Fire Inspectors and Investigators

33-2021.00
Median wage $75,920/yr13,800 employed (US)Rank #821 of 923 scored · top 89% by substitution

Inspect buildings to detect fire hazards and enforce local ordinances and state laws, or investigate and gather facts to determine cause of fires and explosions.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution14
Exposure15
Augmentation48

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

30 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%17

panel mean rating 1.7/5 → substitution pressure 17/100

Technical feasibility todayw 20%12

panel mean rating 1.5/5 → substitution pressure 12/100

Cost vs. human wagew 15%14

panel mean rating 1.6/5 → substitution pressure 14/100

Adoption barriersw 20%inverted — strong barriers lower the score13

panel mean rating 4.5/5 (barrier strength) → substitution pressure 13/100

Sector adoption velocityw 10%9

panel mean rating 1.4/5 → substitution pressure 9/100

Task breakdown (30 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.

Write detailed reports of fire inspections performed, fire code violations observed, and corrective recommendations offered.

57

CI 37–77 · exposure 66 · augmentation 88 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire inspection and code enforcement remain relatively lower-digitization, human-centric operations dominated by public agencies with slower IT adoption cycles. Pilot projects exist, but production deployment of autonomous report writing in municipal fire departments is still uncommon.
Sector adoption velocityclaude-sonnet-52/5Fire inspection remains a physically-grounded, government/municipal-heavy field with low digitization and slow tech adoption compared to information-sector norms.
Augmentation potentialclaude-haiku-4-5-202510015/5AI dramatically accelerates inspector productivity on report drafting by auto-populating violations, structure, and recommendations from inspection notes, freeing the human inspector to focus on judgment and completeness review. This is a textbook case of assistive technology that keeps the inspector in control while multiplying output speed.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully speed up drafting, standardize language, and suggest corrective recommendations based on cited codes, significantly aiding inspectors who then verify and finalize the report.
Task automatabilityclaude-haiku-4-5-202510015/5LLMs can today generate detailed, comprehensive reports from inspection data, checklists, and violation notes with high consistency and speed, meeting the >50% time-saving bar. Current AI systems reliably transform structured inspection inputs into properly formatted, thorough narratives covering violations and recommendations.
Task automatabilityclaude-sonnet-53/5AI can draft structured reports from inspector notes/photos and checklists, but requires accurate field observations and judgment about code applicability that must originate from a human inspection, limiting full end-to-end automation.
Adoption barriersclaude-haiku-4-5-202510013/5Fire inspection reports carry liability and must meet code documentation standards, creating moderate friction around independent AI authorship. Regulatory oversight of the inspection itself (not the report) remains the inspector's responsibility, so barriers are friction-based rather than legal prohibition—oversight and review are typically required before filing.
Adoption barriersclaude-sonnet-54/5Fire code violation reports often carry legal and regulatory weight requiring a certified inspector's sign-off, and liability for missed hazards creates strong incentive for human authorship and review.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI inference and integration cost per report is now a fraction of the loaded hourly cost of a fire inspector manually drafting detailed documentation, especially at volume. Cost advantage is substantial—one to two orders of magnitude cheaper per task.
Cost vs. human wageclaude-sonnet-53/5AI drafting assistance can cut report-writing time substantially, but the inspection itself and verification of code citations still require paid professional time, so overall cost savings are moderate rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510014/5Multiple document-generation products and deployed LLM systems handle report writing at scale in professional contexts. While minor human review for accuracy and liability remains common practice, mature AI systems already produce reportable-quality output in production fire marshal and inspection workflows.
Technical feasibility todayclaude-sonnet-52/5General-purpose LLM drafting tools and some inspection software offer report templating and narrative generation, but no widely deployed product reliably converts fire inspection findings into finished, code-citation-accurate reports without human review.

Prepare and maintain reports of investigation results, and records of convicted arsonists and arson suspects.

53

CI 39–67 · exposure 62 · augmentation 75 · importance 4.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Law enforcement and fire departments are mid-stage in AI adoption; many have implemented document management and reporting tools, but full end-to-end automation of arson records is still being piloted rather than universally deployed in production.
Sector adoption velocityclaude-sonnet-52/5Public safety and fire investigation agencies are typically slow adopters of AI due to legal, budgetary, and procedural constraints, with pilots rare and production use minimal.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can draft reports from inspector field notes and auto-populate arsonist databases, significantly accelerating the inspector's documentation workflow while the human retains control over accuracy and case decisions. This is a strong augmentation use case.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by drafting report templates, organizing case data, and searching records, improving efficiency while a human investigator remains responsible for accuracy and judgment.
Task automatabilityclaude-haiku-4-5-202510014/5AI can reliably extract, structure, and organize investigation findings into standardized reports, and can cross-reference suspect/convict databases for record maintenance with minimal human intervention. However, some judgment calls on case closure and anomaly flagging may still benefit from human review, preventing a full 5.
Task automatabilityclaude-sonnet-53/5Drafting and organizing investigation reports and maintaining structured records is largely text-based work that current AI can support substantially, but final compilation still requires human verification of facts, evidence chain-of-custody, and legal accuracy.},
Adoption barriersclaude-haiku-4-5-202510013/5Records management is subject to FOIA, discovery rules, and evidence-handling standards, creating moderate friction. While no law requires a human sign-off on database entries, audit trails and legal defensibility concerns drive organizational oversight requirements that slow pure automation.
Adoption barriersclaude-sonnet-54/5Fire investigation reports often feed into criminal prosecutions, requiring certified investigator sign-off, accurate legal documentation, and accountability that AI cannot assume.
Cost vs. human wageclaude-haiku-4-5-202510014/5Automating report writing and database maintenance is significantly cheaper than paying inspectors to do clerical work; AI infrastructure costs are low per task, and a single system can serve many inspectors, easily achieving 2–3× cost reduction compared to human-only workflows.
Cost vs. human wageclaude-sonnet-52/5AI can cut drafting time, but human investigators must still verify facts, maintain legal chain-of-custody, and finalize reports, so cost savings are moderate rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510014/5Document processing and database management systems with AI are mature and deployed in law enforcement agencies today. OCR, named-entity recognition, and record linkage work well on investigation reports, though some jurisdictions still rely on manual entry, limiting universal reliability to 4.
Technical feasibility todayclaude-sonnet-53/5Report-writing assistants and records-management software are deployed in many public-safety contexts, but no product independently and reliably compiles complete investigative arson reports without heavy human review.

Review blueprints and plans for new or remodeled buildings to ensure the structures meet fire safety codes.

31

CI 25–37 · exposure 33 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire inspection and code enforcement remain primarily human-driven, with limited automation adoption. Adoption is concentrated in data-entry and document management tasks rather than the core compliance judgment; production deployment of AI for final code review decisions is rare in public sector and municipal fire departments.
Sector adoption velocityclaude-sonnet-52/5Government fire/building departments are typically slow adopters of AI tools, with pilots emerging in some jurisdictions but production use still rare.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by automatically extracting and organizing blueprint data, highlighting potential code violations, and cross-referencing regulations, thereby reducing manual document review time. However, the inspector must retain full judgment authority, making this a moderate augmentation scenario rather than a transformation.
Augmentation potentialclaude-sonnet-54/5AI-assisted plan review tools can flag likely code violations and speed up initial screening, meaningfully boosting inspector productivity even though humans finalize decisions.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can extract and analyze data from blueprints using OCR and computer vision, reviewing for code compliance requires understanding complex, jurisdiction-specific fire safety regulations and making contextual judgments about structural adequacy. Current systems can flag obvious deviations but cannot reliably perform the full code-review task end-to-end without expert human oversight.
Task automatabilityclaude-sonnet-53/5AI can parse blueprints and check against code-compliance rules using computer vision and NLP, but final determinations require domain judgment, site context, and legal accountability that limit full automation today.
Adoption barriersclaude-haiku-4-5-202510014/5Fire safety code compliance is typically mandated by law and often requires sign-off by a licensed fire inspector or official. Regulatory liability, public safety responsibility, and jurisdictional legal requirements for human certification create substantial legal barriers to full automation.
Adoption barriersclaude-sonnet-54/5Fire code compliance sign-off typically requires a licensed/certified fire inspector or AHJ approval, creating a significant legal and liability barrier to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI document processing and code-checking tools are relatively inexpensive per instance, but the task's safety criticality demands high-quality human oversight and re-verification, offsetting any per-task savings. The total cost (AI + required human review + liability) remains comparable to or exceeds direct human review.
Cost vs. human wageclaude-sonnet-53/5Automated plan-checking software can reduce review time substantially, but licensing, integration, and required human verification keep costs roughly comparable to inspector labor once oversight is included.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools exist for blueprint analysis and rule-checking (e.g., automated floor-plan extraction, basic compliance flagging), but production systems are narrowly scoped and still generate material false negatives/positives. No deployed product reliably replaces the human inspector's judgment on fire safety compliance at scale in actual regulatory workflows.
Technical feasibility todayclaude-sonnet-52/5Some AEC-tech products (e.g., automated code-checking software like UpCodes, Clara) exist and are used in permitting workflows, but they are narrow-scope, error-prone on complex fire-specific codes, and not yet standard in fire inspector workflows.

Teach public education programs on fire safety and prevention.

26

CI 18–35 · exposure 20 · augmentation 75 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire departments and inspection agencies are traditionally low-digitization, conservative sectors with strong human accountability requirements. Adoption of AI for public-facing safety education remains minimal and pilot-stage.
Sector adoption velocityclaude-sonnet-52/5Public sector fire departments and inspection agencies are typically slow adopters of AI tools, with most current use limited to administrative or content support rather than program delivery.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can significantly assist inspectors by drafting educational materials, generating presentation content, personalizing messaging for different audiences, and creating multimedia resources—enabling inspectors to expand reach and customize programs while they remain the credible source and delivery mechanism.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully help inspectors draft educational materials, create presentations, generate multilingual content, and design engaging activities, improving efficiency while the human still delivers the program.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can generate fire safety content and instructional materials, the core task requires real-time interaction, audience engagement, and adaptive response to questions—capabilities that current AI systems cannot reliably deliver in live educational settings. Automating the full public education function to a 50% time-saving threshold with equal quality is not feasible today.
Task automatabilityclaude-sonnet-52/5AI can help generate content and materials for fire safety education, but live public teaching, audience engagement, and adapting to community-specific concerns require in-person human delivery that current AI cannot fully replace.
Adoption barriersclaude-haiku-4-5-202510014/5Fire inspectors are often required by law or regulation to conduct public education, and these programs require credentialed personnel for liability and legal authority. Public trust in fire safety information is high, creating organizational and regulatory friction against full substitution with AI.
Adoption barriersclaude-sonnet-52/5No licensing requirement mandates a human deliver this specifically, but public trust, in-person engagement expectations, and government/community relations create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5While AI-generated educational materials have low marginal cost, the task's value derives from credentialed, authoritative instruction with liability and accountability. The all-in cost of reliable AI delivery (including oversight, regulatory compliance, and human backup) does not yet undercut the loaded wage of a fire inspector.
Cost vs. human wageclaude-sonnet-52/5While AI can cheaply produce slides, scripts, or handouts, the actual delivery (school visits, community talks) still requires a paid human presenter, so overall cost savings are modest.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs live public fire safety education at the quality and interaction level required for this occupational task. AI chatbots and content generators exist but do not substitute for in-person or real-time educational delivery by trained inspectors.
Technical feasibility todayclaude-sonnet-52/5There are no deployed products that autonomously deliver public fire safety education programs in communities; AI is at best a content-creation aid used by human instructors.

Identify corrective actions necessary to bring properties into compliance with applicable fire codes, laws, regulations, and standards, and explain these measures to property owners or their representatives.

25

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire departments and building inspection agencies are traditionally risk-averse and slow-moving. Adoption of AI assistive tools is in early pilot phases; most jurisdictions still rely on manual inspections and written reports, with no evidence of significant displacement in production environments.
Sector adoption velocityclaude-sonnet-52/5Fire safety and code enforcement occurs mostly within municipal government and physical inspection contexts, sectors with historically slow AI adoption and limited digitization of the inspection workflow.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can helpfully draft preliminary violation lists from photos, suggest standard corrective measures from code databases, and assist in documentation and reporting. However, the nuanced explanation and negotiation with property owners remains largely human-driven, limiting augmentation to moderate productivity gains in the discovery and documentation phases.
Augmentation potentialclaude-sonnet-53/5AI can meaningfully assist by retrieving relevant code sections, drafting explanatory reports, and organizing violation documentation, improving inspector efficiency even though the core assessment remains human-led.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in identifying some standard code violations through image analysis and rule-matching, the task requires integrating complex contextual judgment about building-specific factors, local code variations, and feasibility trade-offs. End-to-end automation would fall short of the 50% time-saving threshold because the explanation and negotiation of corrective measures to property owners demands human expertise and accountability.
Task automatabilityclaude-sonnet-52/5AI can help draft code citations and suggest corrective actions from inspection notes, but identifying the specific violations requires on-site judgment and physical inspection that current AI cannot perform autonomously.'
Adoption barriersclaude-haiku-4-5-202510014/5Fire inspection and code compliance carry high liability exposure; errors directly affect public safety and building occupancy permits. Most jurisdictions require a licensed fire inspector or qualified professional to legally certify compliance and sign off on corrective measures, creating a hard regulatory barrier to full automation.
Adoption barriersclaude-sonnet-54/5Fire inspection and code enforcement typically require a certified/licensed inspector, government authority, and legal accountability for citations, creating strong regulatory and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI tooling for code compliance analysis requires significant setup, domain customization, and human expert oversight to validate outputs. The all-in cost (models, integration, mandatory human review) remains comparable to or higher than direct human fire inspector labor for reliable compliance work.
Cost vs. human wageclaude-sonnet-52/5Human inspectors' wages are moderate, but AI cannot yet substitute for the on-site judgment and liability-bearing sign-off, so full cost savings are not realized despite cheap inference for supporting research.
Technical feasibility todayclaude-haiku-4-5-202510012/5No production systems reliably perform this task end-to-end. Image recognition tools can flag potential hazards, and rule-checking engines exist in niche domains, but deployed products do not consistently identify compliant corrective actions across diverse properties and code jurisdictions at reliability standards needed for fire safety.
Technical feasibility todayclaude-sonnet-52/5There are no deployed products that reliably conduct fire code compliance determinations and communicate remediation to property owners; existing tools are limited to checklist support or document search.

Develop and coordinate fire prevention programs, such as false alarm billing, fire inspection reporting, and hazardous materials management.

25

CI 25–25 · exposure 25 · augmentation 50 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire departments and municipal agencies adopt digitization slowly; most remain in pilot phases for data management tools. Program development remains largely manual and expert-driven, with limited production-grade AI integration in the sector.
Sector adoption velocityclaude-sonnet-52/5Public sector fire departments and municipal safety services are typically slow adopters of AI tools, with limited production deployment beyond basic recordkeeping software.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist with drafting inspection reports, organizing hazmat databases, and flagging compliance gaps, raising efficiency for the fire inspector performing the work, though the high-level coordination and policy judgment remain human-centered.
Augmentation potentialclaude-sonnet-53/5AI can meaningfully assist by drafting reports, organizing billing data, tracking hazardous materials inventories, and generating compliance checklists, improving efficiency while a human inspector retains responsibility.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with hazardous materials data processing and generate inspection reports from structured inputs, developing and coordinating fire prevention programs requires strategic judgment, stakeholder management, and domain expertise that current systems cannot fully automate. The coordination and policy-development aspects remain firmly in human territory.
Task automatabilityclaude-sonnet-52/5AI can assist with drafting policy documents, tracking billing data, and organizing reports, but developing and coordinating an actual multi-stakeholder prevention program requires site knowledge, regulatory judgment, and inter-agency coordination that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Fire prevention program development and coordination must comply with state/local fire codes, NFPA standards, and often requires sign-off from licensed fire officials. Liability and public safety requirements mean human oversight and authorization are legally mandated, creating hard barriers to automation.
Adoption barriersclaude-sonnet-54/5Fire inspection and hazardous materials oversight are governed by codes and often require certified/licensed fire inspectors to sign off, creating significant regulatory and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted components (report generation, data management) offer modest cost savings, but the bulk of program development and coordination requires skilled human fire inspectors and administrators, making all-in costs comparable to or exceeding the human alternative.
Cost vs. human wageclaude-sonnet-52/5AI could cheaply support documentation and data management portions, but the coordination, inspection, and compliance judgment components still require paid inspector time, keeping overall cost comparable to human-led work.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some deployed products exist for fire inspection documentation and hazmat tracking, but no end-to-end AI system reliably handles program development and coordination across the legal, regulatory, and operational complexities involved. Current tools are narrow adjuncts rather than system replacements.
Technical feasibility todayclaude-sonnet-52/5No deployed product manages fire prevention program coordination as a whole; existing tools address narrow slices like records management or billing software, not the integrated task.

Develop or review fire exit plans.

21

CI 18–25 · exposure 25 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire departments and inspection agencies are laggard sectors in AI adoption; the highly regulated, liability-sensitive nature of fire safety and the requirement for licensed professional involvement slow any AI-driven automation.
Sector adoption velocityclaude-sonnet-52/5Fire safety and building inspection is a traditionally low-digitization, government/public-safety sector with slow AI adoption compared to information or finance sectors.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could usefully assist inspectors by automatically flagging common code violations, comparing plans to building blueprints, or extracting relevant sections of fire codes, but the human expert must retain full authority over plan adequacy and sign-off.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully assist by checking code compliance, generating draft layouts, and flagging potential issues, improving inspector efficiency while the human remains responsible for final judgment.
Task automatabilityclaude-haiku-4-5-202510012/5Reviewing and analyzing existing fire exit plans for compliance could be partially automated using computer vision and checklist matching, but developing plans requires contextual judgment about building layout, occupancy codes, and safety trade-offs that current AI cannot reliably handle end-to-end at 50% time savings.
Task automatabilityclaude-sonnet-52/5AI can help draft or check exit plan layouts against code text, but verifying physical building conditions, egress widths, and site-specific hazards requires human on-site judgment that current systems cannot perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Fire exit plans are legally required to be developed or certified by licensed fire safety professionals or engineers; regulatory frameworks and liability for inadequate egress planning create hard barriers to full automation without human sign-off.
Adoption barriersclaude-sonnet-54/5Fire exit plans typically require review/approval by licensed fire inspectors or code officials under regulatory and life-safety liability frameworks, creating strong legal and professional barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for plan analysis and code-checking might reduce labor by 20–30%, but the integration cost, required oversight, and liability exposure make the all-in cost comparable to or potentially higher than direct human inspection and plan development.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply generate draft layouts or flag code issues, but the human inspection, sign-off, and liability review still dominate the cost, keeping overall savings modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5No mature product reliably develops or reviews fire exit plans independently; while AI can assist with document analysis and code lookup, the task requires expert judgment and professional liability that demands human oversight, limiting deployable automation.
Technical feasibility todayclaude-sonnet-52/5Some CAD/code-compliance software and AI-assisted plan review tools exist, but no deployed product autonomously develops or certifies fire exit plans at scale in production.

Photograph damage and evidence related to causes of fires or explosions to document investigation findings.

19

CI 14–25 · exposure 20 · augmentation 50 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire departments and investigation agencies operate in traditionally low-digitization public-sector environments with slow technology adoption; most still rely on manual photography by human investigators rather than autonomous systems.
Sector adoption velocityclaude-sonnet-51/5Fire investigation is a physical, safety-critical field profession with low digitization and minimal AI agent deployment in production settings.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted features like automated damage detection, recommendation of shot angles, or post-processing to enhance evidence visibility could meaningfully assist inspectors, though the core task of capturing legally admissible evidence remains human-led.
Augmentation potentialclaude-sonnet-53/5AI can assist with organizing, tagging, and enhancing photos, or flagging potential burn patterns, but the core evidence-gathering and photographic judgment remain human-driven.
Task automatabilityclaude-haiku-4-5-202510012/5While AI vision systems can capture and organize photographs automatically, fire investigation requires human judgment to identify legally and technically relevant damage/evidence, frame shots appropriately, and ensure chain-of-custody documentation—tasks that cannot be reliably automated end-to-end today.
Task automatabilityclaude-sonnet-52/5Physical presence at fire/explosion scenes is required to identify, capture, and contextualize evidence, which current AI cannot perform end-to-end; AI can only assist with post-hoc image organization or analysis.5D drones or cameras don't replace the investigative judgment needed on-site.
Adoption barriersclaude-haiku-4-5-202510014/5Fire investigation photography is often required as evidence in legal proceedings; chain-of-custody and admissibility standards typically mandate documentation by a licensed/trained investigator, creating a legal barrier to full automation.
Adoption barriersclaude-sonnet-54/5Evidence documentation for fire investigations often must meet legal chain-of-custody and forensic standards, typically requiring a certified investigator to personally observe, photograph, and attest to conditions for court admissibility.
Cost vs. human wageclaude-haiku-4-5-202510012/5The cost of deploying reliable vision systems, integrating them with forensic workflows, and providing necessary human oversight approaches or exceeds the cost of a trained inspector taking photographs on-site.
Cost vs. human wageclaude-sonnet-51/5There is no AI system substituting for the human presence, judgment, and physical evidence handling required, so no meaningful cost comparison favors AI today.
Technical feasibility todayclaude-haiku-4-5-202510012/5General-purpose camera systems and computer vision exist, but no production product reliably performs fire investigation photography with the legal and forensic standards required; automated systems lack the contextual decision-making about what constitutes material evidence.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously conducts scene photography and evidentiary documentation for fire investigations; this remains a human field task with at most camera/drone hardware assistance.

Conduct inspections and acceptance testing of newly installed fire protection systems.

17

CI 0–35 · exposure 28 · augmentation 38 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection is a heavily regulated, government-led function with minimal digitization; adoption of AI in this sector remains negligible, with inspectors still relying on manual procedures and paper-based documentation.
Sector adoption velocityclaude-sonnet-51/5Fire inspection is a highly physical, regulated public-safety function with minimal digitization or AI agent deployment in the field.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist inspectors by pre-analyzing blueprints, flagging code violations, and organizing inspection checklists, meaningfully reducing preparation time and field review burden while the licensed inspector retains authority and decision-making.
Augmentation potentialclaude-sonnet-52/5AI can help with documentation, scheduling, and referencing code requirements, but offers little assistance during the actual physical inspection and testing process.
Task automatabilityclaude-haiku-4-5-202510014/5AI can autonomously inspect visual documentation, analyze system configurations against code standards, and generate compliance reports with high accuracy. However, some physical testing (pressure gauges, water flow verification) still requires human presence, limiting end-to-end automation to ~70-80% time savings.
Task automatabilityclaude-sonnet-51/5This requires physical presence to visually inspect installed sprinkler heads, alarms, pumps, and wiring, and to conduct hands-on acceptance testing (e.g., flow tests) that AI cannot perform without robotic embodiment.
Adoption barriersclaude-haiku-4-5-202510015/5Fire protection system acceptance testing is legally mandated and typically requires a licensed Fire Protection Engineer or certified inspector to formally approve systems; regulatory frameworks explicitly require qualified human sign-off, creating a hard legal barrier to full automation.
Adoption barriersclaude-sonnet-55/5Fire code compliance inspections and acceptance testing typically require a licensed/certified fire inspector's sign-off, with legal liability tied to human certification of life-safety systems.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI system costs (computer vision infrastructure, regulatory compliance oversight, human validation) remain comparable to or exceed the cost of human inspectors, especially when liability and sign-off requirements are factored in.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical inspection labor, so there is no viable AI cost basis to compare against the human inspector's wage for this task.
Technical feasibility todayclaude-haiku-4-5-202510012/5While computer vision and code-matching algorithms exist in research and pilot projects, no mature, production-deployed system reliably conducts full fire protection system inspections without substantial human oversight and verification.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs physical fire protection system inspections and acceptance testing; this remains a manual, certified-inspector task in practice.

Recommend changes to fire prevention, inspection, and fire code endorsement procedures.

16

CI 13–20 · exposure 16 · augmentation 63 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection and prevention operate in heavily regulated, slow-moving government and municipal sectors with strong human expertise requirements and low digital transformation velocity.
Sector adoption velocityclaude-sonnet-52/5Fire safety and municipal inspection sectors are slow adopters of AI tools relative to information/finance sectors, with most AI use limited to pilots or administrative support.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by summarizing fire incident data, identifying code gaps, and drafting analysis sections, raising inspector productivity in research and initial drafting phases while the human expert makes final judgments.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by analyzing incident trends, benchmarking code standards across jurisdictions, and drafting policy language for human inspectors to refine and finalize.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with data analysis of fire patterns and code gaps, but the task fundamentally requires human judgment about regulatory changes, stakeholder concerns, and practical implementation. Current systems cannot reliably generate credible policy recommendations end-to-end.
Task automatabilityclaude-sonnet-52/5This requires synthesizing field experience, incident data, and regulatory knowledge into judgment-based policy recommendations, which current AI can support but not autonomously perform end-to-end at equal quality.
Adoption barriersclaude-haiku-4-5-202510015/5Fire code changes require regulatory authority, professional licensing, and legal accountability. A human fire inspector or official must legally author and endorse the recommendations, creating a hard barrier to full automation.
Adoption barriersclaude-sonnet-54/5Fire code changes typically require authorized officials, public comment processes, and legal/regulatory approval, creating strong institutional and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI could reduce research and analysis time moderately, but human expert review, stakeholder consultation, and final recommendation formulation remain necessary. Cost savings are partial and offset by oversight requirements.
Cost vs. human wageclaude-sonnet-52/5AI could cheaply draft analysis or summaries, but the overall task still requires expert review, stakeholder engagement, and legal vetting, keeping costs comparable to human-led processes.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs fire code recommendation or policy change endorsement. This requires institutional authority, liability accountability, and domain expertise that AI systems lack in production settings.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously generates fire code and inspection procedure recommendations for jurisdictional adoption; this remains a human expert and committee-driven process.

Test sites and materials to establish facts, such as burn patterns and flash points of materials, using test equipment.

15

CI 5–25 · exposure 13 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire departments and investigation agencies are typically public sector, slow-moving organizations with entrenched processes and regulatory requirements. Adoption of AI tools for analysis is emerging slowly, but replacement of core testing activities remains rare.
Sector adoption velocityclaude-sonnet-51/5Fire investigation is a physical, safety-critical field with low digitization and slow AI adoption; this is a laggard sector for automation of on-site physical testing.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist investigators by analyzing test data, identifying burn patterns from imagery, and cross-referencing material properties against databases, allowing investigators to focus on interpretation and conclusions. However, the core testing itself still requires human judgment and physical execution.
Augmentation potentialclaude-sonnet-53/5AI can assist with analyzing test equipment data, pattern recognition in burn photographs, or drafting reports based on collected data, but the physical testing itself is unassisted by AI.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can analyze images of burn patterns and data from test equipment, the hands-on physical testing and material handling required to establish facts through direct experimentation cannot be automated by current AI systems. Physical presence and equipment operation remain essential to the task.
Task automatabilityclaude-sonnet-51/5This requires physical presence at fire scenes, hands-on manipulation of test equipment (accelerant detectors, flash point testers), and physical sample collection—none of which current AI systems can perform.rating reflects zero end-to-end automation potential today.rationale focuses on physical/manual nature.
Adoption barriersclaude-haiku-4-5-202510014/5Fire investigation involves legal authority, liability, and the requirement that certified investigators conduct and sign off on official test results and findings. Regulatory requirements and liability concerns create strong barriers to full automation of testing and fact-establishment.
Adoption barriersclaude-sonnet-54/5Fire investigation often requires certified/licensed investigators whose findings may be used in legal proceedings (arson, insurance, criminal cases), creating strong credentialing and liability requirements alongside chain-of-custody rules for evidence.
Cost vs. human wageclaude-haiku-4-5-202510012/5The equipment, sensors, and human oversight required for AI-assisted testing approaches comparable cost to a trained fire investigator, while the investigator brings legal authority and judgment that AI cannot replace. Net cost savings remain marginal.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical task, so cost comparison favors the human investigator by default since AI cannot execute the core physical actions at all.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI systems can assist with post-hoc analysis of test results and burn pattern recognition from images, but no deployed products reliably perform the full end-to-end testing and investigation autonomously. Current AI lacks embodied capability to handle materials, operate specialized fire testing equipment, or conduct live tests.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical fire scene testing or material sampling; this remains entirely a human forensic activity requiring physical tools and on-site presence.

Teach fire investigation techniques to other firefighter personnel.

15

CI 14–16 · exposure 9 · augmentation 50 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire departments are traditionally conservative, decentralized, and bound by civil service rules and regulatory mandates for instructor qualifications. Adoption of AI for core safety training has been minimal, and cultural norms strongly favor experienced human mentorship in life-safety domains.
Sector adoption velocityclaude-sonnet-52/5Fire departments and public safety training programs are generally slow adopters of AI-driven instruction due to institutional, budget, and cultural factors.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by generating training materials, simulating scenarios, creating quizzes, and organizing curriculum content, thereby reducing preparation burden on instructors. However, the human instructor remains essential for live mentoring, judgment calls, and credentialing.
Augmentation potentialclaude-sonnet-53/5AI can help create training curricula, case study materials, and quizzes, and support instructors, but does not replace the interactive, experiential nature of teaching investigation skills.
Task automatabilityclaude-haiku-4-5-202510011/5Teaching fire investigation techniques requires adaptive instruction, real-time feedback, assessment of learner understanding, and contextual knowledge-sharing grounded in hands-on demonstrations. Current AI cannot reliably deliver the personalized pedagogical adjustment and mentoring accountability this task demands.
Task automatabilityclaude-sonnet-51/5This requires live instruction, hands-on demonstration, and judgment-based mentoring of investigative techniques that AI cannot deliver end-to-end today.
Adoption barriersclaude-haiku-4-5-202510014/5Fire safety training is subject to regulatory standards and certification requirements (NFPA, state/local fire codes) that typically mandate qualified human instructors to validate competency and sign off on personnel readiness. Legal liability for inadequate fire investigation training creates hard barriers to full automation.
Adoption barriersclaude-sonnet-54/5Fire investigation training often requires certified instructors with recognized expertise and hands-on practical exercises, creating professional and credentialing barriers to full AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510012/5Creating and maintaining effective instructional content with AI assistance costs less than senior inspectors' loaded wages, but oversight and refreshing of safety-critical training material, plus the human instructor still needed to guide practice, keeps the total cost-per-outcome relatively high.
Cost vs. human wageclaude-sonnet-52/5Developing AI-based training content has some cost savings potential, but human instructor expertise, credibility, and hands-on demonstration are still required, limiting overall cost reduction.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can generate training materials and deliver static content, no deployed product reliably performs interactive fire investigation instruction with the quality, safety-critical judgment, and experiential learning that actual firefighter training requires.
Technical feasibility todayclaude-sonnet-52/5AI tools can generate training materials or supplement e-learning modules, but no deployed product independently teaches fire investigation skills to firefighters in a classroom or field-training context.

Arrange for the replacement of defective fire fighting equipment and for repair of fire alarm and sprinkler systems, making minor repairs such as servicing fire extinguishers when feasible.

15

CI 5–25 · exposure 13 · augmentation 50 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire departments and building facilities management organizations have adopted scheduling and tracking software, but automation of the physical inspection and repair arrangement process remains limited. Adoption is slower in the public-sector fire services where this task is concentrated.
Sector adoption velocityclaude-sonnet-51/5Fire inspection and physical facilities maintenance is a low-digitization, physical-labor sector with minimal AI agent adoption in production.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could meaningfully assist inspectors through automated scheduling of replacements, alerting on overdue maintenance, and generating compliance reports. However, the core work of physical assessment and repair coordination remains human-driven, making augmentation useful but not transformative.
Augmentation potentialclaude-sonnet-53/5AI can help by tracking maintenance schedules, generating work orders, and flagging defective equipment from inspection data, but the core arranging/repair work remains human-driven.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could assist with scheduling and documentation of equipment repairs, the physical act of servicing fire extinguishers, inspecting equipment condition, and arranging repairs requires hands-on assessment and human coordination. Current AI cannot perform the tactile inspection and minor repairs component of this task.
Task automatabilityclaude-sonnet-51/5This task requires physical inspection, hands-on equipment servicing, and coordination with vendors/contractors in the physical world, none of which current AI can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Fire safety compliance is heavily regulated, and equipment replacement and system repairs typically require licensed fire safety professionals to verify code compliance and sign off on work. Legal liability for failed fire suppression systems creates strong requirements for human oversight and accountability.
Adoption barriersclaude-sonnet-54/5Fire safety equipment servicing often requires certified technicians and code compliance, and liability for faulty repairs on life-safety systems is high, creating strong regulatory and professional barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI systems for maintenance scheduling are relatively inexpensive, but they do not replace the labor-intensive work of physical inspection, coordination with vendors, and hands-on servicing. The total cost savings compared to a human fire inspector remains modest.
Cost vs. human wageclaude-sonnet-51/5AI cannot perform the physical repair or servicing components, so a human (or technician) is required regardless of cost, making AI substitution infeasible rather than merely expensive.
Technical feasibility todayclaude-haiku-4-5-202510012/5Scheduling and tracking systems exist for maintenance workflows, but no deployed product reliably performs the core task of assessing defective equipment, arranging replacements, and executing minor repairs autonomously. Human inspectors remain essential to evaluate equipment condition and authorize repairs.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical repair scheduling, equipment servicing, or hands-on maintenance of fire safety systems today.

Conduct fire exit drills to monitor and evaluate evacuation procedures.

14

CI 5–23 · exposure 13 · augmentation 25 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection remains a traditional, heavily regulated sector with low automation adoption. Organizations continue to rely on in-person inspectors, and regulatory frameworks do not encourage or permit replacement of live drill oversight with AI systems.
Sector adoption velocityclaude-sonnet-51/5Fire inspection and safety compliance is a low-digitization, physically-grounded sector with minimal AI agent deployment for in-person procedural oversight.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by providing video recording, timing analysis, or post-drill reporting, but current systems offer only marginal productivity gains. The core task—observing occupant behavior, making judgment calls, and ensuring compliance—remains predominantly human-driven work.
Augmentation potentialclaude-sonnet-52/5AI could help schedule drills, analyze evacuation time data, or generate reports afterward, but it offers little assistance during the live monitoring and evaluation itself.
Task automatabilityclaude-haiku-4-5-202510012/5While AI systems could monitor video feeds or compile evacuation timing data, the task fundamentally requires physical presence on-site, real-time judgment of occupant behavior, and safety oversight that cannot be fully automated. Current AI cannot evaluate nuanced evacuation compliance or identify emerging hazards during live drills to meet the 50% time-saving threshold.
Task automatabilityclaude-sonnet-51/5Conducting a fire drill requires physical presence to coordinate a live evacuation, observe human behavior, and assess building egress in real time, none of which current AI systems can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Fire safety regulations typically mandate that a licensed fire inspector directly conduct and sign off on evacuation drills. Liability, safety standards, and legal requirements place hard constraints on delegation to automated systems; human oversight is legally required in most jurisdictions.
Adoption barriersclaude-sonnet-54/5Fire safety drills are often mandated by code and require an authorized inspector to physically oversee and certify compliance, creating strong regulatory and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI monitoring systems (cameras, sensors) require significant infrastructure and integration costs, plus human oversight for safety. A single fire inspector conducting a drill typically costs less than the total deployment and operational cost of AI monitoring systems that would still require substantial human backup.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical coordination task, so AI cost is not comparable—human labor is the only viable option today.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably conducts fire exit drills end-to-end. Computer vision systems can passively monitor some aspects (counting occupants, timing), but they cannot direct drills, assess procedural compliance, or make real-time safety decisions as required. Deployed systems lack the autonomy and judgment needed.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product runs or manages physical evacuation drills; this remains a purely human, on-site activity.

Analyze evidence and other information to determine probable cause of fire or explosion.

13

CI 0–25 · exposure 13 · augmentation 50 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire investigation remains a specialized, regulated sector with slow digital transformation; most agencies are small, public-sector entities with limited procurement agility, and adoption of AI for core investigative conclusions is minimal outside small pilot programs.
Sector adoption velocityclaude-sonnet-51/5Fire investigation is a physical, low-digitization field service function with minimal AI agent deployment; adoption is largely limited to peripheral tools like report drafting, not the core investigative task.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by automating evidence logging, flagging pattern matches in historical databases, and highlighting correlations in multimodal data (images, chemical reports, witness timelines), but the investigator must still synthesize and validate findings for legal defensibility.
Augmentation potentialclaude-sonnet-53/5AI can assist with organizing evidence, cross-referencing historical fire patterns, and drafting reports, providing meaningful support to investigators without performing the core causal analysis itself.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in processing and correlating evidence data (burn patterns, chemical composition, witness statements), determining probable cause requires integration of complex spatial reasoning, contextual judgment, and legal-chain-of-custody standards that current systems cannot reliably execute end-to-end. Manual investigator involvement remains essential for <50% time savings at equal quality.
Task automatabilityclaude-sonnet-51/5Determining probable cause requires physical scene examination, chain-of-custody evidence handling, and expert judgment integrating burn patterns, witness statements, and forensic science that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Legal liability is high—fire-cause determinations directly influence prosecution, insurance claims, and policy; regulatory frameworks (NFPA 921, criminal procedure) typically require a licensed fire investigator's professional endorsement, and errors carry asymmetric legal risk that deters full automation.
Adoption barriersclaude-sonnet-55/5Fire cause determinations often feed into legal proceedings, insurance claims, and criminal investigations requiring certified investigators, expert witness testimony, and legal accountability that only licensed humans can provide.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI systems for evidence analysis require significant human oversight, domain-expert validation, and integration costs; the loaded cost of the human investigator plus all-in AI tooling remains comparable or exceeds the investigator's hourly rate for complex cases.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for on-site physical investigation and expert testimony, so there is no viable AI-only cost comparison—human investigators remain necessary regardless of AI tool costs.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs comprehensive fire-cause determination. AI tools exist for narrow subtasks (image analysis of burn patterns, accelerant detection), but no production system integrates evidence holistically to reach defensible probable-cause conclusions in the forensic/legal sense required.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously conducts fire cause determination; this remains a specialized forensic investigative function performed by trained human investigators in the field.

Inspect buildings to locate hazardous conditions and fire code violations, such as accumulations of combustible material, electrical wiring problems, and inadequate or non-functional fire exits.

10

CI 0–20 · exposure 13 · augmentation 50 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Fire inspection remains a heavily regulated, government-run or government-contracted function with slow technology adoption. Most inspections are still conducted manually by human fire marshals; AI pilot projects exist but are rare and have not displaced inspectors at scale in production fire marshal offices.
Sector adoption velocityclaude-sonnet-51/5Public sector fire safety inspection is a low-digitization, physical-presence-dependent field with minimal AI agent deployment in production.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist inspectors by pre-screening imagery, highlighting potential hazards, organizing inspection checklists, or analyzing drone footage to flag items for closer human examination. These tools raise inspector productivity on documentation and prioritization but still require substantial human judgment and verification on-site.
Augmentation potentialclaude-sonnet-53/5AI can assist with checklist generation, report writing, code lookup, and even computer-vision-based hazard flagging from photos, improving inspector efficiency on parts of the task.
Task automatabilityclaude-haiku-4-5-202510012/5While AI systems can analyze images and identify some visual hazards (e.g., blocked exits via computer vision), fire inspection requires real-time navigation of complex 3D environments, assessment of code compliance across dozens of interconnected systems (electrical, structural, ventilation), and professional judgment on context-dependent violations. Current systems cannot reliably substitute for on-site human inspection and reduce time by ≥50% while maintaining equal quality.
Task automatabilityclaude-sonnet-51/5This requires physical presence, walking through buildings, physically testing exits, wiring, and materials—no current AI system can perform physical inspection end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Fire inspectors are typically state-licensed professionals whose inspections carry legal force; most jurisdictions require a licensed fire inspector to conduct official inspections and certify compliance. Liability for missed hazards and building code authority requirements create hard legal and regulatory barriers to full automation without human sign-off.
Adoption barriersclaude-sonnet-55/5Fire inspections are typically legally mandated to be performed by certified/licensed inspectors with authority to issue citations, creating hard regulatory and liability barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5A human fire inspector's loaded wage is $50–70k annually for 2000+ hours of work. AI vision systems and drone deployments still require significant setup, oversight, and human verification; per-inspection costs remain comparable to or exceed outsourcing to junior inspectors, with no clear cost advantage.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical inspection itself, so there is no viable cost comparison—the human must still be on-site performing the core task.
Technical feasibility todayclaude-haiku-4-5-202510012/5Deployed products exist for narrow subtasks (e.g., image classification of fire hazards, automated floor plans), but no production system reliably performs comprehensive fire code inspection autonomously. Pilot programs using computer vision or drones exist, but human inspectors must still verify findings and make final determinations, making end-to-end automation infeasible at scale today.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously conducts physical fire code inspections; at most, AI assists with report drafting or image analysis for narrow sub-tasks.

Instruct children about the dangers of fire.

9

CI 5–14 · exposure 5 · augmentation 50 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire safety instruction for children remains a core face-to-face service delivered by public safety professionals; AI adoption in this particular task is minimal and limited to supplementary content, not replacement of the instructional relationship.
Sector adoption velocityclaude-sonnet-51/5Fire safety education for children is a low-digitization, community-outreach function within public safety agencies, showing minimal AI adoption for the interactive teaching component.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist fire inspectors by generating lesson plans, creating illustrated safety materials, or producing interactive video content that an inspector then uses in class, thereby improving preparation and engagement without removing the human instructor from the central role.
Augmentation potentialclaude-sonnet-53/5AI can help generate lesson plans, presentation materials, quizzes, and visual aids to support inspectors preparing fire safety talks for children, offering moderate productivity gains in preparation, though not during the live instruction itself.
Task automatabilityclaude-haiku-4-5-202510011/5Instructing children about fire dangers fundamentally requires live human interaction, emotional engagement, and the ability to respond to individual questions and concerns in real time. Current AI systems cannot meaningfully replace this interpersonal teaching relationship, even with significant time investment in setup.
Task automatabilityclaude-sonnet-51/5This task involves live, in-person engagement with children, requiring physical presence, real-time adaptation, and trust-building that current AI cannot replicate end-to-end.dictionary AI could assist with materials but not perform the actual instruction.rating reflects that no meaningful automation of the live delivery exists today.
Adoption barriersclaude-haiku-4-5-202510014/5Schools and public safety institutions have strong preferences for qualified human instructors (fire inspectors) delivering safety education to children, and parents typically expect in-person engagement. Liability concerns and institutional norms create significant friction against full automation.
Adoption barriersclaude-sonnet-54/5Working with children in safety contexts typically requires background-checked, trained personnel (often uniformed fire officials) for trust, liability, and child-safety reasons, creating strong organizational and social barriers to AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-generated content creation is cheap, but integrating it into a live instructional program for children, including monitoring engagement and managing classroom dynamics, still requires substantial human oversight that approaches or exceeds the cost of direct human instruction.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI-only substitute for in-person instruction of children, so cost comparison favors the human since AI cannot deliver the equivalent output at all.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can generate educational content about fire safety, no deployed product reliably delivers interactive, age-appropriate fire safety instruction to groups of children in real-world school or community settings. AI chatbots and video tools exist but are not standard teaching replacements in this domain.
Technical feasibility todayclaude-sonnet-51/5No deployed product independently instructs children in person about fire safety; this remains a human-led community education activity with no production AI substitute.

Supervise staff, training them, planning their work, and evaluating their performance.

9

CI 3–16 · exposure 8 · augmentation 50 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While some fire departments use scheduling and performance tracking software, actual supervisory decision-making remains human-performed. Adoption of AI in this specific supervisory role is minimal due to legal, liability, and organizational constraints, not demand.
Sector adoption velocityclaude-sonnet-52/5Fire inspection agencies are largely public-sector and low-digitization environments where AI adoption for management functions is minimal and slow.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist with administrative components: generating training schedules, aggregating performance data, flagging anomalies in attendance or metrics. However, the human supervisor must remain central to coaching, evaluation, and interpersonal decisions, offering moderate augmentation rather than transformation.
Augmentation potentialclaude-sonnet-53/5AI tools can help draft training materials, schedules, and performance review documentation, offering moderate assistance while the human retains full supervisory responsibility.
Task automatabilityclaude-haiku-4-5-202510012/5Staff supervision and performance evaluation require contextual judgment about individual capabilities, interpersonal dynamics, and organizational goals. While AI can assist with scheduling and documenting metrics, the core supervisory responsibilities—coaching, feedback, conflict resolution—demand human judgment and authority that current systems cannot reliably replicate end-to-end.
Task automatabilityclaude-sonnet-51/5Supervising staff, planning work, and evaluating performance requires interpersonal judgment, accountability, and contextual decision-making that current AI cannot perform end-to-end; at best AI assists with scheduling or drafting evaluation text.
Adoption barriersclaude-haiku-4-5-202510015/5Supervisory authority, performance evaluation, and hiring/firing decisions carry legal liability and are typically reserved for licensed or credentialed management. Employment law and organizational structure require human accountability for staff decisions, creating hard barriers to full automation.
Adoption barriersclaude-sonnet-54/5Supervisory responsibility, performance evaluation, and disciplinary actions typically require accountable human management under labor law and organizational policy, creating strong structural barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI systems cannot replace the supervision function without human supervisors remaining accountable; any cost savings from partial administrative assistance (scheduling, record-keeping) are minimal compared to the necessity of retaining the human supervisor role.
Cost vs. human wageclaude-sonnet-51/5There is no functioning AI substitute for the supervisory role itself, so cost comparison favors the human by default since AI cannot deliver the output.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed commercial product reliably performs staff supervision, training planning, and performance evaluation autonomously. These tasks require legal employment authority, accountability, and nuanced decision-making that remain exclusively human-performed in production environments.
Technical feasibility todayclaude-sonnet-51/5No deployed product manages people, assigns work, and conducts performance evaluations autonomously in production settings; HR software provides tools, not substitution for the supervisory task.

Conduct fire code compliance follow-ups to ensure that corrective actions have been taken in cases where violations were found.

6

CI 0–13 · exposure 8 · augmentation 50 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection remains a physical, regulated, and localized government function with minimal digitization or automation adoption. The sector has been slow to deploy AI or automation due to legal requirements, safety criticality, and the need for human professional judgment and accountability.
Sector adoption velocityclaude-sonnet-51/5Fire inspection and public safety enforcement is a low-digitization, physically-grounded government/municipal function with minimal AI agent deployment in production for site verification.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could assist with pre-inspection tasks such as flagging high-risk sites, summarizing prior violations, scheduling, and post-inspection documentation, but the core inspection and compliance verification must remain under human control.
Augmentation potentialclaude-sonnet-53/5AI can assist with scheduling, generating inspection checklists, drafting compliance letters, and organizing violation records, improving efficiency around the core physical task.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could assist with document review and historical record checking, the core task requires physical site visits to verify corrective actions have been implemented, direct observation of compliance, and professional judgment that cannot be fully automated. AI cannot perform the in-person inspection component that is essential to this task.
Task automatabilityclaude-sonnet-51/5This requires physical, on-site inspection of premises to verify corrective actions were taken, which current AI systems cannot perform end-to-end.6 rely on physical presence and judgment.The task is inherently physical and cannot be automated by software alone.4 requires physical presence.The task is physical.The task is physical.The task is physical.The task requires a physical site visit.The task inherently requires a physical, in-person site visit which current AI cannot perform.There is no way to automate physical site inspection with current AI.The task cannot be automated end-to-end today.Rating reflects the physical, on-site nature of the follow-up inspection.The task requires physical presence and cannot be automated remotely today.The task requires a human physically present to verify compliance.There is minimal automatable portion beyond scheduling and documentation.The task requires physical verification which is not automatable.Overall rating is low due to physical inspection requirement.Rating 1.Final rating 1.Final answer 1.Final.Rating: 1.Rationale: The core of this task is a physical site visit to verify real-world conditions (e.g., cleared exits, installed extinguishers), which current AI systems cannot perform; only administrative sub-tasks like scheduling or record-keeping are automatable, far below the 50% threshold for the full task.
Adoption barriersclaude-haiku-4-5-202510015/5Fire code compliance inspections are regulated by law and typically require a licensed fire inspector to physically verify compliance and sign off on corrective actions. Liability for failures in fire safety is substantial, and jurisdictions mandate human professional authority over compliance determinations.
Adoption barriersclaude-sonnet-55/5Fire code enforcement typically requires a certified/licensed fire inspector to conduct and legally certify compliance verification, with liability and regulatory authority vested in a human officeholder.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI could potentially reduce some pre-visit administrative work (records review, scheduling), but the bulk of the task—physical inspection and professional determination—must still be performed by a human inspector. Overall cost savings remain minimal relative to the full human wage for this specialized role.
Cost vs. human wageclaude-sonnet-51/5Since AI cannot perform the physical verification, there is no comparable AI cost pathway; a human inspector remains necessary, making AI not cost-competitive for the core task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs end-to-end fire code compliance follow-up inspections. AI systems cannot conduct the required physical site visits, visual inspection of corrective measures, or make binding compliance determinations that have legal and safety implications.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product conducts physical fire code compliance verification visits; this remains a human inspector function performed in the field.

Issue permits for public assemblies.

6

CI 0–13 · exposure 8 · augmentation 38 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire departments and permit offices are bound by legal requirements for human authority in permit issuance. Adoption of AI for this task is negligible because the task cannot legally be delegated to a system; it remains firmly under human responsibility.
Sector adoption velocityclaude-sonnet-51/5Municipal fire safety and permitting is a slow-moving, highly regulated public-sector function with minimal AI agent deployment in production despite general digitization pilots.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by pre-populating permit forms, checking document completeness, or flagging common violations, but the core judgment—whether to approve a public assembly—must remain with the inspector. Augmentation potential is limited because the task is primarily discretionary authority rather than information processing.
Augmentation potentialclaude-sonnet-53/5AI can assist by pre-checking applications against code requirements, flagging occupancy issues, and drafting permit documents, improving inspector efficiency without replacing final judgment.
Task automatabilityclaude-haiku-4-5-202510011/5Issuing permits for public assemblies requires human judgment about safety codes, crowd capacity, emergency procedures, and site-specific conditions that current AI cannot reliably evaluate end-to-end. The task fundamentally depends on discretionary authority vested in a licensed inspector.
Task automatabilityclaude-sonnet-52/5Issuing permits involves reviewing application data against fire codes and occupancy limits, which AI could partially support, but final approval requires site-specific judgment and legal authority that current systems cannot fully replicate end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Public assembly permits are issued by licensed fire inspectors under legal authority; only a qualified human can legally approve and sign a permit. Regulatory frameworks explicitly require human inspection and human sign-off, creating a hard legal barrier to automation.
Adoption barriersclaude-sonnet-55/5Permit issuance is a government-authorized function tied to public safety liability, typically requiring a certified fire inspector or municipal authority to sign off, making this a hard regulatory barrier.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task is primarily human judgment and legal authorization; automating it would require replacing the inspector entirely, which is not feasible. The cost of building and maintaining a system to handle the liability and regulatory responsibility far exceeds the salary of a permit-issuing inspector.
Cost vs. human wageclaude-sonnet-52/5While document review software could reduce some administrative cost, the human inspection, liability review, and legal issuance steps still dominate the cost structure, keeping AI only marginally cheaper at best.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can independently issue legal permits for public assemblies; this requires human authorization and signature by a licensed official. While AI might assist in checklist generation or document processing, the permitting decision itself remains a human function in all jurisdictions.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously issues public assembly permits today; this remains a government function requiring human inspector sign-off with only limited digital workflow tools in use.

Examine fire sites and collect evidence such as glass, metal fragments, charred wood, and accelerant residue for use in determining the cause of a fire.

4

CI 0–9 · exposure 8 · augmentation 38 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire investigation remains a traditional, heavily regulated field with limited digitization and slow technology adoption. Most departments rely on manual inspection protocols, and autonomous systems face high barriers to entry due to liability and certification requirements.
Sector adoption velocityclaude-sonnet-51/5Fire investigation is a physical, on-site public safety function with minimal AI integration into the core evidence-collection process.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted tools for image analysis, evidence cataloging, and preliminary cause determination could meaningfully assist inspectors in organizing and interpreting collected evidence, but the physical collection itself remains human-dependent.
Augmentation potentialclaude-sonnet-52/5AI can assist with documentation, pattern analysis of burn patterns from photos, or report drafting after the fact, but offers little help during the actual physical evidence collection process.
Task automatabilityclaude-haiku-4-5-202510012/5Physical collection of fragile evidence from hazardous fire scenes requires dexterous manipulation, site-specific navigation, and chain-of-custody protocols that current robotics cannot reliably perform end-to-end. AI can assist in identifying and photographing evidence, but the actual collection task remains primarily manual.
Task automatabilityclaude-sonnet-51/5This requires physical presence at a hazardous, often unstable scene to identify, handle, and collect physical evidence, which no current AI system can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Chain-of-custody requirements, legal admissibility of evidence, and forensic standards mandated by law and courts require certified human investigators to physically collect, document, and testify about evidence. Liability and regulatory frameworks necessitate human sign-off and responsibility.
Adoption barriersclaude-sonnet-55/5Fire investigation for cause determination often requires certified/licensed investigators whose findings may be used in legal proceedings, insurance claims, or criminal prosecution, creating strong legal and evidentiary chain-of-custody requirements.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current specialized robotics and autonomous systems for hazardous environments are vastly more expensive than employing trained fire inspectors, and would require substantial infrastructure investment, oversight, and re-equipment per incident.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute for physical evidence collection, so AI cost is irrelevant/infinite relative to the human doing the work.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs fire scene evidence collection autonomously. Experimental robotics exist for hazardous environments, but none operate at production scale for forensic evidence gathering with the precision required for legal proceedings.
Technical feasibility todayclaude-sonnet-51/5No deployed product exists that can physically inspect a fire scene and collect evidence; this remains entirely a human field task.

Inspect properties that store, handle, and use hazardous materials to ensure compliance with laws, codes, and regulations, and issue hazardous materials permits to facilities found in compliance.

4

CI 0–9 · exposure 8 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection is a regulated public-sector function with entrenched licensing requirements and low digitization pressure. Adoption of automation is minimal because the role is tied to legal authority and liability that cannot be delegated to AI systems.
Sector adoption velocityclaude-sonnet-51/5Public sector fire safety and hazmat inspection is a low-digitization, physically grounded government function with minimal AI agent deployment in production.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could assist inspectors by automating compliance checklist generation, flagging code violations from photos, and organizing hazmat regulations, but the core judgment and on-site assessment remains human-driven. This would provide useful supplementary support without transforming the inspection process.
Augmentation potentialclaude-sonnet-53/5AI can assist with permit documentation, compliance checklist generation, code lookup, and report drafting, but the core on-site inspection and judgment call remain human-driven.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could assist with document review, code lookup, and compliance checklist generation, the task requires physical inspection of properties, assessment of dynamic conditions, and professional judgment about hazard severity that cannot be fully automated today. Current AI systems cannot reliably perform the on-site visual inspection and decision-making components needed to meet a 50% time-saving threshold.
Task automatabilityclaude-sonnet-51/5This requires physical, on-site inspection of facilities, hazardous materials handling processes, and equipment—AI cannot physically walk a site, verify storage conditions, or observe real-world compliance conditions today.
Adoption barriersclaude-haiku-4-5-202510015/5Fire inspection and hazmat permitting are legally restricted to licensed fire inspectors and investigators; permits issued by AI would not be valid. Regulatory frameworks explicitly require human professional certification and sign-off, creating hard legal barriers to substitution.
Adoption barriersclaude-sonnet-55/5Fire inspectors are licensed government officials with legal authority to issue permits and enforce fire and safety codes; this is a regulated, liability-heavy function requiring human certification and sign-off.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task requires specialized domain knowledge, legal liability assumption, and regulatory authority that only licensed inspectors can provide. Even with AI assistance, a qualified human must remain in the loop and liable, making the all-in cost roughly equivalent to or higher than human-only performance.
Cost vs. human wageclaude-sonnet-51/5AI cannot perform the physical inspection component at all, so there is no viable cost substitution—human inspectors remain necessary regardless of AI tool costs.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed products reliably perform end-to-end fire safety inspections or hazmat compliance assessments in production. While computer vision and document analysis exist as research tools, no operationalized system handles the full task of inspecting properties and issuing regulatory permits at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs autonomous on-site hazardous materials inspections or issues permits; this remains a human-led, government-authorized activity.

Inspect and test fire protection or fire detection systems to verify that such systems are installed in accordance with appropriate laws, codes, ordinances, regulations, and standards.

4

CI 0–9 · exposure 8 · augmentation 50 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection remains a traditional, heavily regulated, government and quasi-public sector function with strong legal and professional licensing requirements. Adoption of AI automation in this sector is minimal; digitization is limited mainly to record-keeping rather than the inspection task itself.
Sector adoption velocityclaude-sonnet-51/5Fire inspection is a physical, safety-critical field profession in the public/quasi-public sector with very low AI adoption and no evidence of production-scale automation.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could usefully assist by automating code lookups, flagging relevant regulations, analyzing inspection records for patterns, and generating compliance reports, but the human inspector remains essential for on-site decision-making and certification. These tools would improve efficiency without replacing the core inspection role.
Augmentation potentialclaude-sonnet-53/5AI can assist with generating checklists, interpreting code requirements, documenting findings, and flagging anomalies from sensor data or photos, improving inspector efficiency on the reporting and knowledge-lookup portions of the task.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could assist with code interpretation and document analysis, the task requires physical inspection and testing of complex systems on-site, hands-on verification of equipment functionality, and professional judgment that current automation cannot perform reliably end-to-end. Partial automation of compliance documentation review is possible, but the core inspection and testing work remains dependent on human presence and tactile verification.
Task automatabilityclaude-sonnet-51/5This requires physical, hands-on inspection and testing of fire protection systems on-site, which current AI systems cannot perform end-to-end without robotic embodiment.'
Adoption barriersclaude-haiku-4-5-202510015/5Fire code compliance and system certification are highly regulated; most jurisdictions legally require a licensed fire inspector or engineer to certify that systems meet codes. Liability for system failures is asymmetric—inadequate inspection creates public safety risks. Human authorization and sign-off are typically mandatory, creating a hard regulatory barrier to automation.
Adoption barriersclaude-sonnet-55/5Fire inspections are typically legally mandated to be performed by licensed/certified fire inspectors, with liability, code enforcement authority, and legal sign-off requirements that cannot be delegated to AI.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of deploying AI systems capable of physical inspection (robotics, sensors, integration) plus the overhead of oversight and liability management would far exceed the loaded wage of a fire inspector, especially given regulatory requirements for human certification and sign-off.
Cost vs. human wageclaude-sonnet-51/5AI cannot yet substitute for the physical inspection labor, so there is no meaningful AI cost basis to compare against human wages for the core task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product performs on-site fire system inspection and testing. Current AI lacks embodied capability to physically test sprinkler systems, alarm functionality, exit signage, or conduct the tactile/visual verification required by standards. Research prototypes for code compliance checking exist but do not perform actual system testing.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously conducts physical fire system inspections and testing; this remains a human field task requiring direct physical interaction with equipment.

Dust evidence or portions of fire scenes for latent fingerprints.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire departments and investigative agencies are slow-moving, tradition-bound organizations with limited automation adoption. This specific forensic task requires human judgment, legal defensibility, and expert decision-making that organizations have shown no inclination to automate.
Sector adoption velocityclaude-sonnet-51/5Fire investigation is a physical, hands-on field profession with minimal digitization of the core evidence-collection task, and no robotics/AI adoption trend exists for this specific action.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist by analyzing images of dusted fingerprints post-collection or flagging areas for investigator attention, but current systems offer minimal productivity boost during the active dusting phase itself. The investigator's expertise and manual skill remain central.
Augmentation potentialclaude-sonnet-52/5AI can assist with documentation, image analysis of collected fingerprints, or scene mapping afterward, but offers no meaningful help during the physical act of dusting for prints.
Task automatabilityclaude-haiku-4-5-202510011/5Dusting for latent fingerprints requires precise manual dexterity, tactile feedback, and real-time adjustment to surface texture and contamination—capabilities far beyond current AI robotics deployed in the field. No end-to-end automation system exists that can reliably perform this delicate forensic technique.
Task automatabilityclaude-sonnet-51/5This is a physical, manual forensic technique requiring hands-on manipulation of fingerprint powder, brushes, and evidence at a physical scene; no AI system can perform the physical dusting act itself.
Adoption barriersclaude-haiku-4-5-202510015/5Chain-of-custody and forensic evidence standards require certified human investigators to handle, document, and testify about evidence collection. Legal and regulatory frameworks mandate human expertise and accountability in fire scene investigation and fingerprint evidence processing.
Adoption barriersclaude-sonnet-54/5Fire investigation evidence collection often requires certified/trained personnel for chain-of-custody and legal admissibility in criminal or insurance proceedings, creating strong procedural and legal barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Even if a robot could theoretically perform dusting, integration, maintenance, and oversight costs would far exceed the wage of a trained fire investigator performing this task themselves. Robotics capex and operational complexity make this economically infeasible today.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute for physical evidence collection, so the cost comparison favors the human entirely since AI cannot perform the task at all.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs latent fingerprint dusting autonomously. While computer vision can analyze dusted prints, the physical act of dusting with brushes and powders on fragile, irregular fire-scene surfaces requires embodied robot manipulation not yet in production use.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical fingerprint dusting; this remains a manual forensic task performed by trained human investigators using physical tools.

Testify in court cases involving fires, suspected arson, and false alarms.

0

CI 0–0 · exposure 0 · augmentation 38 · importance 4.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This task is embedded in courtroom procedure and legal process where human expert testimony is mandated by law; no sector-level adoption of AI replacement is possible or occurring.
Sector adoption velocityclaude-sonnet-51/5Legal and judicial systems evolve slowly regarding evidentiary standards, and there is no trend toward AI-generated testimony in courts.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist an investigator in preparing testimony by organizing evidence, summarizing case details, or drafting talking points, but the courtroom act itself must remain under human control and verbal delivery.
Augmentation potentialclaude-sonnet-53/5AI can help investigators organize evidence, draft reports, and prepare testimony narratives beforehand, but it does not participate in the courtroom act itself.
Task automatabilityclaude-haiku-4-5-202510011/5Courtroom testimony requires real-time judgment, credibility assessment, cross-examination response, and legal argumentation that depend on human presence, professional licensure, and sworn statements. AI cannot appear as a witness or provide legally binding expert testimony.
Task automatabilityclaude-sonnet-51/5Courtroom testimony requires a sworn human witness with firsthand knowledge, professional credibility, and ability to respond to cross-examination; AI cannot legally or practically substitute for this.
Adoption barriersclaude-haiku-4-5-202510015/5Testifying in court requires the witness to be legally qualified, sworn under oath, subject to perjury laws, and present for cross-examination—all hard statutory and procedural barriers that prohibit AI substitution.
Adoption barriersclaude-sonnet-55/5Testimony requires a qualified, sworn human witness under rules of evidence and professional licensure/certification as a fire inspector/investigator, creating an absolute legal barrier to AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5A fire investigator's courtroom testimony cannot be substituted by AI at any price point, as the task is legally mandated to be performed by a credentialed human professional accountable under oath.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI alternative to price against human testimony, so AI cannot be cheaper for delivering sworn court testimony.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product can serve as a replacement expert witness in court; testimony must come from a licensed fire investigator and is a legal requirement. AI has no capability to fulfill this role in any jurisdiction today.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product testifies in court proceedings; this remains entirely outside current AI product capability and legal frameworks.

Package collected pieces of evidence in securely closed containers, such as bags, crates, or boxes, to protect them.

0

CI 0–0 · exposure 0 · augmentation 13 · importance 4.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire investigation is a low-digitization, physically-grounded sector where evidence handling remains a core procedural requirement with no meaningful AI adoption to date.
Sector adoption velocityclaude-sonnet-51/5Fire investigation and forensic evidence handling remain a low-digitization, physically-grounded field with minimal AI adoption for hands-on evidence tasks.
Augmentation potentialclaude-haiku-4-5-202510011/5AI offers minimal assistance for physical evidence packaging; the task is procedural and manual, with no obvious role for AI-driven suggestions or augmentation.
Augmentation potentialclaude-sonnet-52/5AI could help with documentation, labeling suggestions, or checklist generation, but offers little direct assistance with the physical act of packaging evidence.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of evidence in a secure facility environment, including handling fragile or hazardous materials. Current AI systems cannot perform end-to-end physical packaging, securing containers, or the chain-of-custody verification this task demands.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring hands-on collection and secure packaging of physical evidence at a scene, which current AI systems cannot perform without robotic embodiment.None of this can be done by software-based AI today.
Adoption barriersclaude-haiku-4-5-202510015/5Evidence packaging is subject to strict legal chain-of-custody requirements and regulatory standards governing evidence handling. Only authorized personnel can package evidence to maintain its admissibility in court and investigative integrity.
Adoption barriersclaude-sonnet-55/5Chain-of-custody and evidentiary integrity rules typically require certified personnel to collect and package evidence, creating strict legal and procedural barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Robotic systems capable of evidence handling and secure packaging would require substantial specialized hardware and integration costs, far exceeding the wage cost of a fire inspector performing this task.
Cost vs. human wageclaude-sonnet-51/5There is no AI system capable of performing this physical task, so the human is the only viable option and thus far cheaper than any nonexistent AI alternative.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can autonomously package physical evidence while maintaining legal chain-of-custody requirements. Robotics in warehouses exist, but none operate in the constrained, regulated environment of fire investigation evidence rooms.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs physical evidence packaging; this remains firmly a manual, hands-on task performed by trained investigators.

Subpoena and interview witnesses, property owners, and building occupants to obtain information and sworn testimony.

0

CI 0–0 · exposure 0 · augmentation 38 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire inspection is a traditional government service in small-to-medium organizations with limited digitization. Legal and witness interaction requirements mean this sector has shown minimal adoption of automation in core investigative functions.
Sector adoption velocityclaude-sonnet-51/5Fire investigation and legal/investigative functions in public safety sectors show minimal AI agent adoption for interviewing or legal processes.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with scheduling, organizing witness statements, or generating interview templates, but the core act of witness examination and testimony collection remains entirely human-dependent, limiting augmentation value.
Augmentation potentialclaude-sonnet-53/5AI can assist with transcription, summarizing interview notes, flagging inconsistencies, or drafting interview questions, providing moderate productivity support while the human retains all judgment and authority.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires the authority to compel testimony under oath and the judgment to conduct investigative interviews—capabilities fundamentally beyond current AI systems. Legal testimony collection and witness examination cannot be performed end-to-end by AI without human authority and presence.
Task automatabilityclaude-sonnet-51/5Issuing subpoenas and conducting witness interviews to obtain sworn testimony requires legal authority, in-person judgment, and adaptive questioning that AI cannot perform end-to-end today.'
Adoption barriersclaude-haiku-4-5-202510015/5Hard legal barriers: subpoena authority is granted only to licensed fire inspectors and law enforcement; testimony must be sworn by a human official. Regulatory and statutory requirements explicitly mandate human authority and accountability.
Adoption barriersclaude-sonnet-55/5Subpoena power and sworn testimony collection require legal authorization and often licensed/certified investigator status, making this a hard legal barrier to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task requires a licensed human investigator with legal authority; AI cannot substitute for this role at any cost since the human's legal standing is non-negotiable, making the comparison economically irrelevant.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this function, so cost comparison favors the human entirely; any AI involvement would only be a minor support tool, not a replacement.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product can issue subpoenas or conduct binding investigative interviews. These require legal authority vested in human officials and the credibility of an in-person or official proceeding that AI systems cannot replicate in production settings.
Technical feasibility todayclaude-sonnet-51/5No deployed product conducts legally binding interviews or issues subpoenas; this remains a human-only, research-irrelevant task in current AI products.

Conduct internal investigation to determine negligence and violation of laws and regulations by fire department employees.

0

CI 0–0 · exposure 0 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Adoption of AI for formal internal investigations is near-zero; public sectors and fire departments move slowly on automation of sensitive personnel matters, and legal risk prevents autonomous investigation.
Sector adoption velocityclaude-sonnet-51/5Public sector fire departments and internal affairs functions are slow to adopt AI for personnel investigations given legal and labor sensitivities.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by organizing evidence, flagging relevant regulation citations, or summarizing documentation, but the core investigative judgment and legal determination must remain with human investigators.
Augmentation potentialclaude-sonnet-53/5AI can help organize case files, transcribe interviews, search regulations, and draft investigative reports, aiding the investigator's efficiency without replacing judgment.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires human judgment about legal negligence, interpretation of regulations, and assessment of employee conduct—domains where current AI cannot reliably make binding determinations without human authority and accountability. No AI system can independently conduct formal investigations with legal standing.
Task automatabilityclaude-sonnet-51/5This requires interviewing employees, assessing credibility, applying judgment about culpability, and navigating organizational politics and legal standards hat current Aic cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and organizational barriers exist: only authorized human investigators can conduct formal inquiries into employee conduct, legal liability for erroneous findings rests with humans, and regulations typically mandate human sign-off on disciplinary decisions.
Adoption barriersclaude-sonnet-55/5Internal investigations involving employee discipline, due process, union contracts, and legal liability require authorized human investigators, often with statutory or civil service protections.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI provides minimal cost savings here since the task fundamentally requires trained investigators and legal review by qualified humans; automation offers no meaningful economic advantage.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the investigator's role, so there is no meaningful cost comparison—human investigators with authority and judgment are required.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product conducts internal investigations into employee negligence or regulatory violations autonomously; this requires human decision-makers with legal authority and accountability that AI cannot assume.
Technical feasibility todayclaude-sonnet-51/5No deployed product conducts internal disciplinary/negligence investigations of public safety personnel; this remains a human investigative and adjudicative function.

Coordinate efforts with other organizations, such as law enforcement agencies.

0

CI 0–0 · exposure 0 · augmentation 38 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Fire departments are traditionally conservative, physically-grounded organizations with low AI adoption rates; inter-agency coordination involves human relationships and legal accountability that show no production-scale AI automation.
Sector adoption velocityclaude-sonnet-51/5Public safety and investigative sectors show low AI adoption for coordination and liaison functions, which remain manual and relationship-based.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by drafting coordination requests or summarizing prior agency interactions, but the core task of building trust and executing agreements between organizations remains fundamentally human-centered.
Augmentation potentialclaude-sonnet-53/5AI can help by organizing communications, summarizing reports, and tracking case information shared across agencies, aiding but not replacing coordination efforts.
Task automatabilityclaude-haiku-4-5-202510011/5Coordinating efforts requires negotiation, relationship management, and dynamic interpersonal decision-making between organizations with different goals and authorities. Current AI systems cannot autonomously initiate or conduct meaningful multi-stakeholder coordination.
Task automatabilityclaude-sonnet-51/5Interagency coordination requires relationship-building, on-scene judgment, negotiation, and trust that cannot be executed end-to-end by current AI systems.
Adoption barriersclaude-haiku-4-5-202510015/5Legal authority, liability for inter-agency intelligence-sharing, and regulatory oversight of fire investigation place this task firmly in the domain of licensed human officials; no AI system can substitute for the legal standing required.
Adoption barriersclaude-sonnet-55/5Interagency coordination in investigations typically involves legal authority, chain-of-custody, and jurisdictional protocols requiring authorized human officials.
Cost vs. human wageclaude-haiku-4-5-202510011/5The human inspector's coordination role involves trust-building and accountability that cannot be outsourced to AI; attempting to replace it would require human oversight anyway, making the all-in cost higher than a direct human coordinator.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this coordination role, so cost comparison favors the human entirely.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs autonomous inter-agency coordination in production; this requires legal authority, human judgment on sensitive information-sharing, and accountability that AI cannot assume.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs cross-agency coordination for fire/law enforcement investigations; this remains a human liaison function.

Attend training classes to maintain current knowledge of fire prevention, safety, and firefighting procedures.

0

CI 0–0 · exposure 0 · augmentation 38 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This task is in a heavily regulated public safety sector where attendance is mandated and non-negotiable. There is no adoption pathway for AI substitution in training attendance within fire services.
Sector adoption velocityclaude-sonnet-51/5Public safety and fire services are a traditionally low-digitization, physically-oriented sector with slow AI adoption for compliance-driven training tasks.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could provide marginal support through pre-class content summaries, practice exam generation, or post-class reinforcement materials, but these assist learning preparation rather than transforming the training attendance experience itself.
Augmentation potentialclaude-sonnet-53/5AI can help by providing supplementary study materials, summarizing regulations, or offering practice quizzes to support learning, though it cannot replace attendance itself.
Task automatabilityclaude-haiku-4-5-202510011/5Attending training classes is inherently a human learning activity requiring presence, engagement, and real-time interaction with instructors and peers. AI cannot meaningfully substitute for the legal and professional requirement that fire inspectors physically attend and participate in mandated training.
Task automatabilityclaude-sonnet-51/5Attending training and internalizing hands-on procedural knowledge requires human physical presence, participation, and certification; AI cannot attend classes or absorb experiential training on someone's behalf.'
Adoption barriersclaude-haiku-4-5-202510015/5Regulatory and legal barriers are absolute: fire inspectors are required by law and professional licensing to attend training and maintain certifications personally. These attendance requirements cannot be delegated to or replaced by AI systems.
Adoption barriersclaude-sonnet-55/5Continuing education and certification requirements for fire inspectors are typically mandated by licensing bodies and cannot be legally satisfied by an AI system attending in a person's place.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost structure is inverted: fire inspectors must pay for or be provided training as part of their professional licensing. AI tools might reduce preparation time but cannot eliminate the core human attendance requirement, making replacement irrelevant to cost analysis.
Cost vs. human wageclaude-sonnet-51/5There is no AI alternative that fulfills the certification/attendance requirement, so cost comparison favors the human by default since the task cannot be offloaded.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can attend training classes on behalf of a human or fulfill regulatory attendance requirements. While AI can assist with content review or pre-training preparation, it cannot replace the certification-mandated attendance itself.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs 'attending training' as a substitute for a human; this is inherently a personal, in-person learning and certification activity.

Related occupations — Protective Service

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.