Brownfield Redevelopment Specialists and Site Managers

11-9199.11
Median wage $141,900/yr622,190 employed (US)Rank #626 of 923 scored · top 68% by substitution

Plan and direct cleanup and redevelopment of contaminated properties for reuse. Does not include properties sufficiently contaminated to qualify as Superfund sites.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution22
Exposure22
Augmentation57

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

23 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%23

panel mean rating 1.9/5 → substitution pressure 23/100

Technical feasibility todayw 20%20

panel mean rating 1.8/5 → substitution pressure 20/100

Cost vs. human wagew 15%24

panel mean rating 1.9/5 → substitution pressure 24/100

Adoption barriersw 20%inverted — strong barriers lower the score23

panel mean rating 4.1/5 (barrier strength) → substitution pressure 23/100

Sector adoption velocityw 10%18

panel mean rating 1.7/5 → substitution pressure 18/100

Task breakdown (23 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.

Maintain records of decisions, actions, and progress related to environmental redevelopment projects.

53

CI 46–60 · exposure 58 · augmentation 75 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment is a specialized, often small-team domain with lower digitization rates than finance or tech. Adoption of advanced AI for record management remains in pilot phases; most organizations still rely on manual or legacy systems.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and brownfield redevelopment is a niche, moderately digitized sector with slower AI tool adoption compared to fast-moving information/finance sectors.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can substantially assist specialist and site managers by auto-generating initial drafts of project logs, organizing documents, cross-referencing decisions with regulations, and alerting to missing records, significantly reducing administrative burden while keeping humans in oversight roles.
Augmentation potentialclaude-sonnet-54/5AI can significantly speed up drafting, organizing, and summarizing records and progress reports, allowing specialists to focus on decision-making while maintaining oversight of accuracy.
Task automatabilityclaude-haiku-4-5-202510013/5AI can automate significant portions of record-keeping—parsing documents, organizing data into structured formats, and flagging inconsistencies—but requires human judgment to decide which decisions and actions warrant recording and their classification within complex regulatory frameworks. Setup and oversight overhead is material.
Task automatabilityclaude-sonnet-54/5Recordkeeping and documentation of decisions/actions is a structured text task well-suited to AI drafting, summarization, and organization tools, though some human verification is needed for accuracy and completeness.
Adoption barriersclaude-haiku-4-5-202510014/5Environmental redevelopment records are subject to regulatory requirements (EPA, state agencies) that often mandate human accountability and proper documentation practices; liability and audit concerns create a requirement for human review and sign-off rather than pure automation.
Adoption barriersclaude-sonnet-53/5Environmental redevelopment projects often require regulatory-compliant documentation and audit trails, so there is moderate friction from compliance and legal accountability requirements even though the task itself isn't inherently licensed.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-powered document processing and record management tools cost substantially less than hiring administrative staff for manual entry and organization, though oversight and integration labor partially offset the savings. The ratio is favorable but not an order of magnitude.
Cost vs. human wageclaude-sonnet-54/5AI-assisted documentation tools are inexpensive compared to specialist labor hours spent on manual recordkeeping, though integration with regulatory-specific formats adds some cost.
Technical feasibility todayclaude-haiku-4-5-202510013/5Document management and data entry automation products exist and are deployed, but they struggle with domain-specific terminology in environmental redevelopment and often require manual verification of critical records. Error rates on misclassified or missed entries remain above acceptable thresholds for some compliance contexts.
Technical feasibility todayclaude-sonnet-53/5Products like document management systems with AI summarization and note-taking assistants exist and are used in project management, but domain-specific environmental compliance recordkeeping still requires human oversight and customization.

Prepare and submit permit applications for demolition, cleanup, remediation, or construction projects.

44

CI 34–55 · exposure 53 · augmentation 75 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Environmental remediation and brownfield specialists operate in heavily regulated, geographically fragmented markets with low digital maturity; adoption of AI automation in permit workflows remains limited to large firms, with most work still manual.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and construction permitting sectors are traditionally slow to digitize, with heavy reliance on paper-based or agency-specific portals and limited AI tool integration compared to finance or professional services.'
Augmentation potentialclaude-haiku-4-5-202510014/5AI can dramatically assist by auto-populating forms, cross-referencing regulations, flagging missing documentation, and generating drafts; specialists using AI tools can prepare applications far faster and with fewer errors while maintaining final technical authority.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully speed up drafting of narrative sections, checklists, and compiling supporting documentation, letting specialists focus on technical review and agency negotiation.'
Task automatabilityclaude-haiku-4-5-202510014/5Current AI can draft, populate, and substantially complete permit applications by extracting project data, regulatory requirements, and historical templates; however, final review and site-specific technical justifications typically require expert judgment, limiting full end-to-end automation to about 70–80% efficiency gains.
Task automatabilityclaude-sonnet-53/5AI can draft permit application forms and compile boilerplate content from templates and prior filings, but assembling site-specific technical data, engineering specs, and agency-specific requirements still requires substantial human expertise and verification.'
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory bodies often require a licensed engineer or site manager to sign or certify applications, and many jurisdictions mandate human submission or approval; these legal/licensing requirements create hard adoption barriers regardless of AI capability.
Adoption barriersclaude-sonnet-54/5Permit applications for remediation and demolition often require signatures from licensed engineers, environmental professionals, or PE-stamped documents, and regulatory agencies mandate accountable human submission and liability.'
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-assisted application preparation (drafting, data aggregation, compliance checking) costs significantly less than the specialist labor required for manual compilation, reducing preparation time by 60–70% while integration and oversight overhead remains moderate.
Cost vs. human wageclaude-sonnet-52/5While AI can speed drafting, the need for licensed professional review, site-specific data integration, and regulatory accuracy checks keeps human oversight costs high relative to AI savings.'
Technical feasibility todayclaude-haiku-4-5-202510013/5Document automation and form-filling tools exist in production (RPA, document AI), but permit application submission varies widely by jurisdiction and agency system maturity; many agencies still require wet signatures or human interaction, making reliable end-to-end deployment inconsistent across contexts.
Technical feasibility todayclaude-sonnet-52/5Generic document-drafting AI tools exist but no mature product specifically automates environmental/construction permit application preparation and submission reliably across varied jurisdictions today.'

Estimate costs for environmental cleanup and remediation of land redevelopment projects.

44

CI 25–62 · exposure 45 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Environmental and real estate sectors show moderate AI adoption in estimation tools and data analysis, with many firms piloting or selectively deploying AI-powered cost modeling. Adoption is faster in larger firms and specialized consultancies but lags in smaller operators and geographically fragmented projects.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and brownfield redevelopment sectors are slow to adopt AI due to regulatory complexity, site variability, and reliance on physical inspections.
Augmentation potentialclaude-haiku-4-5-202510014/5AI significantly accelerates hypothesis generation, scenario modeling, and data synthesis for environmental engineers, allowing them to explore multiple remediation pathways and cost trade-offs rapidly. The human remains essential for interpreting site complexity and regulatory fit, but AI-assisted estimation dramatically raises analyst productivity.
Augmentation potentialclaude-sonnet-53/5AI can help aggregate historical cost data, generate cost models, and draft estimate reports, offering useful support while humans still validate and finalize the assessments.
Task automatabilityclaude-haiku-4-5-202510014/5AI can analyze soil surveys, contamination reports, regulatory databases, and historical site data to generate detailed cost estimates for remediation work. Modern systems can cross-reference treatment methods, equipment requirements, and labor benchmarks to produce actionable estimates, though human validation of complex site-specific factors remains prudent.
Task automatabilityclaude-sonnet-52/5Cost estimation requires site-specific environmental data, regulatory judgment, and negotiation with contractors/agencies that current AI cannot autonomously synthesize end-to-end, though it can assist with parts of the calculation.
Adoption barriersclaude-haiku-4-5-202510013/5Professional licensure requirements (PE/environmental certification) for sign-off on final estimates, liability exposure for cost underruns, and regulatory audit trails create moderate friction. However, AI-generated estimates are increasingly accepted as a preliminary or supporting analysis rather than a human substitute.
Adoption barriersclaude-sonnet-54/5Remediation cost estimates often feed into regulatory filings and liability determinations, requiring certified professionals (e.g., licensed environmental engineers) to sign off, creating significant professional and legal barriers.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-based estimation tools require modest compute and licensing costs, while professional environmental engineers command high hourly rates ($150–250+). Amortized cost per estimate heavily favors AI, though some human review overhead remains.
Cost vs. human wageclaude-sonnet-52/5AI tools can speed up drafting of estimates but still require expensive human environmental engineers and site assessments, so overall cost savings versus a qualified specialist are modest.
Technical feasibility todayclaude-haiku-4-5-202510013/5Specialized environmental remediation software and AI-augmented platforms exist and are used in practice, but they typically require significant manual data input, expert review, and site-specific calibration. Performance is reliable for routine estimates but material error rates persist on novel contamination scenarios or complex regulatory stacks.
Technical feasibility todayclaude-sonnet-52/5Some cost-estimation software incorporates AI-assisted analytics, but no deployed product reliably produces defensible remediation cost estimates without heavy expert review and site inspection input.

Prepare reports or presentations to communicate brownfield redevelopment needs, status, or progress.

43

CI 30–56 · exposure 42 · augmentation 75 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment is concentrated in real estate and environmental consulting sectors with moderate digitization; while some firms pilot AI-assisted drafting, production adoption remains limited due to high liability sensitivity and reliance on specialized professional judgment.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and municipal redevelopment sectors are slow, moderately digitized, and have limited AI tool adoption compared to finance or professional services, with most usage still pilot-level.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can significantly augment specialist productivity by auto-generating report templates, organizing and visualizing site data, drafting initial sections, and formatting presentations, freeing the specialist to focus on technical analysis, regulatory interpretation, and stakeholder communication—a strong assistive role.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully speed up drafting, summarizing project status, formatting presentations, and generating first drafts, substantially boosting productivity while specialists retain oversight of technical accuracy.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist in drafting sections of reports (e.g., summarizing site data, formatting presentations) but cannot independently assess complex redevelopment needs, validate technical findings, or make judgment calls on site status without human expertise in environmental remediation and regulatory compliance.
Task automatabilityclaude-sonnet-53/5AI can draft substantial portions of reports and presentations from provided data and templates, but synthesizing site-specific technical, regulatory, and stakeholder context still requires human judgment and verification, limiting full end-to-end automation.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory and liability barriers are significant: reports drive environmental and legal decisions, often become regulatory submissions, and must be signed off by qualified professionals; clients and agencies typically require human expert authorship and accountability, limiting full automation.
Adoption barriersclaude-sonnet-52/5There is generally no licensure requirement to write status reports, though regulatory submissions and stakeholder communications may require sign-off by a qualified environmental professional, creating some but not strong barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5While AI tools reduce drafting labor, the specialist's domain expertise and oversight cost remains high; AI integration adds coordination overhead, and the loaded cost of a brownfield specialist reviewing and validating AI output is still comparable to or exceeds purely manual report preparation.
Cost vs. human wageclaude-sonnet-54/5Drafting reports and slides with AI tools is dramatically cheaper per hour than specialist labor, though final review, data verification, and domain expertise still require paid professional time, moderating the full savings.
Technical feasibility todayclaude-haiku-4-5-202510013/5Document-generation AI and presentation tools exist and are increasingly deployed, but they require substantial human oversight to ensure technical accuracy, regulatory correctness, and contextual appropriateness for brownfield-specific audiences and compliance frameworks.
Technical feasibility todayclaude-sonnet-53/5General-purpose AI writing and presentation tools (e.g., LLM-based drafting assistants, slide generators) are deployed widely for report drafting, but no specialized brownfield-reporting product reliably handles the domain-specific technical and regulatory content without human review.

Identify and apply for project funding.

26

CI 23–30 · exposure 25 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Brownfield redevelopment is a specialized, heavily regulated sector with limited digital transformation. Adoption of AI for funding applications remains minimal; most organizations rely on experienced grant managers and consultants rather than automated systems.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and public-sector-adjacent redevelopment work is a lower-digitization, project-based field with slow AI tool adoption compared to finance or professional services.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by automating funding database searches, drafting application text, identifying eligibility criteria, and organizing requirements, reducing a specialist's research and writing time while they focus on strategy and stakeholder relationships.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully speed up funding source research, draft narratives, and organize application materials, giving specialists a strong productivity boost while they retain final control.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can help draft grant applications and identify potential funding sources, the task requires human judgment on eligibility, strategic fit, and relationship management with funders. Current systems cannot reliably perform the full end-to-end process of identifying appropriate funding, meeting eligibility criteria, and submitting competitive applications without substantial human oversight.
Task automatabilityclaude-sonnet-52/5AI can assist with drafting grant narratives and searching funding databases, but identifying appropriate funding sources, tailoring applications to specific site conditions, and navigating relationship-based funder requirements still needs substantial human judgment and effort.
Adoption barriersclaude-haiku-4-5-202510014/5Significant barriers exist: funding agencies often require authorized signatories and accountable personnel; liability for misrepresenting project eligibility rests with the applicant organization; and lender/grantor relationships involve trust and judgment that regulators expect to see exercised by qualified humans, not delegated to automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement to apply for funding, but grant applications often require certified professional judgment, signatures, and compliance documentation that create organizational friction against pure automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI assistance (research tools, draft generation) reduces time by perhaps 20-30%, but the per-task cost when accounting for integration and required human verification remains comparable to or higher than a human specialist's loaded cost for this task.
Cost vs. human wageclaude-sonnet-52/5AI can cut drafting time but the overall task still requires expert review, site-specific data gathering, and relationship management, so all-in cost savings versus a specialist are modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools exist for grant research and application drafting, but no deployed product reliably identifies and applies for project funding in brownfield redevelopment with consistent quality. The specialized domain knowledge, regulatory understanding, and relationship-building required make this largely manual in practice.
Technical feasibility todayclaude-sonnet-52/5Generic grant-writing and research assistant tools exist, but no deployed product reliably identifies niche brownfield redevelopment funding sources and completes compliant applications without heavy human oversight.

Design or conduct environmental restoration studies.

26

CI 14–37 · exposure 20 · augmentation 25 · importance 3.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Environmental remediation and brownfield redevelopment remain heavily regulated, site-specific, and dependent on licensed professional judgment. Adoption of AI automation in this sector is minimal; most work still follows traditional field-assessment and expert-review workflows with limited digitization or AI integration.
Sector adoption velocityclaude-sonnet-51/5placeholder
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist specialists in data processing, literature review, modeling scenario outcomes, and regulatory documentation drafting, improving analytical efficiency. However, field work, site-specific judgment, and regulatory sign-off remain human-led, limiting augmentation to intermediate rather than transformative gains.
Augmentation potentialclaude-sonnet-51/5placeholder
Task automatabilityclaude-haiku-4-5-202510012/5Environmental restoration studies require extensive field sampling, specialized equipment operation, real-world site assessment, and expert judgment about complex ecological systems. While AI can assist with data analysis and literature synthesis, the core investigative and field-work components cannot be automated end-to-end by current systems.
Task automatabilityclaude-sonnet-52/5Designing and conducting environmental restoration studies requires site-specific field investigation, sampling, regulatory judgment, and interpretation of complex contamination data that current AI cannot autonomously perform end-to-end.dominates.){
Adoption barriersclaude-haiku-4-5-202510014/5Strong regulatory barriers exist: environmental assessments must comply with EPA, state, and local regulations, and typically require licensed professional environmental scientists or engineers to sign off. Liability for incorrect restoration designs is high, and regulatory agencies mandate human expert review and certification.
Adoption barriersclaude-sonnet-51/5placeholder
Cost vs. human wageclaude-haiku-4-5-202510012/5Environmental restoration studies are labor-intensive, requiring qualified environmental specialists, equipment, field work, and regulatory oversight. AI tools may reduce some analytical costs but cannot replace the dominant cost driver—qualified human expertise and on-site investigation—making the all-in cost comparable to or higher than human-only delivery.
Cost vs. human wageclaude-sonnet-51/5placeholder
Technical feasibility todayclaude-haiku-4-5-202510012/5Some specialized software exists for environmental modeling and data analysis, but no deployed product reliably performs the full scope of environmental restoration study design and execution independently. Field sampling protocols, regulatory compliance documentation, and site-specific decision-making remain heavily human-dependent in practice.
Technical feasibility todayclaude-sonnet-51/5placeholder

Conduct quantitative risk assessments for human health, environmental, or other risks.

25

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Environmental remediation and brownfield redevelopment are moderately digitized but remain specialized and risk-averse sectors. Adoption of autonomous AI for risk assessment is slow; pilot projects exist but production-scale AI substitution for regulatory-critical tasks is limited due to liability concerns and professional gatekeeping.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and remediation sectors are historically slow adopters of AI, with heavy reliance on fieldwork, regulatory compliance, and conservative professional practices limiting rapid deployment.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist specialists by automating data collection, running exposure or toxicological models, and generating preliminary analyses that humans review and refine. This augmentation improves speed and thoroughness of exploratory work but does not fundamentally transform the task since human judgment remains essential for parameter selection and regulatory interpretation.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by automating data synthesis, running exposure models, flagging anomalies, and drafting reports, significantly boosting analyst productivity while humans retain judgment and sign-off responsibility.
Task automatabilityclaude-haiku-4-5-202510012/5Quantitative risk assessment involves complex reasoning, judgment calls about uncertain parameters, and integration of domain-specific knowledge that current AI struggles with end-to-end. While AI can accelerate data gathering and model execution, the critical steps—determining risk models, selecting appropriate parameters, interpreting results in real-world context, and justifying assumptions—require human expertise and cannot be reliably automated to meet the 50% time-saving threshold today.
Task automatabilityclaude-sonnet-52/5Quantitative risk assessment requires site-specific field data, regulatory judgment, and integration of complex fate-and-transport models that AI cannot fully execute end-to-end without substantial human expert oversight, though it can accelerate parts of the analysis.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory frameworks (EPA, state environmental agencies, CERCLA) often require signed risk assessments by qualified professionals, and liability for incorrect assessments is substantial. Lenders, insurers, and regulatory bodies typically demand a human expert's signature and professional responsibility on risk conclusions.
Adoption barriersclaude-sonnet-54/5Risk assessments for brownfield sites often require certified professionals (e.g., Professional Engineers, licensed environmental consultants) and must meet regulatory standards (EPA, state agencies), creating strong legal and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI tools (risk modeling software, data analysis platforms) still require substantial specialist oversight and integration; the loaded cost of licensing, computing, and required human validation is comparable to or exceeds the cost of a human conducting the work independently.
Cost vs. human wageclaude-sonnet-52/5AI can reduce time on data processing and literature review, but the need for certified professional judgment, site validation, and regulatory defensibility keeps human involvement costly, limiting overall savings.
Technical feasibility todayclaude-haiku-4-5-202510012/5No mature production systems exist that conduct full quantitative risk assessments for brownfield sites autonomously. Tools exist for narrow subtasks (exposure modeling, statistical analysis), but deployed products do not reliably perform end-to-end site risk assessment without significant human oversight and rework due to domain complexity and regulatory specificity.
Technical feasibility todayclaude-sonnet-52/5Some AI-assisted modeling and data analysis tools exist for environmental risk assessment, but no deployed product reliably performs full quantitative human health/environmental risk assessments autonomously in production settings.

Review or evaluate environmental remediation project proposals.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment is a specialized, heavily regulated sector with modest digital maturity. Adoption of AI agents in production is minimal; most tools remain pilots or advisory aids rather than decision-making systems, and organizational risk aversion is high.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and brownfield redevelopment is a niche, moderately digitized sector with slow AI adoption compared to finance or software; pilots for document review exist but production-scale evaluation tools are rare.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist specialists by pre-screening documents, summarizing data from multiple environmental reports, and flagging inconsistencies or gaps, raising their productivity in information synthesis. However, augmentation is limited to support tasks rather than transformation of the core evaluation work.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully speed up review by extracting key data, flagging inconsistencies, and summarizing lengthy technical proposals, significantly aiding the specialist's productivity while they retain final judgment.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with document analysis and preliminary screening of remediation proposals, the task requires synthesizing complex geotechnical, hydrological, and regulatory data with site-specific context and professional judgment that current systems cannot reliably perform end-to-end at quality parity. Even with significant setup, human expertise remains essential for the evaluation.
Task automatabilityclaude-sonnet-52/5AI can help summarize and flag issues in proposals but the core evaluative judgment involves regulatory compliance, site-specific risk assessment, and liability considerations that require expert human review; full end-to-end automation with equal quality is not yet achievable.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory frameworks often require a licensed environmental professional or engineer to sign off on remediation evaluations, and liability for inadequate site assessment is substantial. Client preference for human accountability and the legal requirement for professional credentials create strong adoption barriers.
Adoption barriersclaude-sonnet-54/5Environmental remediation decisions often require licensed professional engineers or certified environmental professionals to sign off, and regulatory frameworks (e.g., EPA, state agencies) mandate qualified human accountability for approvals.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for document analysis are relatively inexpensive, but the integration overhead, required validation by licensed professionals, and liability costs for errors mean total cost-per-evaluation remains comparable to or exceeds the cost of human specialist review.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply assist with document review and drafting summaries, but the specialist's liability-bearing judgment and site expertise still dominate cost, keeping overall cost savings modest rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some tools exist for document review and data extraction from environmental reports, but no deployed products reliably evaluate remediation proposals as a complete task. Production use cases are narrow and heavily dependent on human oversight; error rates remain material for consequential decisions.
Technical feasibility todayclaude-sonnet-52/5Document analysis tools and LLMs can extract and summarize technical content from proposals, but no deployed product performs holistic environmental remediation proposal evaluation reliably in production without expert oversight.

Conduct feasibility or cost-benefit studies for environmental remediation projects.

25

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Environmental remediation is a regulated, risk-sensitive domain where firms adopt AI cautiously and primarily for auxiliary tasks (data management, initial scoping). Production-level AI displacement in feasibility studies remains limited; adoption is slower than in unregulated information sectors.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and remediation sectors are traditionally slow adopters of AI, with limited digitization of site data and heavy reliance on field work and regulatory processes.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by automating literature synthesis, cost database lookups, and preliminary modeling to accelerate specialist review, but the human expert must remain central to site characterization, regulatory navigation, and final recommendations.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully assist by automating data analysis, cost modeling, literature review, and report drafting, significantly speeding up parts of the feasibility study while specialists retain final judgment and oversight.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with data aggregation, cost estimation, and comparative analysis of remediation options, but the task requires site-specific environmental assessment, regulatory interpretation, and professional judgment that current AI cannot reliably perform end-to-end. Human hydrogeologists, engineers, and environmental scientists remain essential for field validation and liability-bearing decisions.
Task automatabilityclaude-sonnet-52/5AI can assist with data compilation, cost estimation templates, and drafting portions of feasibility analyses, but the core task requires site-specific judgment, regulatory interpretation, and integration of engineering/environmental expertise that current AI cannot reliably perform end-to-end.atable.
Adoption barriersclaude-haiku-4-5-202510014/5Environmental remediation feasibility studies typically require sign-off by licensed environmental consultants or engineers, and the liability and regulatory compliance burden (EPA, state environmental boards) create strong barriers to full automation. Professional credentials and legal accountability are often non-negotiable.
Adoption barriersclaude-sonnet-54/5Environmental remediation feasibility studies often require certified professionals (e.g., PE, environmental consultants) and must meet regulatory standards (EPA, state agencies), creating strong liability and licensing barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted analysis reduces overhead in data compilation and preliminary modeling, but the specialist wages for the required site investigation, expert review, and sign-off are high, and AI cost savings are modest compared to the loaded professional labor required for a credible study.
Cost vs. human wageclaude-sonnet-52/5While AI can cut some research and drafting time, the need for licensed environmental professionals, site visits, and regulatory sign-off means overall costs remain dominated by human expert labor, keeping AI cost savings modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI tools can help generate cost models and summarize environmental data, no mature product reliably performs integrated feasibility and cost-benefit studies for remediation projects that meet regulatory and professional standards. Products exist for fragments (cost estimation, literature review) but not for the full, defensible study.
Technical feasibility todayclaude-sonnet-52/5No deployed products autonomously conduct full remediation feasibility or cost-benefit studies; existing tools are limited to spreadsheet modeling, GIS analysis, or document search rather than complete study production.

Negotiate contracts for services or materials needed for environmental remediation.

25

CI 20–30 · exposure 20 · augmentation 63 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield remediation is sector-specific, capital-intensive, and often involves public-sector coordination; adoption of AI negotiation tools remains pilot-stage, with most firms relying on experienced in-house or external counsel.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and brownfield redevelopment is a physical, project-based, moderately-digitized sector with slow AI adoption for negotiation-specific tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by drafting contract language, flagging non-standard clauses, and summarizing terms, allowing negotiators to focus on commercial and liability strategy rather than document mechanics.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by drafting contract terms, analyzing vendor proposals, benchmarking pricing, and flagging risk clauses, improving the human negotiator's efficiency and preparation.
Task automatabilityclaude-haiku-4-5-202510012/5Contract negotiation requires balancing multiple stakeholder interests, legal risk assessment, and judgment calls on acceptable terms. While AI can draft templates and summarize options, it cannot reliably conclude binding agreements or handle novel commercial disputes—these require human discretion and authority.
Task automatabilityclaude-sonnet-52/5Negotiation involves real-time judgment, relationship management, and trade-offs across price, risk, and regulatory compliance that current AI cannot reliably handle end-to-end.dictionar AI can draft contract language but not conduct the negotiation itself.
Adoption barriersclaude-haiku-4-5-202510014/5Contracts require human signature and legal authority; many remediation projects involve state or federal environmental oversight that mandates licensed professionals to validate and sign agreements, creating licensing and liability barriers to full automation.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement mandates a human negotiator, but liability for contract terms, vendor relationship trust, and organizational authority to bind the company create meaningful friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI contract-analysis tools cost hundreds to thousands monthly; human contract managers in brownfield work earn $70k–$120k+ annually. For full negotiation, human oversight remains necessary, making AI a cost supplement rather than replacement.
Cost vs. human wageclaude-sonnet-52/5Human negotiators remain necessary for the interactive back-and-forth; AI assistance (drafting, benchmarking) lowers some prep costs but doesn't replace the human cost of actual negotiation.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools exist for contract drafting and clause analysis, but deployed systems lack the judgment to negotiate substantive terms in real brownfield contexts where site conditions, remediation scope, and liability allocations are highly variable and site-specific.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously negotiates remediation service/material contracts in production; AI contract tools focus on drafting and review, not live negotiation with counterparties.

Identify environmental contamination sources.

23

CI 20–25 · exposure 20 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment is a specialized, regulation-heavy sector with slower digitization than mainstream industries; pilots of AI-assisted analysis exist, but widespread production adoption remains limited due to liability concerns and the need for licensed environmental professionals.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and site remediation are moderately digitized but adoption of AI for contamination source identification is still nascent, with pilots more common than production deployment.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist specialists by pre-processing historical records, satellite data, and sensor readings to highlight anomalies and patterns, reducing manual data review time and supporting evidence synthesis for the expert's final source identification.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by analyzing historical land-use records, satellite/aerial imagery, and sensor data to help specialists narrow down likely contamination sources, improving efficiency significantly while humans retain judgment and field verification roles.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist in analyzing historical records, satellite imagery, and sensor data to flag potential contamination sources, but identifying sources reliably requires field investigation, expert judgment on complex site conditions, and regulatory context that current AI systems handle poorly end-to-end.
Task automatabilityclaude-sonnet-52/5Identifying contamination sources requires physical site inspection, sampling, historical records review, and geological judgment that current AI cannot perform end-to-end; AI can assist with data analysis but not the core field investigation.'
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory frameworks (CERCLA, state environmental laws) and liability exposure create strong barriers: environmental specialists must certify findings, and errors in source identification can trigger costly legal and remediation consequences, requiring human sign-off and professional judgment.
Adoption barriersclaude-sonnet-54/5Environmental site assessments often require certified professionals (e.g., Phase I/II ESAs under regulatory frameworks like CERCLA/ASTM standards), creating strong licensing and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for data processing are relatively cheap, but the total cost per site (integration, validation, required expert review, site investigation) often exceeds the savings from partial automation, especially given the liability and remediation cost implications of errors.
Cost vs. human wageclaude-sonnet-52/5AI tools can cheaply analyze historical documents or GIS data, but the required soil/groundwater sampling, lab analysis, and expert interpretation keep overall costs comparable to or dependent on human specialists.
Technical feasibility todayclaude-haiku-4-5-202510012/5Products exist for processing satellite imagery and historical data, but they have material limitations in real-world contamination detection and source attribution; no mature production systems reliably identify contamination sources without expert human oversight and on-site validation.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously identifies environmental contamination sources in the field; this remains a research/assistive stage with heavy reliance on licensed environmental professionals and lab testing.

Review or evaluate designs for contaminant treatment or disposal facilities.

23

CI 20–25 · exposure 25 · augmentation 50 · importance 3.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield remediation and site management are heavily regulated, compliance-driven sectors with slow digital transformation; adoption of AI for design evaluation remains minimal, with most organizations relying on traditional expert review workflows.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and brownfield redevelopment is a niche, moderately digitized sector with slow AI adoption compared to finance or software, mostly limited to pilot use of AI for document analysis.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can provide useful assistance by extracting design parameters, checking document completeness, flagging design patterns against known treatment standards, and organizing technical data for human review, materially accelerating the human specialist's review process without replacing their judgment.
Augmentation potentialclaude-sonnet-53/5AI can usefully assist by summarizing technical specifications, checking regulatory compliance references, and flagging anomalies in design documents, improving reviewer efficiency without replacing judgment.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with visual inspection of design documents and flag some technical inconsistencies, but evaluating contaminant treatment facility designs requires expert judgment on regulatory compliance, site-specific geology, risk assessment, and long-term liability—elements that demand human expertise and cannot be fully automated to meet the 50% time-saving bar.
Task automatabilityclaude-sonnet-52/5Reviewing engineering designs for contaminant treatment requires site-specific judgment, regulatory knowledge, and liability-bearing sign-off that current AI cannot fully replicate end-to-end.the task involves nuanced risk assessment beyond pattern matching.
Adoption barriersclaude-haiku-4-5-202510014/5Design reviews for contaminant treatment facilities are typically required to be certified or signed off by licensed professionals (PE/environmental engineers) under EPA and state regulations; liability and regulatory requirements create strong legal barriers to full automation.
Adoption barriersclaude-sonnet-55/5Environmental engineering designs typically require review and sign-off by licensed professional engineers under regulatory frameworks (e.g., EPA, state environmental agencies), creating hard legal barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI assistance reduces document preparation time but does not eliminate the need for licensed environmental engineers or site managers to review designs; the loaded cost of human oversight remains dominant, making the combined cost comparable to or higher than human-only review.
Cost vs. human wageclaude-sonnet-52/5While AI can cheaply scan documents, the need for licensed engineer oversight and liability means the effective all-in cost of AI-assisted review remains comparable to or only modestly cheaper than human expert review.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI tools can extract and summarize design specifications from documents, no deployed product reliably performs independent evaluation of treatment facility designs at the quality required for regulatory or safety sign-off; this task remains primarily manual expert review.
Technical feasibility todayclaude-sonnet-52/5AI tools can assist with document review and flagging inconsistencies, but no deployed product independently performs authoritative design review for contaminant treatment facilities in production.

Develop or implement plans for revegetation of brownfield sites.

23

CI 20–25 · exposure 20 · augmentation 63 · importance 2.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment is a specialized, regulated sector with relatively slow digitization; while planning tools are emerging, production adoption of AI-driven revegetation implementation remains limited to pilot programs and early adopters.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and remediation sectors are slower AI adopters compared to information/finance industries, with physical site work and regulatory processes limiting rapid deployment.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can significantly assist specialists by generating species-selection recommendations, optimizing site layouts, predicting soil recovery timelines, and monitoring outcomes via satellite or sensor data, substantially raising human productivity in planning and adaptive management phases.
Augmentation potentialclaude-sonnet-53/5AI tools can help research suitable plant species, summarize soil contamination data, and draft plan documents, offering meaningful assistance while humans retain responsibility for site-specific decisions and implementation.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with generating revegetation plans using data analysis and modeling (site conditions, soil composition, climate), implementing plans requires on-site judgment, real-time environmental assessment, and adaptive management of living systems that current AI cannot perform end-to-end with 50% time savings at equal quality.
Task automatabilityclaude-sonnet-52/5Developing revegetation plans requires site-specific soil/contaminant assessment, species selection, and regulatory compliance that AI cannot fully execute end-to-end, though it can assist with drafting and research components.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory compliance (environmental permits, ecological standards), liability for site restoration outcomes, and requirement for licensed environmental professionals to sign off on plans create substantial legal and organizational barriers to full automation.
Adoption barriersclaude-sonnet-54/5Environmental remediation projects typically require certified professionals, regulatory sign-off, and liability accountability for contamination risks, creating strong barriers to full AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI planning tools reduce some design labor costs, but implementation oversight, site monitoring, and adaptive management still require significant human expertise; all-in costs remain comparable to or higher than human-led approaches due to integration and ongoing site management needs.
Cost vs. human wageclaude-sonnet-52/5AI could reduce time on literature review or template drafting, but the bulk of cost lies in site testing, permitting, and expert judgment that AI doesn't replace, keeping overall cost comparable to human-led work.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools exist for planning support (predictive modeling, species recommendation engines), but no deployed products reliably handle the full task of developing and implementing revegetation plans independently; implementation remains heavily dependent on skilled human specialists and site-specific iteration.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously develops or implements brownfield revegetation plans; this remains a specialized environmental engineering task requiring field expertise and site assessment.

Provide training on hazardous material or waste cleanup procedures and technologies.

21

CI 16–25 · exposure 17 · augmentation 50 · importance 3.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Hazmat and waste cleanup training remains heavily regulated and human-dependent; while some sectors use AI to supplement or augment training content, actual adoption of AI-led autonomous training is slow due to compliance and safety-critical requirements.
Sector adoption velocityclaude-sonnet-52/5Environmental remediation and hazmat services are a highly specialized, safety-regulated niche with low digitization and slow AI adoption compared to office/professional services sectors. Training delivery in this field remains dominated by in-person, certified instruction with limited AI penetration.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by generating draft training materials, creating practice quizzes, translating procedures into multiple formats, and supporting documentation—useful aids that enhance instructor productivity—but human instructors remain central to assessment and certification.
Augmentation potentialclaude-sonnet-53/5AI can help create training slides, quizzes, updated regulatory content, and answer trainee questions, meaningfully aiding preparation and knowledge dissemination. However, it does not replace the instructor's role in supervised practical exercises and safety demonstrations.
Task automatabilityclaude-haiku-4-5-202510011/5Training delivery involves real-time interaction, assessment of learner understanding, demonstration of physical procedures, and adaptive response to questions—all requiring human judgment and presence that current AI cannot replicate end-to-end at equal quality without substantial human oversight.
Task automatabilityclaude-sonnet-52/5Generating training content and materials could be AI-assisted, but delivering hands-on hazmat training with demonstrations, supervised practice, and compliance verification requires human instructors and physical presence. Current AI cannot fully replace the interactive, site-specific, and safety-critical nature of this training.
Adoption barriersclaude-haiku-4-5-202510014/5OSHA, EPA, and state regulations typically require training to be delivered or certified by qualified human instructors; liability and legal compliance create hard constraints on full automation, and learners must demonstrate competency through supervised practical demonstration.
Adoption barriersclaude-sonnet-54/5Hazardous waste training often falls under OSHA HAZWOPER and similar regulatory frameworks requiring certified trainers and documented competency verification. Liability for improperly trained hazmat workers is high, creating strong incentive to retain qualified human oversight.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-generated training content reduces content creation costs, but qualified human instructors remain legally and practically necessary for live training, assessment, and certification—making the total cost per trainee comparable to or only modestly lower than human-only training.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply produce draft training materials, but the full task includes hands-on instruction, equipment demonstration, and certification oversight that still require paid human trainers. When factoring in liability and compliance requirements, cost savings versus a qualified human trainer are modest at best.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI can generate training materials and lecture content, but no production system reliably delivers hazmat training certification or ensures hands-on procedural mastery, compliance verification, and legal accountability that regulations typically demand.
Technical feasibility todayclaude-sonnet-52/5Products exist for generating e-learning content and quizzes on safety topics, but no deployed AI system independently delivers certified hazmat/OSHA-type training programs at scale in production. Most real-world hazmat training still relies on certified human trainers, especially for practical/field components.

Inspect sites to assess environmental damage or monitor cleanup progress.

15

CI 5–25 · exposure 13 · augmentation 50 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Environmental remediation is a regulated, risk-averse sector with slow digitization. Adoption of AI-driven inspection tools is in pilot stage in some large firms but production deployment remains minimal; the sector favors traditional licensed inspectors due to liability and compliance requirements.
Sector adoption velocityclaude-sonnet-51/5Environmental remediation and site management is a physical, low-digitization field with minimal AI agent deployment for on-site inspection tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5Drone imagery, thermal imaging, and remote sensing can usefully assist inspectors by pre-screening sites, identifying hotspots, and documenting progress photos—raising efficiency in data collection and report generation. However, the core judgment about contamination severity and remediation adequacy still rests with the human specialist.
Augmentation potentialclaude-sonnet-53/5AI can assist with analyzing sensor/sample data, generating reports, or processing drone/satellite imagery to support human inspectors, though the core physical inspection remains manual.
Task automatabilityclaude-haiku-4-5-202510012/5Visual inspection for environmental damage requires on-site physical presence and nuanced judgment about contamination types, severity, and remediation progress. While AI can analyze aerial imagery or lab data, it cannot reliably assess conditions requiring tactile verification, spatial reasoning about subsurface contamination patterns, or real-time decision-making about cleanup adequacy without significant human setup and oversight.
Task automatabilityclaude-sonnet-51/5Physical site inspection requires walking contaminated land, taking samples, and visually assessing conditions in real-world environments that current AI cannot perform autonomously."
Adoption barriersclaude-haiku-4-5-202510014/5Environmental site assessments are often contractually and legally mandated to be performed or signed off by licensed environmental professionals (Phase I/II ESAs, remedial action oversight). Regulatory requirements under CERCLA, state environmental codes, and insurance provisions typically require qualified human professional judgment and certification.
Adoption barriersclaude-sonnet-54/5Environmental regulations often require certified professionals (e.g., licensed environmental engineers/geologists) to inspect and certify cleanup progress, creating strong legal/liability barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools (drones, imagery analysis software) have deployment costs but still require trained human inspectors on-site for verification and interpretation. The all-in cost of AI-assisted inspection remains comparable to or exceeds direct human inspection for high-stakes environmental sites where errors carry legal and health consequences.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute for physical inspection, so the human cost is the only viable cost—AI offers no comparable alternative to displace it.
Technical feasibility todayclaude-haiku-4-5-202510012/5Drone imagery analysis and satellite monitoring exist but narrow in scope and prone to false positives/negatives in complex contamination scenarios. No mature product reliably replaces human inspectors; products serve as aids to supplement rather than substitute for on-site expert assessment in production environments.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product can physically inspect a brownfield site; this remains a research-stage robotics challenge, not a fielded solution.

Develop or implement plans for the sustainable regeneration of brownfield sites to ensure regeneration of a wider area by providing environmental protection or economic and social benefits.

14

CI 3–25 · exposure 13 · augmentation 63 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment specialists work in a traditionally regulated, localized sector with strong professional licensing requirements and low digital-native momentum. Adoption of AI for this task remains limited despite digitization in adjacent sectors.
Sector adoption velocityclaude-sonnet-52/5Environmental consulting and urban redevelopment are traditionally slow-adopting, physically grounded sectors with limited AI agent deployment in production planning workflows.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by analyzing contamination data, modeling remediation scenarios, generating environmental impact reports, or suggesting economic/social benefit structures, raising a specialist's analytical productivity while the human retains decision-making authority.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist with data analysis, environmental modeling, report generation, and synthesizing regulatory requirements, significantly aiding specialists while they retain decision authority.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires complex stakeholder engagement, site-specific environmental assessment, long-term strategic planning, and integration of economic, social, and environmental factors that demand human judgment and negotiation. Current AI cannot autonomously develop or implement such multidisciplinary regeneration plans.
Task automatabilityclaude-sonnet-52/5This requires site-specific environmental assessment, stakeholder negotiation, regulatory navigation, and physical remediation planning that AI cannot execute end-to-end; only sub-components like document drafting or data synthesis are automatable.
Adoption barriersclaude-haiku-4-5-202510015/5Brownfield redevelopment is heavily regulated by environmental laws and often requires licensed professionals (environmental engineers, site managers) to sign off on plans. Legal liability for environmental damage, zoning compliance, and stakeholder approval create hard barriers to autonomous automation.
Adoption barriersclaude-sonnet-54/5Environmental remediation plans typically require licensed environmental professionals, regulatory approvals, and legal sign-off, creating strong institutional and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Brownfield redevelopment requires specialized expertise, regulatory compliance oversight, and client engagement that command high professional fees. AI tools would serve as supplements rather than cost-competitive replacements, making the all-in cost comparison unfavorable for full task substitution.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply assist with research and report drafting, but the overall planning process still requires expensive expert labor, site visits, and regulatory liaison, so total cost savings are modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can assist with data analysis, environmental modeling, or documentation, no deployed product reliably performs the core task of developing and implementing sustainable regeneration plans end-to-end. Pilot tools exist for specific sub-components but not integrated execution.
Technical feasibility todayclaude-sonnet-51/5No deployed product manages brownfield redevelopment planning holistically; this remains a human-led, multidisciplinary process involving engineering, regulatory, and community judgment.

Plan or implement brownfield redevelopment projects to ensure safety, quality, and compliance with applicable standards or requirements.

13

CI 0–25 · exposure 13 · augmentation 50 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield redevelopment is a specialized, regulation-heavy sector with limited digitization and slow AI adoption; projects are bespoke, often involve legacy data and complex site histories, and the industry remains reliant on experienced human specialists rather than adopting agent-based automation.
Sector adoption velocityclaude-sonnet-51/5Environmental remediation and construction site management are physical, low-digitization sectors with minimal AI agent deployment in production.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist with compliance research, environmental data analysis, project timeline optimization, and document drafting, but remains subordinate to human judgment on safety trade-offs, regulatory interpretation, and site-specific decision-making, offering modest productivity gains rather than transformation.
Augmentation potentialclaude-sonnet-53/5AI can assist with document review, regulatory research, environmental data analysis, and report drafting, improving efficiency of specialists managing these projects.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with compliance documentation, risk analysis, and project scheduling, the task requires on-site assessment, stakeholder coordination, regulatory judgment, and adaptive decision-making under uncertain environmental conditions—core elements that remain firmly human-dependent and prevent 50% time savings at equal quality.
Task automatabilityclaude-sonnet-51/5This task requires site-specific engineering judgment, regulatory negotiation, physical inspection, and multi-stakeholder coordination that current AI cannot perform end-to-end.atables cannot execute this fully.
Adoption barriersclaude-haiku-4-5-202510014/5Significant legal and regulatory barriers exist: licensed professionals (environmental engineers, site managers) are often required to sign off on remediation plans and compliance certifications; liability for environmental safety and human health is asymmetric; and regulatory agencies typically mandate human professional sign-off.
Adoption barriersclaude-sonnet-55/5Brownfield redevelopment is heavily regulated (EPA, state environmental agencies) and typically requires licensed professionals (PE, environmental consultants) to sign off on remediation plans and compliance documentation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Brownfield project management is specialized, high-liability work; even with AI assistance for planning and document review, the overhead of domain-specific training, site oversight, regulatory review, and liability exposure keeps total cost per project competitive with or higher than expert human specialists.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the licensed engineers, environmental consultants, and project managers required, so there is no meaningful cost displacement possible today.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools exist for document review, compliance checking, and project management dashboards, but no deployed system can reliably handle end-to-end brownfield planning and implementation, which demands site-specific environmental expertise, regulatory interpretation, and real-time field adjustments beyond current product capabilities.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product manages or plans brownfield redevelopment projects; this remains a human-led, field-based project management function.

Coordinate on-site activities for environmental cleanup or remediation projects to ensure compliance with environmental laws, standards, regulations, or other requirements.

8

CI 0–16 · exposure 5 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Brownfield remediation is a specialized, capital-intensive sector with high regulatory stakes and strong reliance on licensed professionals; adoption of AI coordination tools remains slow and limited to data management and scheduling aids rather than autonomous decision-making.
Sector adoption velocityclaude-sonnet-51/5Environmental services and site remediation are a physically-oriented, moderately digitized sector with low AI agent deployment for field coordination tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist specialists by automating regulatory checklist generation, monitoring weather and scheduling constraints, drafting compliance reports, and tracking equipment/personnel logistics—useful productivity aids but not transformative given the irreducible human expertise required for site decisions.
Augmentation potentialclaude-sonnet-53/5AI tools can help with document review, compliance checklist generation, scheduling, and report drafting, meaningfully supporting the specialist even though on-site coordination itself remains human-led.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time coordination of complex on-site activities, human judgment about regulatory compliance in dynamic conditions, and adaptive decision-making based on environmental conditions—capabilities that current AI systems cannot perform end-to-end without extensive human oversight.
Task automatabilityclaude-sonnet-51/5This task requires physical on-site coordination, real-time supervision of workers and equipment, and adaptive decision-making in dynamic field conditions that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Environmental remediation projects are heavily regulated under federal (EPA, CERCLA, RCRA) and state law, requiring licensed environmental professionals to sign off on compliance decisions and site management plans. Liability for environmental violations is severe and creates legal barriers to full automation.
Adoption barriersclaude-sonnet-55/5Environmental remediation is heavily regulated (RCRA, CERCLA, state programs) and typically requires licensed professionals (PE, environmental consultants) to sign off on compliance, creating hard legal barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5The specialized expertise, liability exposure, and need for real-time decision-making mean that AI assistance would still require a qualified specialist to oversee and validate recommendations, keeping total cost near or above human labor cost for the coordinating role.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical presence and liability-bearing judgment required, so there is no viable AI cost comparison for the core task.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed AI system reliably coordinates on-site environmental remediation projects or ensures compliance with the nuanced, context-dependent environmental regulations that vary by jurisdiction and contamination type. AI can assist with documentation and scheduling, but not with live site coordination.
Technical feasibility todayclaude-sonnet-51/5No deployed product manages on-site remediation coordination and compliance oversight in production; this remains a human field-management role.

Design or implement plans for surface or ground water remediation.

7

CI 0–15 · exposure 8 · augmentation 50 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Brownfield remediation is a small, specialized sector of professional services with slow digitization. Firms are typically mid-sized or small, risk-averse, and rely on established licensed professionals; adoption of AI for autonomous plan design is minimal or non-existent in production.
Sector adoption velocityclaude-sonnet-52/5Environmental engineering and remediation is a slow-adopting, physically-grounded sector with limited AI agent deployment in production compared to information-sector fields.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist with data processing, contaminant modeling, literature review of treatment options, and cost estimation, raising the efficiency of hydrogeologists and remediation engineers during planning phases. However, the augmentation is limited to support roles; human experts must retain responsibility for final design and regulatory compliance.
Augmentation potentialclaude-sonnet-53/5AI can assist with contaminant transport modeling, literature review, regulatory document drafting, and data visualization, meaningfully aiding specialists without replacing core judgment.
Task automatabilityclaude-haiku-4-5-202510011/5Designing or implementing water remediation plans requires expert judgment on site-specific hydrogeology, regulatory compliance, cost-benefit trade-offs, and real-world conditions. Current AI systems cannot autonomously assess contamination plumes, select remediation technologies, or manage the iterative refinement that real projects demand, making end-to-end automation infeasible.
Task automatabilityclaude-sonnet-52/5Designing water remediation plans requires site-specific hydrogeological analysis, regulatory judgment, and engineering design that AI cannot fully perform end-to-end, though it can assist with data analysis and drafting portions of the plan.
Adoption barriersclaude-haiku-4-5-202510015/5Environmental remediation design is tightly regulated; EPA and state regulations require a licensed engineer or environmental professional to design and certify remediation plans. Liability for contamination and human health risks creates high barriers to AI autonomy, and client liability requirements effectively mandate human professional sign-off.
Adoption barriersclaude-sonnet-55/5Remediation plans typically must be designed and sealed by licensed professional engineers or geologists and approved by environmental regulatory agencies, creating hard legal barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Even supportive AI applications require expert hydrogeologists or engineers to review, validate, and adapt outputs, so the all-in cost (inference plus integration plus expert oversight) exceeds the cost of the expert alone doing the work directly, offering no labor cost savings.
Cost vs. human wageclaude-sonnet-52/5AI tools can reduce time on data analysis and modeling, but the overall cost is dominated by field investigation, engineering design, and regulatory compliance that still require expensive human expertise.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product performs autonomous design or implementation of water remediation plans in production. While AI can assist with data analysis and modeling components, the synthesis into a coherent, site-specific, legally defensible remediation strategy remains a human expert function.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously designs or implements groundwater/surface water remediation plans in production; this remains a highly specialized engineering task requiring licensed professional oversight.

Design or implement plans for structural demolition and debris removal.

6

CI 0–11 · exposure 5 · augmentation 50 · importance 3.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Construction and demolition remain low-digitization sectors with limited AI adoption; most firms rely on traditional planning, equipment operators, and manual oversight. Pilots using AI for site modeling exist, but production-scale displacement in actual demolition work is minimal.
Sector adoption velocityclaude-sonnet-51/5Construction and demolition/environmental remediation sectors are among the least digitized and slowest to adopt AI in physical execution tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist planners through 3D modeling, debris volume estimation, equipment scheduling, and environmental compliance checks, boosting efficiency in the design phase. However, augmentation is confined to planning; site execution remains largely human-driven.
Augmentation potentialclaude-sonnet-53/5AI can assist with document drafting, hazard checklists, permit research, and preliminary data analysis (e.g., site surveys, material inventories) to support the human planners, though the core structural design remains human-led.
Task automatabilityclaude-haiku-4-5-202510011/5Structural demolition and debris removal are inherently physical operations requiring on-site presence, equipment operation, and real-time spatial reasoning in dynamic, hazardous environments. Current AI systems cannot perform or meaningfully automate the execution of these tasks.
Task automatabilityclaude-sonnet-51/5Designing demolition and debris removal plans requires site-specific structural engineering judgment, hazard assessment, and regulatory coordination that current AI cannot perform end-to-end reliably.
Adoption barriersclaude-haiku-4-5-202510015/5Demolition and debris removal are heavily regulated by OSHA, EPA, local building codes, and environmental laws; licensed operators and site supervisors are legally required to oversee and sign off on work. Liability for structural safety and hazardous material handling creates hard legal barriers to full automation.
Adoption barriersclaude-sonnet-55/5Demolition plans typically require licensed professional engineer sign-off, permits, and regulatory compliance (safety, environmental, structural codes), creating hard legal barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5While AI-assisted planning tools reduce some preliminary design costs, the bulk of labor and equipment expense occurs in physical execution, which cannot be automated. AI's cost savings are marginal relative to total human labor and heavy machinery costs involved.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this task, so cost comparison favors human experts entirely; AI cannot deliver the output at all.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI can assist with planning phases (e.g., 3D site modeling, schedule optimization) through deployed tools, but no production system reliably executes or independently manages demolition/debris removal operations. Deployed products address fragments of planning only, not task execution.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product designs or implements structural demolition plans; this remains firmly in the domain of licensed engineers and demolition contractors.

Coordinate the disposal of hazardous waste.

4

CI 0–9 · exposure 8 · augmentation 50 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Brownfield remediation is a capital-intensive, heavily regulated sector with small specialist firms; adoption of autonomous AI systems is minimal, and regulatory frameworks actively require certified human decision-makers in the waste disposal process.
Sector adoption velocityclaude-sonnet-51/5Environmental remediation and brownfield redevelopment are physical, compliance-heavy, low-digitization sectors with minimal AI agent deployment for operational hazardous waste tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by automating hazmat inventory tracking, compliance documentation, vendor database management, and permit filing—raising a coordinator's productivity on administrative overhead—but humans must retain authority over disposal decisions.
Augmentation potentialclaude-sonnet-53/5AI can assist with generating waste manifests, tracking regulatory deadlines, summarizing compliance requirements, and organizing documentation, improving efficiency of the human coordinator's paperwork and tracking tasks.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could assist with documentation, compliance tracking, and vendor coordination, the core task requires on-site hazmat assessment, real-time decision-making about disposal methods, and chain-of-custody management. Current systems cannot reliably handle the variability of hazardous materials or replace the expert judgment needed for safe disposal routing.
Task automatabilityclaude-sonnet-51/5This task requires physical logistics coordination, regulatory compliance verification, on-site inspection, and legal accountability for hazardous materials—none of which current AI can execute end-to-end without extensive human action.
Adoption barriersclaude-haiku-4-5-202510015/5Heavy regulatory barriers exist: EPA rules, DOT hazmat shipping requirements, and state/federal licensing mandate human handlers and sign-off. Liability for improper disposal creates asymmetric error costs that prevent full automation, and only licensed waste management professionals can legally authorize disposal routes.
Adoption barriersclaude-sonnet-55/5Hazardous waste disposal is heavily regulated (RCRA, EPA manifests, state environmental law) requiring certified personnel and legal accountability, creating hard barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The liability costs, regulatory compliance overhead, and need for human oversight make AI-assisted coordination more expensive or equally costly compared to employing specialists who carry professional responsibility and insurance.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical coordination, site visits, and liability-bearing sign-offs required, so there is no meaningful AI cost basis to compare against human labor for this task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product performs end-to-end hazardous waste disposal coordination in production. This task demands legal accountability, certified handler qualifications, and real-time site assessment that current AI systems are not operationalized to do independently.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product manages hazardous waste disposal coordination in production; this remains a human-operated regulatory and logistics function involving licensed haulers, manifests, and inspections.

Design or implement measures to improve the water, air, and soil quality of military test sites, abandoned mine land, or other contaminated sites.

3

CI 0–5 · exposure 5 · augmentation 50 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Brownfield remediation remains heavily regulated and site-specific, with slow digital adoption relative to other sectors. Organizations rely on accredited specialists and long regulatory timelines; AI automation is not a production reality in this domain.
Sector adoption velocityclaude-sonnet-51/5Environmental remediation and site management is a slow-moving, physically-grounded, heavily regulated sector with minimal AI agent deployment in production.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist specialists by automating regulatory document preparation, summarizing contamination data, suggesting remediation literature, and analyzing soil/water test results. However, the core design and implementation judgment remains human-dependent.
Augmentation potentialclaude-sonnet-53/5AI can meaningfully assist with data analysis, contamination modeling, regulatory document drafting, and monitoring data interpretation, though the physical design and implementation remain human-led.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires on-site assessment, regulatory interpretation, engineering judgment, and stakeholder coordination that depend on physical site conditions and legal/environmental context. Current AI cannot autonomously design or implement remediation measures without human specialists directing field work, lab analysis, and regulatory compliance.
Task automatabilityclaude-sonnet-51/5This task requires physical remediation design, on-site engineering judgment, regulatory compliance, and hands-on implementation of environmental controls, none of which current AI can execute end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Strict regulatory barriers exist: environmental remediation typically requires licensed Professional Engineers (PE) or Environmental Professionals (EP) to design and certify compliance with EPA, state, and CERCLA standards. Liability for failed remediation is severe, making autonomous AI deployment legally and contractually infeasible.
Adoption barriersclaude-sonnet-55/5Contaminated site remediation is heavily regulated (CERCLA, RCRA, state environmental agencies) and typically requires licensed professional engineers/geologists and government sign-off, creating hard legal barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current AI systems cannot replace the specialized environmental engineers, hydrogeologists, and site managers whose expertise and licensed sign-off are legally required. The cost of deploying AI oversight and integration would exceed the marginal value for this high-liability task.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical fieldwork, sampling, and construction-level implementation involved, so there is no meaningful AI cost basis to compare against human labor for the full task.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools can assist with literature review, regulatory compliance documentation, and basic design calculations, but no deployed product reliably performs end-to-end site remediation design or implementation. Field verification, soil sampling, contaminant testing, and adaptive decision-making remain human-driven.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product designs or implements physical environmental remediation systems; this remains a research-stage and human-engineered domain.

Provide expert witness testimony on issues such as soil, air, or water contamination and associated cleanup measures.

0

CI 0–0 · exposure 0 · augmentation 50 · importance 2.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This task occurs in highly regulated legal and environmental sectors with strong human-expert gatekeeping. No measurable displacement by AI systems exists; courts and professional standards actively resist automated or AI-only testimony.
Sector adoption velocityclaude-sonnet-51/5Legal and environmental consulting fields adopt AI slowly for the actual testimony function itself, given the entrenched procedural and evidentiary requirements of courts.
Augmentation potentialclaude-haiku-4-5-202510012/5AI can assist a human expert in preparing testimony via data synthesis, literature review, or modeling visualization, but these are pre-testimony support tasks. The testimony delivery itself offers minimal augmentation surface since it is narrative, cross-examinable testimony requiring human judgment and credibility.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist in analyzing contamination data, drafting reports, summarizing regulations, and preparing testimony outlines, boosting expert productivity even though it cannot replace the witness role.
Task automatabilityclaude-haiku-4-5-202510011/5Expert witness testimony requires legal qualification, court-room presence, cross-examination response, and professional liability judgment that current AI cannot reliably deliver. While AI can assist in document review or technical analysis, the testimony itself demands a licensed human expert who accepts legal accountability.
Task automatabilityclaude-sonnet-51/5Expert witness testimony requires live human presence, credential-based credibility, real-time cross-examination response, and legal accountability that no AI system can perform end-to-end today.
Adoption barriersclaude-haiku-4-5-202510015/5Legal and regulatory barriers are absolute: only a qualified, licensed human expert can provide sworn testimony in court. Expert witness standards, Daubert criteria, and professional responsibility rules mandate human accountability; automation is legally prohibited.
Adoption barriersclaude-sonnet-55/5Courts require a qualified, credentialed human expert to testify under oath and be subject to cross-examination; this is a hard legal and licensing barrier that cannot be automated away.
Cost vs. human wageclaude-haiku-4-5-202510011/5Expert witness services command high hourly rates ($300–$1,000+) reflecting specialized expertise, liability, and court time. AI assistance is negligible in total cost compared to human expert fees, and cannot replace the billable testimony itself.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the human testifying role at all, so there is no meaningful cost comparison—the human expert must be paid regardless of any AI assistance used in preparation.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product performs expert witness testimony end-to-end; this requires court acceptance, professional licensure, and real-time adversarial interaction that AI systems are not legally positioned or operationally equipped to perform today.
Technical feasibility todayclaude-sonnet-51/5No deployed product provides courtroom expert testimony; this remains entirely a human professional function with only research-stage tools for supporting analysis.

Related occupations — Management

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.