Electrical Engineers

17-2071.00

Median wage $120,630/yr198,750 employed (US)Rank #354 of 923 scored · top 38% by substitution

Research, design, develop, test, or supervise the manufacturing and installation of electrical equipment, components, or systems for commercial, industrial, military, or scientific use.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution32

Exposure31

Augmentation71

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

22 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

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%32

panel mean rating 2.3/5 → substitution pressure 32/100

Technical feasibility todayw 20%29

panel mean rating 2.2/5 → substitution pressure 29/100

Cost vs. human wagew 15%29

panel mean rating 2.2/5 → substitution pressure 29/100

Adoption barriersw 20%inverted — strong barriers lower the score34

panel mean rating 3.6/5 (barrier strength) → substitution pressure 34/100

Sector adoption velocityw 10%32

panel mean rating 2.3/5 → substitution pressure 32/100

Task breakdown (22 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.

Estimate labor, material, or construction costs for budget preparation purposes.

CI 43–62 · exposure 53 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Engineering and construction sectors are adopting AI cost tools (pilots and early production deployments are visible), but adoption remains uneven; large firms move faster than small ones, and many still rely on manual spreadsheets and legacy software.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering and construction sectors are historically slower AI adopters compared to information/finance industries, with pilots for cost estimation tools still emerging rather than widespread production use.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI-generated cost estimates with embedded material databases and labor multipliers substantially boost engineer productivity by automating data lookup and baseline calculations, while the engineer focuses on scope validation, risk adjustment, and client communication.
Augmentation potential	claude-sonnet-5	4/5	AI-powered estimating tools and generative models can significantly speed up drafting of cost breakdowns, flag pricing anomalies, and pull historical data, meaningfully boosting engineer productivity while they retain final judgment.
Task automatability	claude-haiku-4-5-20251001	4/5	AI can automate most of cost estimation by ingesting project specs, historical data, and current material/labor rates, then generating estimates with significant time savings. However, final sign-off typically requires domain expertise and judgment for novel project conditions, leaving ~20-30% of the cognitive work with the engineer.
Task automatability	claude-sonnet-5	3/5	AI can generate cost estimates using historical data, pricing databases, and parametric models, but accurate estimation requires site-specific judgment, current supplier quotes, and risk assessment that still need human validation.
Adoption barriers	claude-haiku-4-5-20251001	3/5	Professional liability for cost estimates creates oversight friction; engineers remain accountable if AI estimates prove wrong, and many organizations require human sign-off. Regulatory requirements vary by jurisdiction and project type, adding friction but not hard legal prohibition.
Adoption barriers	claude-sonnet-5	3/5	While no formal licensing mandates a PE for every cost estimate, liability for budget overruns and client/organizational trust in engineer judgment create meaningful friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI-based estimation tools cost roughly $50–200 per estimate in inference and licensing, while electrical engineers billing at $100–150/hour spend 2–4 hours estimating projects, making AI 5–10× cheaper per task-equivalent when deployed at scale.
Cost vs. human wage	claude-sonnet-5	3/5	AI tools can speed up preliminary estimates and reduce some labor hours, but licensing, integration, and required human review keep costs roughly comparable to traditional estimating workflows for complex projects.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	Commercial cost estimation software exists (e.g., RSMeans, Buildr, AI-enhanced tools), but they often require manual input validation, scope clarification, and expert review before budget submission. Products work reliably on routine projects but struggle with complex or non-standard requirements.
Technical feasibility today	claude-sonnet-5	2/5	Some construction cost estimating software includes AI-assisted features, but reliable engineering-grade estimates for electrical projects still depend heavily on human expertise and are not fully automated in production at scale.

Compile data and write reports regarding existing or potential electrical engineering studies or projects.

CI 48–50 · exposure 50 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Adoption is uneven: some engineering firms and tech-forward teams pilot AI-assisted report writing and data compilation, but widespread production deployment remains limited due to verification requirements and conservative engineering cultures.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering and technical consulting sectors have been slower than software/finance to deploy AI agents in production for core technical documentation, though pilots with AI writing assistants are emerging.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI substantially assists engineers by rapidly synthesizing data, generating first drafts, and producing visualizations, allowing engineers to focus on interpretation and validation rather than manual compilation and formatting—this is an active augmentation use case in professional practice.
Augmentation potential	claude-sonnet-5	4/5	AI tools are already useful for organizing data, generating first-draft report text, and summarizing findings, meaningfully speeding up the writing portion of this task while engineers verify content.
Task automatability	claude-haiku-4-5-20251001	3/5	AI can automate significant parts of report writing (structuring, literature synthesis, data compilation) and generate visual summaries from existing datasets, but typically requires human engineering judgment to validate technical conclusions, interpret results in context, and ensure accuracy of complex electrical analyses.
Task automatability	claude-sonnet-5	3/5	AI can draft report sections and compile structured data summaries from provided inputs, but engineering-specific data compilation and technical accuracy checks still require significant human setup and verification.rate about half automatable with substantial oversight.
Adoption barriers	claude-haiku-4-5-20251001	3/5	Engineering reports often feed into design decisions and safety-critical documentation, creating organizational and reputational friction; no strict licensing barrier to automation, but liability concerns and requirement for engineer sign-off on conclusions create meaningful adoption friction.
Adoption barriers	claude-sonnet-5	3/5	While no license is strictly required to draft a report, professional engineers often must review and stamp deliverables, and liability for technical errors in engineering studies creates meaningful friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	3/5	AI inference and integration costs are now low, but the task requires oversight and revision by qualified engineers to verify technical soundness, bringing total cost close to a human writing the report directly, though faster turnaround may offer modest savings.
Cost vs. human wage	claude-sonnet-5	3/5	AI drafting tools reduce time spent on report writing but still require engineer review and data validation, so total cost savings are moderate rather than an order of magnitude cheaper.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	Deployed tools (LLMs, document automation, BI platforms) can compile routine data and draft reports with reasonable accuracy, but struggle with domain-specific technical validation and often require material human review; production use exists but with quality gatekeeping overhead.
Technical feasibility today	claude-sonnet-5	3/5	LLM-based drafting and document assistants are deployed in engineering firms for report writing, but they lack native access to specialized electrical engineering data sources and require heavy human editing for technical correctness.

Perform detailed calculations to compute and establish manufacturing, construction, or installation standards or specifications.

CI 40–49 · exposure 45 · augmentation 75 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Large engineering firms and capital-intensive sectors (aerospace, utilities, energy) show moderate pilot adoption of simulation and computational assistants, but deployment remains cautious given liability and regulatory constraints; smaller practices lag significantly.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering firms are adopting AI-assisted design and calculation tools at a moderate pace, with pilots and partial integration into CAD/simulation workflows but not yet widespread autonomous deployment.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI-powered simulation, formula lookups, standard-library integration, and error-checking substantially boost engineering productivity within established workflows; tools like generative CAD and parametric design amplify engineer output while preserving human decision-making on trade-offs and safety.
Augmentation potential	claude-sonnet-5	4/5	AI tools significantly speed up calculations, generate first-pass specifications, and check for errors, meaningfully boosting engineer productivity while the engineer retains responsibility for final standards.
Task automatability	claude-haiku-4-5-20251001	3/5	AI systems can automate significant portions of routine calculations (load balancing, circuit dimensioning, standard compliance checks) and generate specifications from parameterized inputs, but typically require human review of assumptions, edge cases, and integration with domain-specific safety standards that demand professional judgment.
Task automatability	claude-sonnet-5	3/5	AI can perform many engineering calculations (load, thermal, tolerance analysis) using tools/scripts, but establishing standards requires validated judgment, code compliance checks, and integration with proprietary specs that current systems can't fully own end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Professional licensing (PE/Professional Engineer) requirements, liability exposure if automated specifications fail in the field, and regulatory mandates that a licensed engineer sign off on critical installations create hard gatekeeping; automation can support but not legally replace the engineer's certification.
Adoption barriers	claude-sonnet-5	4/5	Specifications often require a licensed Professional Engineer's stamp/sign-off, especially for construction and manufacturing standards tied to safety codes, creating strong regulatory and liability barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	Computational tools (cloud-based simulation, AI-assisted code generation) cost significantly less per task-instance than billable engineering hours, though integration and validation overhead moderates the advantage; rough 3–10× cost savings depending on complexity.
Cost vs. human wage	claude-sonnet-5	3/5	AI-assisted calculation tools can reduce time spent on repetitive computation, but licensed engineer review and liability oversight keep total cost comparable to traditional methods in regulated contexts.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	CAD software, simulation tools (MATLAB, ANSYS), and computational modules exist in production, but end-to-end specification generation from high-level requirements remains narrow in scope; most deployments handle narrow subtasks rather than the full calculation-to-specification pipeline independently.
Technical feasibility today	claude-sonnet-5	2/5	Engineering calculation tools with AI assistance exist (e.g., simulation copilots, code generation for calcs) but production use for authoritative standard-setting is narrow and still requires heavy engineer verification.

Prepare specifications for purchases of materials or equipment.

CI 34–50 · exposure 45 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Engineering organizations are cautious adopters; while AI-assisted drafting is emerging in some firms, production-level deployment of specification automation remains limited and most engineering teams still rely on manual or semi-manual processes.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering and manufacturing sectors are adopting AI-assisted drafting and documentation tools at a moderate pace, with pilots more common than full production deployment for specification writing specifically.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can significantly assist engineers by auto-populating routine fields, suggesting compliant standards, cross-referencing past specifications, and accelerating literature review, while the engineer retains control over final approval and design decisions.
Augmentation potential	claude-sonnet-5	4/5	AI tools substantially speed up drafting, formatting, and initial compilation of specifications, letting engineers focus on technical judgment and standards compliance rather than starting from scratch.
Task automatability	claude-haiku-4-5-20251001	3/5	AI can assist with routine specification generation, component research, and standard requirement documentation, potentially saving 40–50% of time on formulaic parts; however, critical design decisions, vendor evaluation, and integration with legacy systems typically require expert human judgment.
Task automatability	claude-sonnet-5	3/5	AI can draft specification documents from requirements and templates, but final specs require engineering judgment, verification against standards, and site-specific constraints that limit full end-to-end automation today.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Regulatory standards (IEC, IEEE, safety codes), liability for incorrect specifications, and implicit requirement that a licensed engineer take responsibility for specifications create strong legal and professional barriers to full automation.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement mandates a human write specs, but liability for equipment failures and compliance with codes/standards creates organizational caution about fully automating this output.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI tools have modest per-task cost (API calls or subscriptions), but oversight and correction by qualified engineers is substantial; total all-in cost remains high relative to direct human specification writing in most organizations.
Cost vs. human wage	claude-sonnet-5	3/5	AI drafting reduces time spent on boilerplate and formatting, but engineer oversight and validation costs remain significant, keeping overall cost roughly comparable to fully human-authored specs in many firms.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No mature product reliably generates complete, legally defensible equipment specifications end-to-end; AI tools can draft sections and suggest standards, but significant manual review and domain expertise remain necessary for accuracy and compliance.
Technical feasibility today	claude-sonnet-5	3/5	Generative AI tools and CAD/PLM-integrated systems can produce draft specifications, but deployed engineering products still require substantial human review before specs are used for procurement.

Operate computer-assisted engineering or design software or equipment to perform engineering tasks.

CI 34–46 · exposure 45 · augmentation 75 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Engineering firms are experimenting with AI-assisted design (generative design, topology optimization), but adoption remains in the pilot and augmentation phase rather than replacement. Regulatory conservatism, the need for human accountability, and high-value contract design work slow displacement of the core task.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering and manufacturing sectors show moderate AI tool adoption with growing pilots in generative design and simulation, but production-scale autonomous design remains uncommon.
Augmentation potential	claude-haiku-4-5-20251001	4/5	Generative design tools, AI-powered constraint solving, and simulation acceleration significantly boost engineer productivity—AI can rapidly explore design spaces, check feasibility, and suggest optimizations while the engineer maintains decision authority and validation. This is one of the higher-augmentation domains in engineering.
Augmentation potential	claude-sonnet-5	4/5	AI significantly boosts productivity in CAD/EDA tasks—auto-generating layouts, running simulations, flagging errors—while engineers retain control over final design decisions.
Task automatability	claude-haiku-4-5-20251001	3/5	CAD/CAM and simulation tools can automate routine design workflows, parametric updates, and constraint solving, but complex engineering decisions, trade-off analysis, and novel problem-solving still require human judgment. The task involves both automatable tool use and non-automatable expertise.
Task automatability	claude-sonnet-5	3/5	AI can assist with parts of CAD/EDA workflows (generating scripts, suggesting layouts, automating routine drafting) but full end-to-end operation of specialized engineering software for complex designs still requires substantial human setup and judgment.rating
Adoption barriers	claude-haiku-4-5-20251001	4/5	Engineering design output (schematics, CAD models) requires sign-off by licensed Professional Engineers in most jurisdictions, and liability for design errors rests with the responsible engineer. Client relationships, regulatory compliance, and safety certifications create strong organizational and legal barriers to full substitution.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement blocks AI use, but liability for design errors, safety certification needs, and organizational validation processes create meaningful friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Operating CAD software requires domain expertise that justifies the engineer's wage. While inference costs for AI assistance are falling, integrating autonomous design systems into existing workflows, maintaining quality checks, and oversight still make the total cost comparable to or higher than hiring the skilled human.
Cost vs. human wage	claude-sonnet-5	2/5	Licensing, integration, and oversight costs for AI-assisted design tools plus need for skilled engineer review keep costs comparable to or only modestly below human-only costs.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	Many deployed CAD systems (AutoCAD, SolidWorks, Fusion 360) exist with AI-assisted features for drafting and optimization, but full autonomous operation of professional-grade design tasks remains limited. Current AI can handle template-based designs and variations but struggles with novel constraints and validation.
Technical feasibility today	claude-sonnet-5	2/5	Some AI copilots exist in CAD/EDA tools (e.g., generative design, autorouting assistants) but they are narrow, error-prone for complex circuits, and not yet reliably autonomous in production engineering workflows.

Collect data relating to commercial or residential development, population, or power system interconnection to determine operating efficiency of electrical systems.

CI 30–50 · exposure 38 · augmentation 75 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Utilities are investing in digital transformation and data analytics tools, but adoption remains uneven; many organizations pilot these systems while human experts retain decision authority, typical of regulated infrastructure sectors.
Sector adoption velocity	claude-sonnet-5	2/5	Utilities and engineering firms are historically slower adopters of AI-driven data pipelines compared to pure information-sector firms, though smart grid analytics is growing.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI substantially augments engineer productivity by automating routine data aggregation, pattern detection in system performance, and generating preliminary efficiency reports, allowing engineers to focus on interpretation, anomaly investigation, and strategic recommendations.
Augmentation potential	claude-sonnet-5	4/5	AI tools can significantly speed up data aggregation, pattern detection, and preliminary efficiency analysis, letting engineers focus on interpretation and decision-making.
Task automatability	claude-haiku-4-5-20251001	3/5	AI can automate parts of data collection from structured sources (SCADA systems, public databases, GIS data) and initial efficiency analysis, but determining operating efficiency often requires domain expertise interpretation, site-specific context, and validation against real-world constraints that still require human judgment.
Task automatability	claude-sonnet-5	2/5	Data collection spans field surveys, utility records, and engineering measurements that require physical access, coordination with stakeholders, and domain judgment about relevance, limiting full automation despite AI's ability to aggregate and process structured data.
Adoption barriers	claude-haiku-4-5-20251001	3/5	Electrical utilities often require licensed Professional Engineers to sign off on efficiency determinations and system recommendations; regulatory compliance and liability concerns create moderate friction, though data collection itself faces fewer barriers.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement for data collection itself, but utility interconnection data often involves regulatory reporting, confidentiality, and coordination requirements that create friction.
Cost vs. human wage	claude-haiku-4-5-20251001	3/5	AI-driven data collection and initial analysis can reduce labor costs significantly, but integration with existing power system infrastructure, validation workflows, and the need for expert review keep overall costs roughly comparable to a qualified engineer's time.
Cost vs. human wage	claude-sonnet-5	2/5	Sensor networks, site visits, and utility coordination still require significant human labor and capital investment, so AI-driven data collection doesn't yet undercut human costs by an order of magnitude.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	Products exist for automating data aggregation and preliminary analysis (data pipeline tools, SCADA integration platforms), but no single system reliably performs end-to-end efficiency determination at production scale without human oversight and domain validation.
Technical feasibility today	claude-sonnet-5	2/5	Some data analytics and GIS-integration tools exist for utilities, but end-to-end automated collection across commercial/residential development and interconnection data is not a mature deployed product category.

Prepare technical drawings, specifications of electrical systems, or topographical maps to ensure that installation and operations conform to standards and customer requirements.

CI 34–41 · exposure 45 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	While some engineering firms pilot AI-assisted design tools, widespread production adoption remains limited. Many firms rely on traditional CAD workflows, and regulatory requirements and client expectations for human-verified work slow faster transition to AI automation.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering/design sectors are adopting AI-assisted CAD and generative design at a moderate pace, with pilots and partial integration common but full automation of certified deliverables still rare.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI tools meaningfully assist engineers by automating routine routing, generating initial layouts, checking standards compliance, and flagging potential errors. This substantially raises engineer productivity in drawing preparation while the engineer retains critical design control.
Augmentation potential	claude-sonnet-5	4/5	AI markedly speeds up drafting, boilerplate specification writing, and error-checking against standards, letting engineers focus on validation and customization while staying in the loop.
Task automatability	claude-haiku-4-5-20251001	3/5	AI can partially automate drawing generation from textual specifications and standards, including layout assistance and component placement. However, the task requires domain expertise to ensure safety compliance and customer-specific requirements, necessitating significant human review and manual refinement for full compliance.
Task automatability	claude-sonnet-5	3/5	AI/CAD tools can draft initial schematics and specification documents from parametric inputs, but final engineering drawings require validated engineering judgment, code compliance checks, and iterative customer-specific customization that current tools cannot fully replace.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Electrical drawings and topographical maps often require professional licensure (PE stamp in many jurisdictions), and liability for incorrect designs rests with the licensed engineer who must sign off. Regulatory frameworks and customer contracts typically mandate human professional responsibility for final drawings.
Adoption barriers	claude-sonnet-5	4/5	Electrical drawings for installations often require a licensed Professional Engineer's stamp/sign-off for code compliance and liability, creating a strong regulatory barrier to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI-assisted CAD systems reduce drawing time modestly but require expensive infrastructure, licensing, and significant skilled human oversight. The combined cost of AI tools plus required engineer review remains comparable to or higher than direct human drawing production.
Cost vs. human wage	claude-sonnet-5	2/5	AI can speed up drafting but licensed engineering software, integration, and mandatory human review/stamping keep the all-in cost comparable to or only modestly cheaper than an engineer's time.
Technical feasibility today	claude-haiku-4-5-20251001	3/5	CAD tools with AI-assisted features (e.g., automated routing, parametric design) exist and are deployed in practice, but they still require substantial human oversight and frequently require rework to meet precise standards and customer needs. These tools augment rather than replace the full task.
Technical feasibility today	claude-sonnet-5	2/5	CAD/BIM plugins and generative design tools exist but are narrow-scope assistive features embedded in engineering software rather than end-to-end autonomous drawing production used reliably in production without engineer review.

Develop software to control electrical systems.

CI 28–46 · exposure 38 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Technology and engineering firms are piloting AI-assisted code generation, but adoption remains cautious in regulated electrical and power domains; most teams use AI as a helper rather than a replacement.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering and embedded software sectors are adopting AI coding assistants steadily, but electrical/controls engineering is less digitized and slower than pure software industries.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI code assistants meaningfully accelerate drafting, debugging, and documentation for electrical engineers, allowing faster iteration while the engineer retains design and safety decisions.
Augmentation potential	claude-sonnet-5	4/5	AI coding assistants meaningfully speed up writing, debugging, and documenting control software, letting engineers focus on system design and validation while AI handles boilerplate and drafting.
Task automatability	claude-haiku-4-5-20251001	2/5	Only isolated parts of software development for electrical systems can be automated (code generation snippets, boilerplate), but the full task requires domain expertise, system-level reasoning, safety constraints, and integration with hardware—areas where current AI falls short of 50% time savings at equal quality.
Task automatability	claude-sonnet-5	3/5	AI coding assistants can generate significant portions of control software (boilerplate, drivers, standard control loops) but embedded/real-time control code for electrical systems still requires substantial human design, testing, and safety validation.'},'
Adoption barriers	claude-haiku-4-5-20251001	4/5	Electrical system software often falls under safety standards (IEC, UL) and regulatory frameworks requiring human engineer sign-off and accountability; liability for system failures creates strong organizational and legal barriers to full automation.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement to write software itself, but safety certification, liability for faulty control systems, and engineering sign-off requirements create meaningful friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI coding tools reduce junior-level drafting costs but do not yet eliminate the need for expert review, testing, and redesign; total integrated cost (with oversight) remains substantial relative to a mid-career engineer's output.
Cost vs. human wage	claude-sonnet-5	2/5	AI tools reduce coding time but the overall cost of developing, testing, and certifying control software still requires expert engineers, oversight, and hardware-in-loop validation, keeping savings moderate rather than order-of-magnitude.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	AI coding assistants exist and can draft routine code, but production software for electrical systems demands rigorous validation, safety certification, and real-time constraints that deployed products have not reliably achieved end-to-end.
Technical feasibility today	claude-sonnet-5	3/5	Deployed coding copilots (Copilot, Cursor, etc.) are used in production for embedded and control software development, but reliability for safety-critical electrical control systems remains limited and requires heavy human review.

Investigate or test vendors' or competitors' products.

CI 25–35 · exposure 25 · augmentation 75 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	While engineering firms use some AI-assisted tools for competitive intelligence and specification parsing, full automation of vendor testing and investigation remains rare in production. Adoption is limited to analytics acceleration rather than replacement of the core investigation task.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering and hardware-testing functions show slower AI adoption than purely digital/informational tasks, with pilots more common in data analysis than physical testing workflows.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI systems can significantly assist engineers by automating literature searches, aggregating competitor specifications, generating test protocols, and analyzing performance data, allowing engineers to focus on critical judgment and hands-on validation. Tools for design comparison and failure-mode analysis demonstrably raise productivity on this task.
Augmentation potential	claude-sonnet-5	4/5	AI tools can meaningfully assist with literature/datasheet review, competitive analysis synthesis, test planning, and report drafting, significantly speeding up parts of the investigative process while humans perform actual testing.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist with some aspects like data analysis and benchmarking of competitor specs, investigating and testing products requires hands-on evaluation, judgment about practical performance, and contextual decision-making that current AI cannot execute end-to-end. The core investigative and testing work remains dependent on human engineers.
Task automatability	claude-sonnet-5	2/5	Physical testing, benchmarking, and hands-on evaluation of hardware/products requires lab work, instrumentation, and judgment that AI cannot perform end-to-end; AI can assist with analysis and documentation but not the core investigative/testing process.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Professional judgment, engineering liability, and warranty/safety considerations mean that final recommendations and test conclusions typically require sign-off by a licensed engineer. Many jurisdictions and standards bodies (IEC, IEEE) require human engineering responsibility for product assessment, creating structural legal and regulatory barriers to full automation.
Adoption barriers	claude-sonnet-5	2/5	No licensing requirement specifically for this task, but organizational reliance on engineering expertise and physical lab access creates moderate friction against automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI can reduce some overhead (competitive analysis, initial data gathering) but the core testing and investigation still requires significant human engineering time. The all-in cost of AI assistance plus human oversight does not yet achieve substantial savings over direct human investigation.
Cost vs. human wage	claude-sonnet-5	2/5	Physical test equipment, lab access, and engineering judgment still require human labor; AI reduces some analysis and reporting time but doesn't replace the bulk of hands-on cost.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed AI system reliably conducts independent product investigations or testing in production. Existing tools can parse specifications and summarize competitor datasheets, but cannot autonomously perform physical testing, design evaluation, or reliability assessment at the quality an engineer would deliver.
Technical feasibility today	claude-sonnet-5	2/5	No deployed AI product autonomously conducts competitive product testing or vendor evaluation for electrical engineering; some AI-assisted data analysis and report generation tools exist but the physical/technical testing itself remains manual.

Plan or implement research methodology or procedures to apply principles of electrical theory to engineering projects.

CI 25–32 · exposure 25 · augmentation 63 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Engineering firms are relatively cautious with automation in core technical functions; adoption remains in the pilot and assistive phase rather than deep production displacement. The sector lags information/finance in adoption velocity.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering firms are adopting AI tools for simulation, code generation, and literature synthesis at a moderate pace, with pilots common but full methodological planning still human-led in most organizations.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can usefully augment engineers by rapidly generating literature summaries, suggesting methodological frameworks, or drafting procedural documents, which can accelerate planning cycles. However, the engineer must retain full control over validation and final design.
Augmentation potential	claude-sonnet-5	4/5	AI can significantly assist by rapidly synthesizing literature, suggesting experimental designs, performing calculations, and drafting documentation, meaningfully speeding up parts of the methodology development process while the engineer retains overall control.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist in drafting research plans and suggesting methodologies based on existing literature, the task requires domain expertise, creative problem-solving, and contextual judgment about novel electrical projects that current systems cannot reliably execute end-to-end. AI cannot independently validate that a proposed methodology will successfully apply electrical theory to a specific project's constraints.
Task automatability	claude-sonnet-5	2/5	Designing research methodology requires domain judgment, novel problem framing, and integration with physical constraints that current AI cannot reliably perform end-to-end; AI can assist with literature review, drafting, and calculations but cannot independently plan valid research procedures for engineering projects.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Regulatory and professional standards often require licensed engineers to sign off on research methodology for projects affecting public safety or regulatory compliance. Professional responsibility and liability asymmetries create strong barriers to full automation of these decisions.
Adoption barriers	claude-sonnet-5	3/5	While no license is strictly required to plan research methodology, professional engineering practice often requires PE oversight or sign-off for safety-critical designs, and organizational trust in novel methodologies from unproven AI systems creates moderate friction.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	An electrical engineer's loaded cost is substantial ($80k–$130k annually for senior roles), and AI systems would need significant integration, validation, and expert review overhead. The cost savings do not approach an order of magnitude given the domain-specific oversight required.
Cost vs. human wage	claude-sonnet-5	2/5	Because AI cannot perform the full task reliably, human engineers must still do the core methodology design and validation, so AI usage adds cost as a supplementary tool rather than substituting the human's costly time.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Some research tools and literature review systems exist, but no deployed product reliably plans or implements complete research methodologies for electrical engineering projects independently. Existing systems operate at the suggestion/draft level, requiring heavy human oversight and modification.
Technical feasibility today	claude-sonnet-5	2/5	No deployed product autonomously plans or implements electrical engineering research methodologies; existing AI tools (copilot-style coding/design assistants, literature summarizers) support fragments of this but are not production systems performing the full task.

Oversee project production efforts to assure projects are completed on time and within budget.

CI 28–28 · exposure 25 · augmentation 75 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Project management software and AI-assisted dashboards are widely adopted in engineering firms, but automation is limited to reporting and flagging; actual oversight and corrective action remain human-driven. Adoption of AI as a partial assistant is moderate, not rapid displacement.
Sector adoption velocity	claude-sonnet-5	3/5	Engineering and construction-adjacent sectors are adopting AI-based project analytics and scheduling tools at a moderate pace, with pilots more common than full production reliance.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI tools that integrate real-time cost and schedule data, predict slippages, and surface risks meaningfully augment a project manager's ability to oversee work. Dashboards, predictive analytics, and anomaly detection can raise human productivity in oversight tasks while the engineer retains full decision authority.
Augmentation potential	claude-sonnet-5	4/5	AI project-tracking, forecasting, and reporting tools can meaningfully augment an engineer's ability to monitor timelines and budgets, improving decision-making while the human remains responsible for oversight.
Task automatability	claude-haiku-4-5-20251001	2/5	Project oversight requires real-time judgment on budget and schedule deviations, stakeholder coordination, and adaptive decision-making that current AI cannot fully handle without human oversight. AI can assist with schedule tracking and budget monitoring, but cannot autonomously manage the interpersonal and contingency-response aspects that dominate oversight work.
Task automatability	claude-sonnet-5	2/5	Overseeing production efforts involves real-time coordination, judgment calls, vendor negotiation, and physical/site accountability that current AI cannot autonomously execute end-to-end.dient AI can support scheduling and budget tracking but not the oversight function itself.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Legal and professional responsibility heavily favor human oversight: a licensed engineer or recognized project lead typically must sign off on budget and schedule decisions, especially in regulated industries. Organizational and contractual norms require human accountability for project delivery.
Adoption barriers	claude-sonnet-5	4/5	Engineering managers often bear legal/contractual accountability for project delivery, safety, and budget sign-off, creating strong organizational and liability barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI project management and monitoring tools cost money but still require a human project manager to oversee them, so the all-in cost of AI+human often exceeds the cost of a skilled engineer/PM doing the work directly. Automation does not yet eliminate the human.
Cost vs. human wage	claude-sonnet-5	2/5	AI tools can reduce some reporting and tracking overhead, but human oversight, escalation, and accountability still dominate the cost structure, keeping overall cost comparable to or only modestly cheaper than a human manager.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While AI tools exist for project tracking and cost analysis, no deployed product reliably performs full project oversight autonomously. Existing systems support dashboarding and alerts but require human project managers to interpret, prioritize, and act on exceptions.
Technical feasibility today	claude-sonnet-5	2/5	Project management software with AI-assisted forecasting exists, but no deployed product independently 'oversees' engineering production to ensure timeline/budget adherence without a human manager driving decisions.

Investigate customer or public complaints to determine the nature and extent of problems.

CI 25–30 · exposure 25 · augmentation 63 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Electrical engineering and utilities are moderately digitized but deeply regulated; AI adoption in complaint investigation remains limited to intake tools, with human investigation still the norm due to safety and liability constraints.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering and utility sectors have historically slower AI adoption for field investigation tasks compared to purely digital information-processing industries, with pilots more common than production deployment.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by auto-categorizing complaints, summarizing issues, and flagging patterns, improving triage efficiency, but the engineer remains essential for actual root-cause diagnosis and safety sign-off.
Augmentation potential	claude-sonnet-5	4/5	AI tools can help summarize complaint histories, detect patterns across data, and draft investigation reports, meaningfully speeding up the engineer's analysis while the engineer still directs the investigation.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can help categorize and summarize complaint text, determining the true nature and extent of problems typically requires domain expertise, site visits, safety judgment, and contextual understanding of complex electrical systems that current systems struggle with end-to-end.
Task automatability	claude-sonnet-5	2/5	Investigation requires gathering field data, judging credibility, and often physical inspection or diagnostic testing that AI cannot perform autonomously; AI can assist with triage but not fully substitute for the investigation itself.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Professional licensure (PE/journeyman requirements in many jurisdictions), liability for missed safety hazards, regulatory compliance in electrical work, and customer expectation for qualified human engineers create strong adoption barriers.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement to investigate complaints, but liability concerns, safety issues in electrical systems, and need for accountable engineering judgment create moderate friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI can reduce initial intake overhead but the core investigation—site assessment, diagnostics, safety evaluation—remains labor-intensive; total cost savings are modest compared to the loaded wage of an electrical engineer.
Cost vs. human wage	claude-sonnet-5	2/5	AI can cheaply process complaint text and flag patterns, but the core investigative work (site visits, technical diagnosis, stakeholder communication) still requires costly human engineering time, keeping overall cost savings limited.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	AI tools can assist with complaint intake and initial triage, but no deployed products reliably investigate the full scope and root cause of electrical problems without substantial human engineering judgment and field work.
Technical feasibility today	claude-sonnet-5	2/5	Some products support complaint triage, log analysis, or anomaly detection, but no deployed system reliably conducts full technical investigations of electrical complaints without human engineers directing and interpreting findings.

Assist in developing capital project programs for new equipment or major repairs.

CI 25–30 · exposure 25 · augmentation 75 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Engineering firms and industrial organizations are early-stage in adopting AI for capital planning; most adoption remains at pilot or analytical-support levels rather than production deployment of autonomous program development systems.
Sector adoption velocity	claude-sonnet-5	2/5	Capital planning in engineering-heavy industrial and utility sectors adopts AI tools slowly, with more emphasis on traditional project management systems than generative AI agents.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can meaningfully assist engineers by automating literature review, generating cost and schedule baselines, drafting sections of project documentation, and flagging inconsistencies—substantially raising productivity while the engineer maintains final authority and judgment.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully assist by drafting proposals, running cost-benefit scenarios, summarizing technical specifications, and generating reports, significantly speeding up parts of the planning process while the engineer retains decision authority.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can help gather data, cost estimates, and generate preliminary project summaries, developing capital project programs requires domain expertise, technical judgment, and stakeholder coordination that currently cannot be fully automated end-to-end. Most of the value-add involves synthesis of business requirements, technical constraints, and organizational strategy—areas where human oversight remains essential.
Task automatability	claude-sonnet-5	2/5	This task involves cross-functional planning, stakeholder negotiation, budgeting judgment, and physical equipment assessment that current AI cannot execute end-to-end; AI can assist with drafting and analysis but not replace the core coordination and judgment work.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Capital projects carry significant financial and operational risk; liability, sign-off requirements, and organizational governance mandate that licensed engineers review and approve program development. Regulatory frameworks and corporate accountability structures create strong friction against full automation.
Adoption barriers	claude-sonnet-5	3/5	While no formal licensing mandates a PE sign capital project programs specifically, engineering judgment and organizational accountability for capital expenditures create meaningful friction against pure automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI-assisted tools reduce overhead in research and documentation phases, but the cost of infrastructure, integration, and human oversight combined with the senior-level expertise required means total delivered cost is not substantially cheaper than a skilled engineer's time on this task.
Cost vs. human wage	claude-sonnet-5	2/5	AI can cheaply generate drafts, cost estimates, or schedules, but the human engineer still must validate technical feasibility, negotiate budgets, and integrate site-specific constraints, keeping overall cost savings modest.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Some tools exist to assist with budget modeling, timeline visualization, and document generation, but no deployed system reliably performs the full scope of capital project program development independently. Current products handle narrow subtasks (e.g., cost estimation templates) with material gaps in contextual judgment and cross-functional integration required in practice.
Technical feasibility today	claude-sonnet-5	2/5	No deployed product autonomously develops capital project programs; existing tools (project management software, cost estimators) provide support but require heavy engineer involvement and customization per project.

Conduct field surveys or study maps, graphs, diagrams, or other data to identify and correct power system problems.

CI 25–30 · exposure 25 · augmentation 63 · importance 3.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Utilities and power companies are large, regulated, risk-averse organizations with legacy systems and slow digitization. While some data analytics adoption exists, AI-driven autonomous field survey and problem identification remains largely in pilot or experimental stages rather than broad production deployment.
Sector adoption velocity	claude-sonnet-5	2/5	Utility and power sectors are traditionally slow adopters of AI compared to information/finance industries, with pilots for predictive maintenance more common than full production deployment.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can meaningfully assist engineers by highlighting anomalies in grid data, automating routine diagram analysis, and cross-referencing patterns across maps and historical records, improving the speed of problem identification. However, the human expert remains essential for judgment, validation, and field-level decision-making.
Augmentation potential	claude-sonnet-5	4/5	AI tools can significantly help analyze maps, sensor data, and historical grid data to flag likely problem areas, improving engineer efficiency even though field verification remains human-led.
Task automatability	claude-haiku-4-5-20251001	2/5	AI can assist with data interpretation and anomaly detection in power system diagrams and maps, but the task requires contextual field knowledge, real-time system state judgment, and integration of multiple data sources that AI cannot perform end-to-end reliably. Field surveys in particular demand physical presence and expert judgment about on-site conditions.
Task automatability	claude-sonnet-5	2/5	Field surveys require physical presence and hands-on inspection of infrastructure, which AI cannot perform; only the data/map analysis portion is partially automatable.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Electrical engineers performing field surveys and identifying system problems face significant regulatory and liability barriers; utilities and grid operators are heavily regulated, require licensed professionals to sign off on safety-critical findings, and bear high costs for incorrect diagnoses. Organizational risk aversion and safety standards create strong friction against full automation.
Adoption barriers	claude-sonnet-5	3/5	Utility infrastructure work often involves safety regulations, licensing for field engineers, and liability for misdiagnosed power system faults, creating moderate barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI solutions (vision models, data analysis) still require substantial integration, calibration to domain-specific systems, and human expert review, making total cost per task comparable to or exceeding the loaded cost of an experienced electrical engineer.
Cost vs. human wage	claude-sonnet-5	2/5	Physical field survey work still requires human labor and equipment, so AI only reduces cost for the analytical/desk portion of the task, not the full task cost.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While AI tools exist for circuit diagram analysis and some anomaly detection in power grid data, production systems are narrow in scope and require significant human oversight. No deployed product reliably conducts independent field surveys or integrates field observations with map/diagram analysis at the required accuracy level.
Technical feasibility today	claude-sonnet-5	2/5	Some GIS and diagnostic tools assist with power system data analysis, but no deployed product autonomously conducts field surveys or reliably identifies and corrects power system problems end-to-end.

Integrate electrical systems with renewable energy systems to improve overall efficiency.

CI 25–28 · exposure 25 · augmentation 75 · importance 3.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Engineering firms and energy companies use computational tools to accelerate design, but most renewable energy integration projects still follow traditional engineer-led workflows. Adoption of autonomous AI agents for end-to-end integration is nascent; pilots exist but production replacement is limited.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering/utility sectors are traditionally slower adopters of AI tools compared to software or finance, though renewable energy modeling software adoption is growing steadily.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI-powered simulation, optimization, and code generation tools significantly assist engineers in modeling renewable integration scenarios, running parametric studies, and generating documentation. These tools measurably boost productivity while the engineer retains design and validation authority.
Augmentation potential	claude-sonnet-5	4/5	AI-assisted simulation, load calculations, design optimization, and documentation can meaningfully speed up the engineer's analysis and iteration cycles while the engineer retains responsibility for final integration decisions.
Task automatability	claude-haiku-4-5-20251001	2/5	Integrating electrical systems with renewable energy requires domain expertise, custom design decisions, and site-specific optimization that current AI cannot execute end-to-end reliably. AI can assist with calculations, simulations, and documentation, but the complexity of system design, regulatory compliance, and real-world constraints demands human engineering judgment.
Task automatability	claude-sonnet-5	2/5	This involves physical system integration, site-specific engineering judgment, and hands-on design work that current AI cannot execute end-to-end; AI can assist with calculations and modeling but not perform the full integration task.
Adoption barriers	claude-haiku-4-5-20251001	4/5	System integration for renewable energy involves regulatory compliance (grid interconnection standards, building codes, safety certifications) and often requires licensed professional engineers to design and certify systems. Liability for system failures and legal sign-off requirements create substantial barriers to full automation.
Adoption barriers	claude-sonnet-5	4/5	Electrical engineering work often requires a licensed PE for sign-off, especially on grid-connected renewable systems, plus safety codes and utility interconnection standards that mandate human accountability.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI tools (simulation software, design assistants) reduce some analysis time, but the integrated cost of the tool, required oversight, and human verification remains comparable to or higher than the labor cost for a skilled electrical engineer to perform the task.
Cost vs. human wage	claude-sonnet-5	2/5	Human engineers remain necessary for physical integration, compliance, and site-specific decisions, so AI tools reduce some analysis time but don't replace the bulk of billable engineering cost.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While simulation and analysis tools exist (some AI-enhanced), no deployed product reliably executes full system integration autonomously. Tools require significant human direction on design parameters, code verification, and validation against safety standards, limiting production-ready autonomy.
Technical feasibility today	claude-sonnet-5	2/5	AI tools exist for simulation and design assistance (e.g., PVsyst-like modeling, load flow analysis aids) but no deployed product autonomously performs full system integration in production engineering workflows.

Design, implement, maintain, or improve electrical instruments, equipment, facilities, components, products, or systems for commercial, industrial, or domestic purposes.

CI 25–25 · exposure 25 · augmentation 75 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Adoption is concentrated in narrow areas (circuit simulation, PCB layout suggestion, code generation for firmware) but full design automation remains rare in production. Conservative risk tolerance in regulated industries and the complexity of integrating AI into legacy workflows limit velocity of deep adoption.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering and industrial sectors adopt AI tools for design assistance at a moderate pace, but physical implementation and maintenance work remains slow to digitize compared to purely digital professional services.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI systems demonstrably augment engineer productivity: circuit simulators, CAD tools with generative features, code assistants, and optimization solvers all help engineers explore designs faster and reduce routine drafting work. These tools remain assistive rather than autonomous, keeping expert judgment in the loop.
Augmentation potential	claude-sonnet-5	4/5	AI significantly aids engineers in circuit design, simulation, documentation, and troubleshooting diagnostics, meaningfully boosting productivity while the engineer retains responsibility for physical implementation and judgment.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist with narrow components (circuit analysis, simulation, code generation), the full end-to-end design cycle requires domain judgment, tradeoff analysis, safety validation, and integration testing that current systems cannot reliably perform without expert oversight. No single AI tool achieves 50% time savings on the complete design-implement-maintain workflow at equal quality.
Task automatability	claude-sonnet-5	2/5	This is a broad, physical-world engineering task spanning design, hands-on implementation, and maintenance of hardware; AI can assist with parts like schematic drafting or simulation but cannot execute the full end-to-end task at equal quality with major time savings.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Electrical engineering designs are subject to strict regulatory approval, building codes, safety standards (UL, IEC), and professional licensing requirements in many jurisdictions. A licensed professional engineer must typically sign off on designs, and liability for failures falls on accountable humans, creating hard barriers to unvetted AI automation.
Adoption barriers	claude-sonnet-5	4/5	Electrical engineering work is often subject to professional licensure (PE requirements), safety codes, and liability for equipment failures, creating strong barriers to full automation and sign-off substitution.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI tools (cloud-based simulation, GitHub Copilot, design software subscriptions) reduce some engineering hours, but the full loaded cost of integration, validation, liability oversight, and human review remains comparable to or higher than the marginal cost of hiring engineers for complex systems.
Cost vs. human wage	claude-sonnet-5	2/5	AI tools reduce some design/analysis time but the task still requires licensed engineers, physical installation, and field maintenance, keeping overall AI-driven cost savings modest relative to full human labor cost.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Deployed tools (simulation software, CAD copilots, code assistants) support isolated subtasks, but no integrated product reliably performs full design, implementation, maintenance, and improvement cycles in production. High error costs in electrical systems (safety, regulatory compliance) limit real-world deployment of end-to-end automation.
Technical feasibility today	claude-sonnet-5	2/5	CAD/EDA tools with AI features and simulation software exist and are used in production, but they support engineers rather than autonomously performing design, installation, and maintenance of electrical systems.

Direct or coordinate manufacturing, construction, installation, maintenance, support, documentation, or testing activities to ensure compliance with specifications, codes, or customer requirements.

CI 25–25 · exposure 25 · augmentation 63 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Adoption of AI for manufacturing coordination and compliance oversight remains in pilot phases. Most manufacturing and construction firms rely on traditional project management and human engineers for oversight, with slow adoption of AI-driven coordination systems.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering/construction/manufacturing sectors have historically slower AI adoption for coordination and compliance oversight roles compared to purely digital information sectors.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can meaningfully assist engineers by automating compliance checking, generating documentation, analyzing test results, and flagging deviations from specifications, thereby reducing manual review overhead while the engineer retains oversight authority and final decision-making.
Augmentation potential	claude-sonnet-5	4/5	AI tools can meaningfully assist by automating documentation, flagging code compliance issues, tracking project status, and drafting reports, freeing engineers to focus on judgment-heavy coordination.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist with documentation review, compliance checking, and testing data analysis, the core responsibility of directing and coordinating multiple teams, making real-time decisions on construction/installation sites, and ensuring compliance requires human judgment and accountability. Current AI cannot independently oversee complex, dynamic manufacturing environments end-to-end with sufficient reliability to meet the 50% time-saving threshold.
Task automatability	claude-sonnet-5	2/5	This is a coordination and oversight task requiring real-time judgment, cross-team communication, and physical-site awareness that current AI cannot perform end-to-end; AI can support scheduling and documentation but not direct the overall activity.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Regulatory and liability barriers are substantial: engineering responsibility for compliance, codes, and customer requirements often rests legally on a qualified engineer; certification and professional accountability requirements limit substitution. Professional licensure and customer expectations for human accountability create significant adoption friction.
Adoption barriers	claude-sonnet-5	4/5	Compliance with codes and customer requirements often requires a licensed professional engineer's sign-off and accountability, creating strong regulatory and liability barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	The cost of AI systems (specialized compliance/coordination software, integration, human oversight) combined with ongoing human supervision still approaches or exceeds the loaded wage of an electrical engineer, particularly given liability exposure.
Cost vs. human wage	claude-sonnet-5	2/5	Human engineers remain necessary for site coordination, stakeholder communication, and liability sign-off, so AI supplementation reduces some documentation costs but doesn't replace the bulk of labor cost involved in coordination roles.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Limited deployed products exist for comprehensive coordination of manufacturing/construction activities. AI can handle narrow components (compliance document review, test data logging) but no mature, production-scale systems reliably direct or coordinate multi-team operations with accountability for quality and safety.
Technical feasibility today	claude-sonnet-5	2/5	No deployed product autonomously directs or coordinates multi-disciplinary engineering activities; existing tools (project management software, compliance checkers) assist but require human direction and decision-making.

Inspect completed installations and observe operations to ensure conformance to design and equipment specifications and compliance with operational, safety, or environmental standards.

CI 25–25 · exposure 25 · augmentation 63 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Adoption remains limited to pilot projects and routine visual anomaly detection in large-scale operations. Most electrical and utilities sectors retain human inspection protocols for compliance and liability reasons, with AI tools used only as secondary aids rather than primary inspectors.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering and industrial inspection sectors are adopting AI-assisted monitoring tools gradually, but widespread deployment of autonomous inspection replacing engineers is still nascent, concentrated in pilots and augmented tools.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by flagging visual anomalies, automating data logging, and summarizing specification deviations for engineer review, moderately accelerating the inspection workflow. However, the requirement for expert judgment on safety and environmental compliance limits the transformative potential of pure augmentation.
Augmentation potential	claude-sonnet-5	4/5	AI-powered sensors, drones, thermal imaging, and anomaly-detection software can significantly speed up data collection and flag issues, letting the engineer focus judgment on flagged anomalies and final compliance decisions.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI-powered visual inspection and monitoring systems can detect some deviations from specifications, the task requires nuanced judgment about safety compliance, environmental standards, and operational readiness across diverse equipment types. Current systems struggle with context-dependent decision-making and can miss subtle non-conformances that require expert electrical knowledge.
Task automatability	claude-sonnet-5	2/5	Physical inspection of installations and observation of operations requires on-site sensory presence, judgment, and often hands-on verification that current AI cannot fully replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Electrical installation compliance and safety sign-off typically require licensed engineers; regulatory frameworks across jurisdictions mandate human professional responsibility for certification. Liability for missed safety violations creates strong institutional and legal barriers to full automation.
Adoption barriers	claude-sonnet-5	4/5	Engineering sign-off on safety and code compliance often legally requires a licensed professional engineer, and liability for missed defects is high, creating strong regulatory and professional barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI inspection systems require substantial setup, calibration, and integration costs, plus continuous human expert review of flagged items. For this high-stakes task, the total cost per installation check remains comparable to or exceeds a single engineer's inspection, especially when accounting for liability and re-verification.
Cost vs. human wage	claude-sonnet-5	2/5	Sensor/camera-based monitoring systems can be cost-effective for narrow anomaly detection, but full inspection requiring engineering judgment and physical presence still requires costly integration and human oversight, keeping overall cost comparable to or higher than a human engineer for full task scope.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Computer vision systems exist for routine visual inspections, but deployed products today lack the reliability needed for critical safety and compliance checks in electrical installations. Most operational systems are narrow in scope (e.g., detecting obvious physical damage) rather than comprehensive conformance auditing, and require significant human oversight.
Technical feasibility today	claude-sonnet-5	2/5	Some computer-vision and IoT sensor-based monitoring tools exist for equipment condition and anomaly detection, but comprehensive compliance inspection combining design specs, safety codes, and physical observation is not reliably automated in production.

Develop systems that produce electricity with renewable energy sources, such as wind, solar, or biofuels.

CI 25–25 · exposure 25 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Renewable energy engineering spans utilities, manufacturing, and consulting—moderate digitization sectors. While pilot adoption of AI-assisted design tools is increasing, production-level displacement of core system development work remains limited; adoption is slower than in software or finance.
Sector adoption velocity	claude-sonnet-5	2/5	Energy and engineering sectors have historically been slower to adopt AI-driven design automation compared to software or finance, though simulation tools are gradually being integrated.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI substantially augments engineers through system simulation, optimization algorithms, performance modeling, and forecasting tools that speed design iteration and analysis; the engineer remains in the loop for critical decisions, making AI a high-productivity multiplier for the task.
Augmentation potential	claude-sonnet-5	4/5	AI significantly aids engineers through simulation, optimization algorithms, predictive modeling of energy yield, and generative design exploration, meaningfully boosting productivity while humans retain design authority.
Task automatability	claude-haiku-4-5-20251001	2/5	Current AI can assist with simulation, design optimization, and performance analysis of renewable energy systems, but the full task requires complex system integration, site-specific engineering decisions, regulatory compliance, and cross-disciplinary coordination that AI cannot perform end-to-end with 50% time savings at equal quality.
Task automatability	claude-sonnet-5	2/5	System-level renewable energy design involves physical prototyping, site-specific engineering, regulatory compliance, and multidisciplinary tradeoffs that current AI cannot execute end-to-end; AI can assist with modeling and calculations but not the full development process.
Adoption barriers	claude-haiku-4-5-20251001	4/5	High barriers exist: electrical systems require licensed Professional Engineer (PE) sign-off, strict grid interconnection standards, safety certifications, environmental permitting, and liability exposure mean a qualified human engineer must legally review and approve the design before deployment.
Adoption barriers	claude-sonnet-5	4/5	Professional engineering licensure (PE stamps), safety codes, and utility interconnection regulations require licensed human engineers to sign off on system designs, creating strong legal barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI inference and simulation tools have low marginal cost, but the task requires significant human expert integration, site surveys, regulatory review, and validation; the total cost of AI-assisted development remains comparable to or higher than traditional engineering for complex renewable systems.
Cost vs. human wage	claude-sonnet-5	2/5	AI tools reduce some analysis and simulation time, but the engineering, testing, and certification work still requires substantial human labor, keeping overall costs comparable to human-led processes with modest AI cost savings.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While AI tools exist for component design (e.g., turbine blade optimization via generative models) and energy forecasting, no deployed product reliably performs the complete system development task covering feasibility studies, hardware integration, grid interconnection, and project management with acceptable error rates.
Technical feasibility today	claude-sonnet-5	2/5	Simulation and design-assistance tools (e.g., for solar layout optimization, load modeling) exist and are used, but no deployed product autonomously develops complete renewable energy generation systems.

Design electrical systems or components that minimize electric energy requirements, such as lighting systems designed to account for natural lighting.

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

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Engineering design remains a high-touch, site-specific discipline with slow digital transformation. While energy modeling tools are adopted, full AI-driven system design automation in production is rare; most firms use AI as an optimization aid rather than autonomous design agent.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering design firms adopt simulation and modeling tools steadily but cautiously, with AI-driven design generation still in pilot or narrow-tool stages rather than widespread production use.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI excels at energy simulation, parametric optimization, daylighting analysis, and design variant generation—significantly accelerating an engineer's exploration of efficient solutions. When combined with BIM and simulation platforms, AI can substantially boost engineering productivity while the engineer validates and integrates solutions.
Augmentation potential	claude-sonnet-5	4/5	AI-powered energy modeling, daylighting simulation, and generative design tools meaningfully speed up iteration and optimization for engineers who remain responsible for final designs.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist in energy optimization calculations and suggest design alternatives, this task requires creative problem-solving, site-specific constraints, and integration of building physics principles that currently demand human judgment. AI tools lack the ability to autonomously navigate tradeoffs between aesthetics, functionality, code compliance, and energy efficiency across a full system design.
Task automatability	claude-sonnet-5	2/5	Energy-efficient electrical system design requires integrating building geometry, code compliance, load calculations, and site-specific factors that current AI can support but not fully execute end-to-end without significant human engineering judgment.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Electrical system design is subject to stringent building codes, safety regulations (NEC, NFPA), and often requires licensed Professional Engineer (PE) sign-off. Liability for failures, energy code compliance verification, and customer accountability create hard legal barriers that prevent full substitution.
Adoption barriers	claude-sonnet-5	4/5	Electrical system designs typically require a licensed Professional Engineer's stamp and compliance with building codes and safety regulations, creating strong legal barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Simulation and optimization software licensing plus integration overhead remains substantial, and expert engineering oversight is still mandatory for safety and code compliance. The loaded cost of an electrical engineer significantly exceeds the incremental cost of AI-assisted tools, so no order-of-magnitude savings exists yet.
Cost vs. human wage	claude-sonnet-5	2/5	While simulation software reduces some iteration time, the overall cost of AI-assisted design still requires substantial licensed engineer oversight, keeping costs closer to comparable rather than dramatically cheaper.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	CAD tools and energy simulation software (LEED calculators, building performance models) exist but require significant human setup and validation. Current AI products cannot independently produce production-ready electrical system designs; they support calculation and visualization but engineers must verify feasibility, safety codes, and real-world applicability.
Technical feasibility today	claude-sonnet-5	2/5	Some CAD/BIM tools include energy modeling and daylighting simulation plugins, but these are decision-support tools requiring an engineer to configure, interpret, and finalize designs rather than autonomous design generation.

Confer with engineers, customers, or others to discuss existing or potential engineering projects or products.

CI 5–30 · exposure 13 · augmentation 63 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Engineering sectors show minimal adoption of AI for independent stakeholder conferencing; this task remains firmly human-owned in practice due to professional standards, client expectations, and the high cost of miscommunication. Pilots are uncommon and adoption is lagging across the industry.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering firms are moderate adopters of AI tools for documentation and analysis, but live consultative discussions with customers remain largely untouched by automation in practice.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by preparing meeting notes, summarizing prior discussions, generating draft agendas, and organizing technical data, which would moderately improve an engineer's preparation and documentation efficiency. However, the core interactive task still requires the engineer's judgment and presence.
Augmentation potential	claude-sonnet-5	4/5	AI tools can meaningfully assist by transcribing, summarizing meetings, preparing talking points, and drafting follow-up documentation, enhancing engineer productivity around these conversations.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can draft communications and summarize existing information, authentic two-way conferencing with engineers and customers requires real-time understanding of implicit requirements, negotiation, and relationship building that current AI systems cannot reliably perform autonomously. The task fundamentally depends on iterative human dialogue and context that exceeds AI's current conversational reliability.
Task automatability	claude-sonnet-5	1/5	This task is fundamentally interpersonal, real-time discussion involving negotiation, judgment, and relationship-building that current AI cannot substitute for end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Significant professional and legal barriers exist: engineers are expected to personally represent their expertise and commitments to customers; liability for design discussions typically requires a licensed engineer's signature or accountability. Organizations and clients generally require human engineering judgment for project conferencing.
Adoption barriers	claude-sonnet-5	3/5	No formal licensing requirement mandates a human for conversations, but strong organizational and customer-relationship norms, trust, and accountability favor human engineers leading such discussions.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	The cost of deploying AI systems with necessary oversight, error correction, and human fallback for critical engineering discussions approaches or exceeds the cost of having an engineer participate directly. Integration overhead remains substantial for mission-critical conversations.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI-only substitute for this task, so cost comparison favors the human; any AI use is supplementary rather than replacing the conferring itself.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed products reliably conduct independent engineering discussions with external stakeholders; AI chatbots exist but produce frequent misunderstandings on technical specifics and cannot commit to design decisions. Current systems require heavy human supervision and cannot replace human engineers in actual project conferences.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product autonomously conducts substantive engineering project discussions with customers or engineers on behalf of an engineer; AI meeting tools only support, not replace, this.

Supervise or train project team members, as necessary.

CI 5–21 · exposure 13 · augmentation 50 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Even in digitized sectors, the supervision and training of engineering teams remains firmly a human responsibility; no evidence of AI agents replacing these functions in production settings.
Sector adoption velocity	claude-sonnet-5	2/5	Engineering firms are adopting AI for technical tasks but management and supervisory functions see minimal AI deployment; adoption in this specific sub-task is slow.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can usefully assist by generating training materials, tracking team progress metrics, suggesting performance improvement areas, or scheduling sessions, but the core supervisory interaction remains fundamentally human-centered.
Augmentation potential	claude-sonnet-5	3/5	AI can assist with generating training content, scheduling, performance tracking, and knowledge resources, moderately aiding supervisors but not transforming the core interpersonal supervisory task.
Task automatability	claude-haiku-4-5-20251001	2/5	Supervising and training humans requires ongoing judgment about individual performance, learning curves, and interpersonal dynamics that current AI cannot reliably assess or adapt to in real-time, though AI could handle narrow subcomponents like scheduling training sessions or logging performance data.
Task automatability	claude-sonnet-5	1/5	Supervising and training team members requires interpersonal leadership, mentoring, motivation, and contextual judgment that current AI cannot perform end-to-end; no off-the-shelf system replaces a human supervisor/trainer role.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Supervision and training of professional staff carry significant organizational and cultural resistance: employees expect human mentorship, performance feedback requires legal/HR accountability, and liability for inadequate training typically falls on human management.
Adoption barriers	claude-sonnet-5	4/5	Supervisory authority typically requires organizational accountability, HR/legal responsibility, and often professional engineering oversight (PE sign-off chains), creating strong structural barriers to AI substitution.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task demands contextual understanding of individual team members, project-specific technical knowledge, and the ability to influence behavior and learning—capabilities AI cannot provide today at lower cost than a qualified engineer-supervisor.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI substitute performing this managerial task, so cost comparison favors the human entirely; any AI role is supplementary at added cost, not replacement.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No production system today reliably supervises or trains human team members end-to-end; current AI tools can support administrative aspects (scheduling, document generation) but cannot replace the real-time coaching, motivation, conflict resolution, and adaptive instruction that supervision demands.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product autonomously supervises or trains engineering staff; this remains a human management function, though AI tools may support training materials.

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.