replacing core design tasks. Firms use generative design tools and energy modeling software, but architectural planning remains human-driven, reflecting the profession's entrenched design processes and the low incentive to fully displace licensed architects.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction sectors have historically been slower to adopt AI compared to information/finance industries, though green building analytics tools are gradually gaining use.
Augmentation potentialclaude-haiku-4-5-202510014/5AI augments this task significantly: energy simulation tools, parametric design, and generative alternatives can accelerate exploration of sustainable solutions and reduce design iteration time. Architects leverage these systems to improve productivity and explore more optimization scenarios than they could manually.
Augmentation potentialclaude-sonnet-54/5AI-driven energy modeling, material analysis, and generative design tools meaningfully speed up early-stage sustainable design exploration and analysis while architects retain final design authority.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with energy modeling, sustainability calculations, and generating design alternatives, but the task requires integrating complex trade-offs, client requirements, regulatory constraints, and aesthetic judgment. Current AI systems cannot autonomously synthesize these factors to produce a complete, defensible green building design meeting the ≥50% time-saving bar.
Task automatabilityclaude-sonnet-52/5Generative AI can assist with energy modeling, sustainable material suggestions, and drafting concepts, but the core design synthesis, site-specific judgment, and integration of client/regulatory constraints still require substantial human expertise beyond simple automation.
Adoption barriersclaude-haiku-4-5-202510014/5Licensing and liability are substantial barriers: only licensed architects can legally design buildings and sign construction documents in most jurisdictions. Building code compliance, environmental certifications (LEED, Passive House), and liability for structural/performance failures create hard legal requirements for human professional sign-off.
Adoption barriersclaude-sonnet-54/5Architectural design and construction documents typically require a licensed architect's stamp and legal responsibility, creating strong regulatory and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI energy simulation and design tools require significant training, integration, and human oversight by licensed architects. The all-in cost of these tools plus human review is comparable to or may exceed traditional design methods for complex green projects.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce time on analysis and iteration but still require licensed architects for design decisions, code compliance, and stamped drawings, so overall cost savings are moderate rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI tools for energy simulation and parametric design exist (e.g., Autodesk, Ladybug), they support rather than perform the full planning task independently. No deployed system reliably designs green building projects end-to-end; architects use AI as a tool within their workflow, not as an autonomous performer.
Technical feasibility todayclaude-sonnet-52/5Some products (energy simulation tools, generative design plugins like Autodesk Forma) assist parts of green building design, but no deployed system reliably performs full sustainable design planning end-to-end in production.

Design environmentally sound structural upgrades to existing buildings, such as natural lighting systems, green roofs, or rainwater collection systems.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 2.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI-assisted design in architectural practice is slow and fragmented, confined mainly to large firms experimenting with generative tools for early-stage ideation. The sector remains conservative due to licensing, liability, and the craft-oriented nature of practice; deployment of fully autonomous or minimal-oversight systems is minimal.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction are traditionally slow adopters of AI in production compared to information/finance sectors, with most AI use still in pilot or supplementary tool form.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-powered building simulation, generative design exploration, and environmental analysis (daylighting, thermal modeling, rainwater flow) can meaningfully assist architects in evaluating options and iterating faster. However, the human architect must remain central to site assessment, code compliance review, cost-benefit tradeoffs, and final design decisions.
Augmentation potentialclaude-sonnet-54/5AI tools significantly help architects with energy modeling, rendering options, precedent research, and generating design variations for sustainable retrofits, substantially boosting productivity while the architect retains design authority.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with generating design concepts and analyzing building data, this task fundamentally requires creative problem-solving, site-specific environmental assessment, regulatory compliance evaluation, and integration with existing building constraints that demand human professional judgment. Current AI falls far short of the ≥50% time-saving threshold for end-to-end reliable execution.
Task automatabilityclaude-sonnet-52/5This requires site-specific assessment, structural analysis, creative synthesis of building science with aesthetics, and coordination with engineers, which current AI cannot fully perform end-to-end despite being able to generate concept ideas or draft calculations.
Adoption barriersclaude-haiku-4-5-202510014/5Substantial legal and professional barriers exist: architects must be licensed and personally responsible for designs affecting building safety, structural integrity, and regulatory compliance. Liability for code violations and performance failures creates high error-cost asymmetry, and most jurisdictions require a licensed architect to stamp and sign structural design documents.
Adoption barriersclaude-sonnet-54/5Licensed architects must stamp and take legal responsibility for structural and building-code-related designs, creating a hard regulatory barrier to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI design tools and simulations still require specialized human expertise to interpret outputs, verify compliance, and adapt designs to site constraints. The integrated cost of AI systems, integration, and required professional oversight remains comparable to or exceeds the cost of direct architect time on this complex task.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply generate design options and run simulations, but the overall task still requires expensive professional judgment, site visits, and liability-bearing sign-off, keeping all-in costs close to human-comparable.
Technical feasibility todayclaude-haiku-4-5-202510012/5Generative design tools and building simulation software exist in research and niche commercial settings, but no mainstream deployed products reliably automate structural upgrade design from concept to specification. Most systems are narrow (e.g., single daylighting analysis) or require extensive manual intervention to produce buildable, code-compliant designs.
Technical feasibility todayclaude-sonnet-52/5AI design tools (generative design, energy modeling software) exist and are used for early concepting, but no deployed product reliably produces complete, code-compliant environmental retrofit designs without heavy architect oversight.

Develop final construction plans that include aesthetic representations of the structure or details for its construction.

25

CI 20–30 · exposure 30 · augmentation 75 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architectural firms are cautious adopters; while generative tools for concept sketches and visualization are spreading, production adoption for final construction documents is limited due to liability, licensing requirements, and client/regulatory risk aversion in this regulated sector.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction sectors are historically slow to digitize and adopt AI at production scale compared to software or finance industries.
Augmentation potentialclaude-haiku-4-5-202510014/5AI excels at rapidly generating multiple aesthetic representations, draft layouts, and technical visualizations that architects then refine; this significantly amplifies designer productivity in the exploratory and iteration phases while keeping human judgment and sign-off central.
Augmentation potentialclaude-sonnet-54/5AI-assisted design tools, generative design, and automated drafting significantly speed up iteration and detail generation while architects retain control over final decisions.
Task automatabilityclaude-haiku-4-5-202510012/5Current AI can generate 2D/3D visualizations and technical drawings, but creating final construction plans requires integrating aesthetic judgment with precise technical specifications, building code compliance, and site-specific constraints that demand substantial human oversight and revision.
Task automatabilityclaude-sonnet-52/5AI can help generate design options and drafts, but producing final, code-compliant construction documents with precise technical details requires professional judgment, coordination, and liability that current tools cannot fully replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Architects must be licensed professionals; construction plans must bear an architect's seal and signature, creating a hard legal barrier—only a qualified, licensed human can sign off on final plans regardless of AI assistance in generation.
Adoption barriersclaude-sonnet-55/5Final construction plans typically require a licensed architect's stamp/signature for legal and code compliance, creating a hard regulatory barrier to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI visualization and drafting tools reduce some design labor, but the oversight, refinement, and compliance verification by licensed architects remain substantial; AI cost is lower for raw generation but total cost per final deliverable is still comparable because human review is mandatory.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce drafting time but licensed architects must still verify, stamp, and finalize plans, so the all-in cost including professional oversight remains close to traditional costs.
Technical feasibility todayclaude-haiku-4-5-202510013/5Generative AI and CAD-integrated tools can produce draft plans and renderings at scale, but professional-grade construction documents remain largely human-drafted; AI tools are used for acceleration and visualization, not reliable end-to-end plan generation that meets regulatory and quality standards.
Technical feasibility todayclaude-sonnet-52/5BIM-integrated AI tools and generative design plugins exist (e.g., in Revit, Autodesk) but they assist rather than reliably produce complete final construction sets without extensive architect review.

Direct activities of technicians engaged in preparing drawings or specification documents.

25

CI 20–30 · exposure 20 · augmentation 50 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While architecture firms increasingly use AI tools for drafting assist, actual displacement of supervisory roles directing technicians is minimal. Adoption remains at the tool-assist level; deep automation of management activities in architectural practices remains rare.
Sector adoption velocityclaude-sonnet-52/5Architecture firms are adopting AI drafting tools slowly and unevenly; managerial/supervisory functions see minimal AI adoption compared to drafting tasks themselves.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist architects by auto-generating specification sections, flagging consistency errors in drawings, and organizing technician work queues, raising productivity on the technical side. However, the interpersonal and decision-making aspects of directing teams see moderate benefit.
Augmentation potentialclaude-sonnet-53/5AI can help review drafted drawings for errors, suggest revisions, and support communication of specifications, aiding the architect's oversight role without replacing the directive function.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in generating or reviewing drawing elements and specifications, the task of *directing* technician activities requires real-time human judgment, priority-setting, and accountability that current systems cannot reliably handle end-to-end. Supervision and quality control remain fundamentally human responsibilities.
Task automatabilityclaude-sonnet-52/5Directing and supervising other people's work involves judgment, communication, and quality control that current AI cannot fully replicate end-to-end, though AI can assist with reviewing outputs.'
Adoption barriersclaude-haiku-4-5-202510014/5Professional liability, design accountability, and client-facing responsibility are typically held by licensed architects. Regulatory and contractual requirements mean the architect must legally sign off on drawings and direct their production—an AI cannot assume that authority or liability.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement for directing technicians specifically, but organizational structures, liability for design accuracy, and managerial trust create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted drafting or specification generation may reduce costs for component parts, but a human architect directing technicians provides irreplaceable judgment and legal/contractual accountability. The full cost of replacing that oversight role exceeds the savings from document automation.
Cost vs. human wageclaude-sonnet-52/5Managing human staff requires human oversight and interpersonal coordination that AI cannot substitute cheaply; any AI assistance layers cost on top of the architect's existing role rather than replacing it.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably directs team activities or makes supervisory decisions in production. AI can draft documents or flag errors, but orchestrating technician work, resolving conflicts, and ensuring accountability remain beyond current system capabilities in real architectural offices.
Technical feasibility todayclaude-sonnet-51/5No deployed product manages or directs technician teams; AI tools assist with drafting review but not personnel/task direction in production settings.

Perform predesign services, such as feasibility or environmental impact studies.

24

CI 23–25 · exposure 25 · augmentation 63 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Architecture and engineering firms remain conservative in automating predesign services due to professional licensure requirements, client expectations for human expertise, and the high cost of errors in feasibility determinations. Adoption of AI for these tasks remains experimental and confined to specialized analytical subtasks.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction sectors are historically slow AI adopters, with predesign analytics still largely manual and pilot-stage AI tools not yet widespread in production workflows.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist architects by automating literature searches, preliminary code compliance checks, data visualization, and draft report generation, which can speed up early-stage analysis and allow professionals to focus on judgment and stakeholder engagement.
Augmentation potentialclaude-sonnet-54/5AI tools significantly speed up data collection, site analysis, and drafting of preliminary reports, giving architects substantial productivity gains while they retain final judgment and sign-off.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with data gathering, literature review, and preliminary analysis for feasibility and environmental studies, but cannot independently conduct site assessments, stakeholder engagement, or synthesize complex interdisciplinary judgments required for credible predesign services. The task requires substantial human expertise and on-site evaluation that current AI systems cannot fully replace.
Task automatabilityclaude-sonnet-52/5AI can assist with data gathering, summarizing regulations, and drafting parts of feasibility reports, but synthesizing site-specific judgments, stakeholder input, and regulatory nuance still requires substantial human expertise and site visits.imestamp
Adoption barriersclaude-haiku-4-5-202510014/5Strong legal and regulatory barriers exist: architects must sign and stamp professional documents, clients expect licensed professional judgment, and liability for incorrect feasibility assessments falls on the architect. Regulatory standards (e.g., NEPA for environmental impact) require documented professional responsibility that cannot be delegated to AI alone.
Adoption barriersclaude-sonnet-54/5Environmental impact studies and feasibility reports often require licensed professional certification, regulatory compliance, and legal accountability, creating strong barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI infrastructure and oversight costs for generating credible feasibility studies are substantial relative to typical consulting labor, especially when accounting for liability and the need for human expert review and revision of AI-generated analyses.
Cost vs. human wageclaude-sonnet-52/5AI can cut research and drafting time modestly, but licensed professional review, site visits, and liability sign-off keep overall costs close to human-driven processes.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI tools exist for document analysis and report drafting, no deployed product reliably performs end-to-end feasibility or environmental impact studies at the quality expected in professional practice. Products lack the domain expertise integration, site-specific analysis, and regulatory compliance assurance that clients and regulators require.
Technical feasibility todayclaude-sonnet-52/5There are AI tools for zoning lookups, site analysis, and environmental data aggregation, but no mature end-to-end product reliably performs full feasibility or environmental impact studies in production without heavy architect oversight.

Plan or design structures such as residences, office buildings, theatres, factories, or other structural properties in accordance with environmental, safety, or other regulations.

23

CI 20–25 · exposure 25 · augmentation 75 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architecture remains a design-intensive, relationship-driven profession where clients expect human judgment and custom solutions. Adoption of AI is mainly in supportive roles (visualization, analysis) within larger practices; full replacement or agent-led design workflows remain rare in production.
Sector adoption velocityclaude-sonnet-52/5Architecture firms are adopting AI tools for visualization and early massing studies, but the sector overall is slow to adopt AI for core design due to liability, licensing, and industry conservatism.
Augmentation potentialclaude-haiku-4-5-202510014/5Generative design, real-time compliance feedback, parametric modeling, and energy simulation can substantially accelerate an architect's iteration and analysis phases. AI can enhance productivity on conceptual layout and performance evaluation while the human retains design authority and regulatory responsibility.
Augmentation potentialclaude-sonnet-54/5AI substantially assists architects via rapid generative concept iterations, code-checking software, rendering, and parametric design tools, meaningfully boosting productivity while the licensed architect remains responsible for final designs.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with generating design layouts and checking code compliance, the task requires integrated judgment across aesthetics, site-specific environmental factors, client needs, and regulatory nuance that current systems cannot fully synthesize end-to-end at production quality. Meaningful parts (e.g., parametric layouts, code checking) can be partially automated, but client-facing design and final liability still rest with human architects.
Task automatabilityclaude-sonnet-52/5Architectural design requires synthesizing client needs, site constraints, building codes, and creative form-making into coherent structural proposals; AI can generate schematic massing or draft options but cannot yet independently produce code-compliant, buildable designs end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Architects must be licensed professionals who take legal liability for safety and code compliance; buildings require permit sign-off and often third-party inspection. Regulatory frameworks explicitly require a licensed architect's seal, creating a hard legal barrier to full automation regardless of AI capability.
Adoption barriersclaude-sonnet-55/5Architectural design for buildings legally requires a licensed architect to stamp and take liability for drawings in nearly all jurisdictions, making this a hard-barrier, regulator-protected task.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI tools require significant setup, integration with CAD/BIM ecosystems, and human architect review to meet professional standards, making per-project costs still substantial relative to junior architect labor. Full replacement of high-touch design work is not yet economically compelling.
Cost vs. human wageclaude-sonnet-52/5Licensed architects command high wages, so AI-assisted early-stage design generation is cheaper for concepting, but full design work still requires substantial human hours for compliance, coordination, and liability, keeping overall cost savings modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5Generative design and compliance-checking tools exist in labs and some firms, but no production system reliably designs complete structures meeting all regulations, environmental requirements, and client briefs without substantial human oversight and revision. Deployed products address narrow sub-tasks (facade generation, energy modeling) rather than the full design process.
Technical feasibility todayclaude-sonnet-52/5AI-assisted generative design tools (e.g., in Autodesk, Hypar) exist and produce concept options, but no production system reliably delivers full building designs with regulatory compliance without extensive human architect revision.

Represent clients in obtaining bids or awarding construction contracts.

23

CI 20–25 · exposure 20 · augmentation 50 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architectural firms are adopting AI for design and documentation tasks, but actual deployment of AI for client-facing contract representation and bid management remains minimal. This task involves high-stakes judgment and client relationship, slowing adoption despite digitization of the sector.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction sectors are known for slower digitization and AI adoption compared to information or finance industries, with procurement processes still largely manual and relationship-driven.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist architects by analyzing bid documents, flagging non-compliant terms, summarizing contractor qualifications, and organizing contract data. These tools raise productivity in bid evaluation and comparison while the architect retains final judgment and client representation authority.
Augmentation potentialclaude-sonnet-53/5AI tools can help organize bid documents, compare submissions, flag inconsistencies, and draft contract language, meaningfully speeding up parts of the process while the architect retains representation and decision authority.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in analyzing bids and contract terms, the core task requires legal judgment, negotiation, relationship management, and client representation that demands human authority. Partial automation of bid analysis and document review is feasible, but end-to-end client representation with equivalent quality and time savings falls short of the 50% threshold.
Task automatabilityclaude-sonnet-52/5This involves negotiation, judgment about contractor reliability, and client representation in a legally consequential process that AI cannot fully execute end-to-end today.assessments require relationship management and accountability beyond drafting or data comparison.
Adoption barriersclaude-haiku-4-5-202510014/5Architects must represent clients legally and professionally; liability for poor bid selection or contract terms rests on the licensed architect. Professional licensure, fiduciary duty, and the legal requirement for a qualified professional to sign off on contract awards create substantial barriers to full automation.
Adoption barriersclaude-sonnet-54/5Architects often have contractual and professional liability obligations to represent clients faithfully in procurement, and construction contracts carry significant legal and financial stakes requiring licensed professional judgment and signature authority.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for contract analysis and bid summaries carry meaningful setup and integration costs, but the human architect's loaded wage for this task remains lower than the all-in cost of AI systems that would need to replicate judgment and client interaction at scale.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply assist with bid comparison spreadsheets or document drafting, but the overall representation, negotiation, and liability-bearing decision still requires costly human oversight, keeping total cost comparable to human-only delivery.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs full client representation in bid solicitation and contract awarding. AI tools can help draft RFPs or summarize bids, but actual representation—managing client expectations, negotiating terms, and making binding decisions—requires human professionals and remains research-stage in production contexts.
Technical feasibility todayclaude-sonnet-51/5No deployed product independently represents architects' clients in bid solicitation or contract award decisions; this remains a human-led professional service function.

Inspect proposed building sites to determine suitability for construction.

23

CI 20–25 · exposure 20 · augmentation 50 · importance 2.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architecture remains a human-centered, relationship-driven profession with slow AI adoption in core decision-making tasks. Preliminary data gathering tools are spreading, but autonomous suitability assessment is not yet a documented practice in production workflows.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction remain relatively low-digitization, slow-adopting sectors for full task automation, though GIS and remote sensing tools are gaining some traction for preliminary analysis.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by preprocessing satellite imagery, extracting zoning data, flagging flood or environmental constraints, and organizing site survey information, reducing the architect's manual data-gathering work and enabling more informed site visits.
Augmentation potentialclaude-sonnet-53/5AI-powered satellite imagery, GIS mapping, soil/flood data analysis, and drone surveys can meaningfully assist architects in preparing for and supplementing physical site visits.
Task automatabilityclaude-haiku-4-5-202510012/5AI can analyze aerial imagery, geological data, and basic site surveys to flag obvious constraints (flooding risk, slope, soil type), but determining suitability requires nuanced judgment about zoning compliance, utilities access, environmental sensitivity, and client-specific constraints that typically demands human site visits and expert interpretation. Current systems cannot replace the end-to-end inspection and decision.
Task automatabilityclaude-sonnet-52/5Site inspection requires physical presence to assess terrain, drainage, access, context, and unforeseen conditions that current AI cannot perceive directly; AI can assist with data analysis but not replace the physical inspection itself.
Adoption barriersclaude-haiku-4-5-202510014/5Professional liability, design responsibility, and regulatory approval requirements mean the determination of suitability typically must be made or signed off by a licensed architect who can be held accountable. Clients and building authorities expect human professional judgment on feasibility decisions.
Adoption barriersclaude-sonnet-54/5Architects are licensed professionals whose site assessments carry legal and liability weight for permitting and safety; many jurisdictions require a licensed professional's sign-off on site suitability determinations.
Cost vs. human wageclaude-haiku-4-5-202510012/5Integrating satellite imagery, GIS platforms, and AI analysis tools still requires skilled human interpretation and site visits, keeping total cost comparable to or exceeding a single architect's inspection. Economies of scale are limited per site.
Cost vs. human wageclaude-sonnet-52/5AI tools (GIS analysis, satellite/drone imagery) can reduce some preliminary desk research costs, but the core physical inspection still requires a human architect on-site, so overall cost savings are limited.
Technical feasibility todayclaude-haiku-4-5-202510012/5Computer vision tools and GIS analysis exist and can extract some site features from images and maps, but production systems for autonomous suitability determination are rare and unreliable. Most architectural practices still rely on human inspection supplemented by data tools rather than AI-driven suitability assessment.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs autonomous physical site suitability inspection; drone/satellite imagery tools exist for data gathering but the professional judgment-based inspection remains research-stage or human-only.

Monitor the work of specialists, such as electrical engineers, mechanical engineers, interior designers, or sound specialists to ensure optimal form or function of designs or final structures.

16

CI 7–25 · exposure 13 · augmentation 50 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI-assisted design compliance tools is emerging in larger firms, but monitoring specialist work remains a core architect responsibility integrated into professional practice and liability; displacement is slow and limited to narrow compliance checking rather than holistic oversight.
Sector adoption velocityclaude-sonnet-52/5Architecture and construction sectors are slow AI adopters overall, with AI tools used mainly for drafting or visualization rather than cross-disciplinary oversight.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can augment an architect's oversight by flagging code violations, detecting clashes between disciplines, and summarizing specialist submittals, allowing the architect to focus on higher-level trade-offs and validation—but the human remains essential for judgment and sign-off.
Augmentation potentialclaude-sonnet-53/5AI tools can help flag clashes in BIM models, summarize specialist reports, or track design compliance, aiding but not replacing the architect's oversight role.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can flag technical discrepancies or inconsistencies in designs (e.g., code violations, spatial conflicts), the task requires judgment about 'optimal form or function' across multiple specialist domains, contextual trade-offs, and often subjective aesthetic or functional priorities that demand human expertise and accountability.
Task automatabilityclaude-sonnet-51/5This requires cross-disciplinary judgment, coordination, and real-time evaluation of specialist work against project intent, which AI cannot perform end-to-end today.
Adoption barriersclaude-haiku-4-5-202510014/5Legal and professional liability falls on the licensed architect; most jurisdictions require a licensed architect to certify design adequacy and coordinate specialist work, creating a hard barrier to full automation even if technically feasible.
Adoption barriersclaude-sonnet-54/5Licensed architects bear legal and professional responsibility for design coordination and sign-off, creating strong liability and licensing barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted compliance tools exist but require significant human expert review and integration; the all-in cost of an AI system plus required architect oversight is comparable to or exceeds the incremental cost of architect time spent directly monitoring.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this oversight role, so cost comparison favors the human architect who provides accountable judgment and coordination.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools can assist with technical compliance checking and visualization review, but no deployed product reliably monitors and coordinates the work of diverse specialists end-to-end; such oversight is task-specific, context-dependent, and typically requires deep domain knowledge across multiple engineering and design disciplines.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously monitors and coordinates multi-disciplinary specialist work on building projects; this remains a human project-management function.

Conduct periodic on-site observations of construction work to monitor compliance with plans.

15

CI 5–25 · exposure 13 · augmentation 50 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Architecture and construction remain relatively low in digital adoption. While some large firms pilot drone inspection, most projects still rely on traditional architect site visits. Widespread AI-driven compliance monitoring is rare and progresses slowly.
Sector adoption velocityclaude-sonnet-51/5Construction and on-site architectural oversight is a physical, low-digitization sector with minimal AI agent deployment for this specific inspection task.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted image analysis and automated deviation detection can help an architect prepare for or focus site visits more efficiently, and drone imagery can supplement in-person inspection. However, the core judgment task remains largely manual and human-led.
Augmentation potentialclaude-sonnet-53/5AI-powered photo documentation, drone imagery analysis, and computer vision tools can help flag discrepancies or track progress, assisting the architect's review process even though the physical observation itself remains human-led.
Task automatabilityclaude-haiku-4-5-202510012/5While AI-powered computer vision systems can detect some deviations from plans (e.g., structural positioning via drone imagery), the task requires real-time judgment about code compliance, material quality, and contextual decision-making that current systems cannot reliably perform end-to-end. Significant human oversight remains necessary.
Task automatabilityclaude-sonnet-51/5This requires physical presence at a construction site to visually inspect work against plans, which current AI systems cannot perform end-to-end; while photo/video analysis tools exist, they cannot replace the physical observation and professional judgment involved.
Adoption barriersclaude-haiku-4-5-202510014/5Building codes and professional liability frameworks often require a licensed architect or engineer to personally certify compliance; many jurisdictions legally mandate in-person observation and sign-off. Organizational and regulatory requirements strongly favor human presence.
Adoption barriersclaude-sonnet-54/5Architects often have contractual and sometimes legal obligations to certify compliance with plans and specifications, creating liability exposure that requires a licensed professional's sign-off, though not universally mandated by statute in all jurisdictions.
Cost vs. human wageclaude-haiku-4-5-202510012/5Drone inspection and AI analysis add infrastructure and subscription costs, but an architect's wage for site observation remains lower than the full stack of automation plus necessary human oversight per visit. Cost advantage is marginal at best.
Cost vs. human wageclaude-sonnet-51/5There is no AI system capable of performing this physical site visit task, so any cost comparison favors the human by default since AI cannot deliver the output at all.
Technical feasibility todayclaude-haiku-4-5-202510012/5Products exist for drone inspection and image analysis, but they lack the comprehensive judgment needed to assess full compliance with architectural plans and building codes. Deployed systems are narrow (e.g., spotting cracks) and require architect review; no mature product performs complete on-site compliance monitoring independently.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously conducts on-site construction observations; this remains a physical, in-person professional activity with no production-scale AI substitute.

Meet with clients to review or discuss architectural drawings.

9

CI 5–13 · exposure 5 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Architecture remains a relationship-intensive, high-stakes profession where client contact is a core value proposition and differentiator; adoption of AI agent replacement for client meetings is negligible and unlikely to accelerate without fundamental changes to professional licensing and liability frameworks.
Sector adoption velocityclaude-sonnet-52/5Architecture firms are adopting AI for drafting and visualization but client-facing meetings remain untouched, reflecting slower adoption in this specific interaction type despite firm-level digitization.
Augmentation potentialclaude-haiku-4-5-202510012/5AI can assist by pre-analyzing drawings for errors, generating alternative design options, or preparing background summaries before a meeting, but the meeting itself remains human-driven and the augmentation value is peripheral to the core task of client dialogue and decision-making.
Augmentation potentialclaude-sonnet-54/5AI can generate renderings, visualizations, and drawing revisions in real time or beforehand that architects use to enrich client discussions, meaningfully improving the meeting's productivity.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time interpersonal negotiation, client feedback interpretation, and dynamic responsiveness to concerns—capabilities where current AI systems cannot substitute for a human architect's judgment and presence. The task is fundamentally about human communication and relationship management rather than information processing.
Task automatabilityclaude-sonnet-51/5This is an interpersonal client meeting requiring real-time relationship building, negotiation, and reading nuanced feedback that current AI cannot substitute for end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: clients expect to meet with a licensed architect who is legally accountable for design decisions and sign-offs; liability and professional responsibility rest on the human architect; regulatory requirements typically mandate direct architect-client engagement for formal design review.
Adoption barriersclaude-sonnet-54/5Client trust, professional liability, and licensure expectations mean a licensed architect must personally engage with clients on design decisions and sign-offs.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of deploying an AI agent capable of authentic client meetings—including overhead for setup, integration with design systems, and required human oversight to prevent liability—would exceed the cost of having an architect conduct the meeting directly.
Cost vs. human wageclaude-sonnet-51/5There is no AI product performing this task, so cost comparison favors the human by default since AI cannot substitute the deliverable.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can draft text summaries or prepare materials for meetings, no deployed product reliably conducts actual client meetings with architectural drawings, interprets nuanced client feedback, or makes real-time design decisions in conversation. This remains largely a human-performed task with minimal AI product deployment.
Technical feasibility todayclaude-sonnet-51/5No deployed product conducts client meetings autonomously; AI tools support drawing prep but the meeting itself remains fully human-led.

Related occupations — Architecture & Engineering

How to read this

A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.

What would change this score

New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.