Computer, Automated Teller, and Office Machine Repairers
49-2011.00Median wage $47,810/yr65,600 employed (US)Rank #295 of 923 scored · top 32% by substitution
Repair, maintain, or install computers, word processing systems, automated teller machines, and electronic office machines, such as duplicating and fax machines.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution34
Exposure28
Augmentation53
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
25 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
12%
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%30
panel mean rating 2.2/5 → substitution pressure 30/100
Technical feasibility todayw 20%24
panel mean rating 2.0/5 → substitution pressure 24/100
Cost vs. human wagew 15%28
panel mean rating 2.1/5 → substitution pressure 28/100
Adoption barriersw 20%inverted — strong barriers lower the score59
panel mean rating 2.6/5 (barrier strength) → substitution pressure 59/100
Sector adoption velocityw 10%25
panel mean rating 2.0/5 → substitution pressure 25/100
Task breakdown (25 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.
Complete repair bills, shop records, time cards, or expense reports.
91CI 85–97 · exposure 95 · augmentation 88 · importance 4.1/5 · click for rater detail
Complete repair bills, shop records, time cards, or expense reports.
91| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Automotive repair, facilities management, and office administration sectors have actively adopted automation for time tracking, billing, and expense management over the past 5–10 years. Measured adoption is deep in digitized firms and accelerating, particularly in fleet and service operations. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Field service and repair trades adopt digital tools moderately fast, but this is a lagging-technology sector compared to pure information/finance work, so uptake is uneven across small shops. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assists technicians and administrators by auto-populating forms from scanned receipts, work orders, or voice notes, flagging inconsistencies, and suggesting categorizations. This meaningfully raises human productivity while keeping humans in control of final approval and review. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-assisted transcription, auto-filled forms, and expense categorization tools substantially speed up this documentation task while the technician still reviews and submits it. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Completing repair bills, shop records, time cards, and expense reports involves primarily data entry, categorization, and documentation tasks that are straightforward and rule-based. Current AI systems (including LLMs and RPA) can reliably extract information from source documents, populate structured forms, calculate totals, and generate reports with >50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 5/5 | This is a structured documentation task (filling forms, logging time/expenses) that off-the-shelf AI plus mobile/OCR/voice-to-text apps can already complete with major time savings at equal or better quality. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal or regulatory barriers prevent automation of record completion; it is not a licensed activity and does not require human sign-off by statute. Minor friction exists from internal audit trails or organizational preference for manual verification, but nothing structurally prevents substitution. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human complete these internal administrative records. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven data entry and document processing costs (API calls, software licensing) are typically an order of magnitude cheaper than paying a technician or administrator hourly wages to manually fill out forms and records. Inference and integration overheads are minimal for this standardized task. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated expense/time-tracking and documentation tools cost a small subscription fee versus the technician's loaded hourly wage spent on paperwork, an order-of-magnitude cost reduction. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products (accounting software with AI data entry, RPA platforms, document processing APIs) demonstrably perform invoice/record completion, time tracking, and expense report automation reliably in production across many organizations today. These are mature, widely-used solutions with low error rates on well-structured data. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Field-service management software with AI-assisted note-taking, receipt scanning, and auto-populated time/expense entries is widely deployed in production, though some manual review still occurs. |
Maintain parts inventories and order any additional parts needed for repairs.
78CI 72–84 · exposure 75 · augmentation 75 · importance 4.2/5 · click for rater detail
Maintain parts inventories and order any additional parts needed for repairs.
78| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Inventory automation and AI-driven procurement are already widely deployed in high-digitization sectors (retail, logistics, manufacturing, telecommunications support). Adoption is rapid among larger service organizations and equipment repair chains, though smaller independent repair shops lag. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | placeholder |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI strongly augments human inventory management by flagging stock shortages, recommending reorder quantities based on seasonal demand, and highlighting slow-moving parts—allowing technicians to focus on repair strategy rather than manual stock monitoring and order logistics. |
| Augmentation potential | claude-sonnet-5 | 4/5 | placeholder |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can nearly fully automate inventory tracking, predictive ordering based on historical repair patterns, and supplier management with minimal human oversight. Current systems can integrate with existing inventory databases and automatically generate purchase orders, saving well over 50% of the time spent on manual inventory counts and ordering decisions. |
| Task automatability | claude-sonnet-5 | 4/5 | Inventory tracking and reordering parts based on stock thresholds is a structured, data-driven task well-suited to existing inventory management and procurement software with automated reorder triggers."},"feasibility":{"rating":4,"rationale":"Deployed inventory management systems (ERP, CMMS) already automate stock tracking, reorder point alerts, and purchase order generation reliably in production across many industries."},"cost_ratio":{"rating":4,"rationale":"Automated inventory software costs a small fraction of a technician's time spent manually tracking and ordering parts, especially at scale."},"barriers":{"rating":1,"rationale":"No licensing, liability, or human-contact requirements attach to inventory management; it's a purely administrative/logistics function."},"adoption_velocity":{"rating":3,"rationale":"Field service and repair industries have moderate digitization with inventory software common, but many small repair shops still rely on manual tracking, giving mixed adoption."},"augmentation":{"rating":4,"rationale":"AI-enhanced inventory systems can predict parts needs, flag low stock, and auto-generate orders, significantly reducing technician administrative burden while they remain in charge of final decisions."}}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Minimal regulatory or legal barriers exist for automating inventory and ordering tasks; no licensing requirement applies. Primary friction comes from organizational inertia and incumbent system integration, but these are surmountable without hard legal or liability obstacles. |
| Adoption barriers | claude-sonnet-5 | 1/5 | placeholder |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven inventory and procurement automation costs a small fraction of the human labor it replaces—combining low inference cost with cloud-based SaaS tools makes the cost per task far lower than the loaded wage of a human inventory specialist or repair technician managing stock. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | placeholder |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature inventory management and procurement software with AI-driven forecasting is deployed in production across retail, manufacturing, and service sectors. While some organizational integration and tuning is required, reliable end-to-end inventory automation and automated reorder systems are demonstrably working in similar fields today. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | placeholder |
Maintain records of equipment maintenance work or repairs.
78CI 72–84 · exposure 75 · augmentation 88 · importance 4.1/5 · click for rater detail
Maintain records of equipment maintenance work or repairs.
78| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Field service and IT operations sectors are already rapidly adopting automated record-keeping via ITSM platforms, fleet management systems, and IoT-linked maintenance logs. Adoption is deep in the tech and enterprise sectors. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Field service and repair industries are moderately digitized with growing CMMS/FSM adoption, but many smaller repair shops still use manual or basic digital logs. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistance here is already transformative: technicians can dictate work notes or photos and have systems auto-populate records, checklists, and compliance documentation, freeing human attention for hands-on repair and decision-making. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-assisted dictation, auto-fill, and summarization tools substantially speed up and improve consistency of maintenance recordkeeping while the technician remains the source of truth. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Recording maintenance and repair work is highly structured data entry that involves documenting work performed, parts replaced, and timestamps. Current AI can reliably extract this information from technician notes, photos, or voice logs and populate databases or ticketing systems, achieving substantial time savings with minimal quality loss. |
| Task automatability | claude-sonnet-5 | 4/5 | Recording maintenance work is largely structured data entry (parts used, time, actions taken) that can be captured via voice-to-text, forms, or integration with ticketing/CMMS systems with high time savings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No legal or regulatory requirement mandates that a human manually record maintenance data. Field service records are business documentation, not licensed professional work, so automation faces minimal friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for recordkeeping itself, though some employer/warranty documentation standards and audit trail integrity requirements create minor friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automating record maintenance through AI costs mere cents per incident when integrated into existing software stacks, versus technician labor time (often billed at $40–80/hour). The cost differential is at least 10–50x in favor of AI. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated logging/transcription tools cost a small fraction of technician time spent manually typing reports, though some integration and review overhead remains. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature field service management platforms already integrate OCR, voice-to-text, and form-filling capabilities that are deployed at scale in enterprise settings. These systems reliably capture and organize maintenance records in production environments, though some manual review is typically retained. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Field service management software with AI-assisted dictation, auto-populated fields, and structured logging is widely deployed in production for technician workflows today. |
Analyze equipment performance records to assess equipment functioning.
61CI 47–75 · exposure 58 · augmentation 75 · importance 3.9/5 · click for rater detail
Analyze equipment performance records to assess equipment functioning.
61| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Predictive maintenance and automated diagnostics have achieved rapid, measurable adoption in manufacturing, telecom, data center, and enterprise IT sectors, with many firms already deploying ML-based monitoring in production. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Repair and field-service trades adopt AI slowly compared to information/finance sectors, with predictive analytics tools still in early or pilot phases in this niche. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems already augment technicians by pre-screening records, surfacing likely failure modes, and prioritizing work queues, substantially shortening the time spent on manual log review and pattern-spotting. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist technicians by summarizing performance logs, flagging anomalies, and suggesting likely failure points, improving diagnostic speed while the technician still verifies and acts. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Analyzing equipment performance records against baseline standards is highly structured and data-driven; modern AI systems can extract key metrics, identify anomalies, and generate diagnostic summaries automatically. While some complex troubleshooting may still require human judgment, the core analysis function easily meets the ≥50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 3/5 | Analyzing structured performance logs to identify trends or anomalies is well within current AI data-analysis capability, but integrating with diverse legacy equipment data formats and drawing repair-relevant conclusions requires setup and domain-specific tuning.dis Full end-to-end automation at scale is not yet standard.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No regulatory licensing mandate exists for automated analysis of equipment logs; organizations face only organizational inertia and quality-assurance oversight preferences. The task itself carries low legal/liability friction compared to repair execution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks automated data analysis, though organizations may still want a technician to validate findings before acting, creating minor procedural friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-based performance analysis (inference on logs, pattern detection, report generation) costs significantly less than technician labor per analysis cycle, especially when amortized across many devices. The cost is typically 5–10× lower than a technician review, though integration and tuning add overhead. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Once data pipelines are built, AI analysis is cheap per run, but the integration and validation overhead for heterogeneous equipment logs keeps costs roughly comparable to a technician's routine review in many shops. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products (predictive maintenance platforms, log analysis tools, anomaly detection systems) reliably perform performance record analysis in production environments across industrial, IT, and service sectors. Minor gaps remain in integrating novel equipment types, but the general capability is mature and field-proven. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Predictive maintenance and log-analysis tools exist in some industrial contexts, but for this specific occupation (ATM/office machine repair) there are few widely deployed production systems doing this analysis autonomously. |
Enter information into computers to copy programs from one electronic component to another or to draw, modify, or store schematics.
60CI 44–76 · exposure 58 · augmentation 63 · importance 3.5/5 · click for rater detail
Enter information into computers to copy programs from one electronic component to another or to draw, modify, or store schematics.
60| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | The electronics repair and IT maintenance sectors show moderate adoption of automation tools, with routine data entry and file management increasingly automated. However, deployment is still mixed—many smaller repair shops and legacy operations continue manual processes, limiting rapid sector-wide velocity. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field service and repair for physical equipment like ATMs is a low-digitization, hardware-centric sector with slow AI adoption compared to fully digital, information-based industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistants can significantly augment technician productivity by auto-populating schematics, suggesting design modifications, automating routine data entry, and managing component databases. A human technician in the loop benefits substantially from AI-assisted drafting and information lookup tools. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted CAD tools and diagnostic software can speed up data entry, schematic editing, and program transfer tasks, providing solid but not transformative productivity gains for the technician. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can reliably perform data entry, program copying, schematic drawing, modification, and storage tasks with high fidelity. Current tools can automate most of the underlying operations (file transfers, CAD software automation, database population) with minimal human intervention, achieving well over 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 3/5 | Data entry and copying programs between components can be scripted or automated, but drawing/modifying schematics for physical hardware still requires human interpretation of device-specific wiring and layouts, limiting full end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | This is primarily a data and software operation with minimal regulatory constraints. Some tasks may require physical access credentials or equipment authorization, but the information-handling portion faces minimal legal or licensing barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No formal licensing generally required, but liability for miscopied firmware or faulty schematics in financial/office equipment creates moderate risk-based friction and need for human verification. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven automation of data entry, file copying, and schematic storage is vastly cheaper than human technician labor per task instance. The computational cost of inference and storage is orders of magnitude lower than the loaded wage of a repair technician. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Specialized technician tools plus AI assistance still require human setup, verification, and physical access, so cost savings versus a technician's loaded wage are modest rather than order-of-magnitude. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature products exist for automated data entry, schematic CAD tools with programmatic APIs, and electronic component programming. While some edge cases (legacy hardware formats, non-standard interfaces) may require human oversight, production systems reliably handle the routine core of this task. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Software tools exist for CAD/schematic editing and firmware flashing, but there is no deployed product that autonomously performs technician-style program copying and schematic modification across diverse ATM/office machine hardware today. |
Read specifications, such as blueprints, charts, or schematics, to determine machine settings or adjustments.
54CI 35–72 · exposure 50 · augmentation 75 · importance 3.8/5 · click for rater detail
Read specifications, such as blueprints, charts, or schematics, to determine machine settings or adjustments.
54| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Field service, manufacturing maintenance, and IT repair sectors are beginning to adopt AI-assisted specification reading (e.g., mobile apps with document-upload features), but deployment remains patchy and pilots outnumber deeply integrated production systems. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field service and repair technician trades are a laggard sector for AI adoption compared to information/finance industries, with pilots for AI-assisted diagnostics still uncommon in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at rapidly parsing, highlighting, and recommending adjustments from complex technical documents, dramatically reducing the time technicians spend hunting through manuals and enabling faster, more accurate work with human oversight intact. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist technicians by quickly parsing complex schematics, cross-referencing manuals, and suggesting settings, improving speed and accuracy while the technician still performs the physical work. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems can reliably extract, interpret, and act on technical specifications from blueprints, schematics, and charts with high accuracy, saving substantial time compared to manual reading and manual lookup of settings. However, complex or degraded documents and final judgment calls on edge cases may still require human verification. |
| Task automatability | claude-sonnet-5 | 2/5 | AI vision-language models can interpret many blueprints or schematics to extract settings, but translating that into precise physical machine adjustments requires hands-on verification that current systems cannot fully replace.But partial reading/interpretation can be automated with significant setup and validation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing requirement or legal mandate requires a human to read specifications; adoption is primarily constrained by organizational familiarity and integration friction, not regulatory barriers. Customer preference for technician oversight is modest. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human read schematics, but practical barriers exist since physical machine adjustment still requires a technician present, limiting how much automation replaces this specific reading task alone. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference, vision model queries, and integration are now inexpensive (pennies per image), whereas a skilled technician reading and interpreting specifications costs $20–$50+ per hour loaded; the cost ratio heavily favors AI. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI document interpretation is cheap, the overall task still requires a technician on-site to apply adjustments, so cost savings are limited to the reading/interpretation sub-step only. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed computer vision and OCR products (including GPT-4V, Claude Vision, and specialized document-parsing tools) demonstrably perform specification reading and interpretation reliably in production settings, including for technical drawings and schematics. Minor gaps exist in handling extremely old or non-standard formats. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some products (CAD-reading AI, technical document parsers) exist but are not widely deployed specifically for ATM/office machine repair technicians reading schematics to set physical adjustments in the field. |
Advise customers concerning equipment operation, maintenance, or programming.
50CI 41–59 · exposure 42 · augmentation 75 · importance 4.3/5 · click for rater detail
Advise customers concerning equipment operation, maintenance, or programming.
50| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | IT service and repair sectors show middling adoption of AI advisors; some organizations have deployed chatbots and knowledge systems for tier-1 support, but human technicians remain primary for complex advice, and uptake varies widely by company size and risk tolerance. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Field service and hardware repair sectors are moderately digitizing with AI-assisted support tools and knowledge bases, but adoption lags behind pure information/professional-service sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI substantially augments technician productivity by retrieving relevant documentation, suggesting troubleshooting steps, and handling routine customer inquiries, allowing human advisors to focus on complex or edge-case problems where judgment and contextual knowledge are essential. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can effectively support technicians by providing quick access to manuals, troubleshooting guides, and diagnostic suggestions, significantly speeding up advisory conversations with customers. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can provide generic troubleshooting guidance and documentation-based answers, the task requires understanding customer-specific equipment configurations, prior maintenance history, and real-time diagnostic interaction—demanding contextual judgment that current systems cannot reliably deliver end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 3/5 | AI chatbots can answer common operational and maintenance questions from manuals, but nuanced troubleshooting of specific hardware faults or programming issues still requires human diagnostic judgment and physical inspection. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Moderate friction exists: organizations may require human sign-off on critical advice due to liability concerns, customer preference for human contact on complex issues, and the need for oversight to catch AI hallucinations or incomplete guidance that could damage equipment or lead to costly downtime. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for advising customers on office equipment, though liability concerns exist if bad advice causes equipment damage or downtime, creating some caution around full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-driven customer service and knowledge systems are substantially cheaper to scale than hiring and training technical advisors; the per-interaction cost of inference and knowledge retrieval is orders of magnitude lower than loaded technician wages, especially for high-volume routine queries. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI chat support is cheap per interaction, but when combined with human escalation for unresolved issues, overall cost savings versus a technician's advisory time are moderate rather than dramatic. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed chatbots and knowledge-base systems can handle routine customer inquiries about standard equipment operation and maintenance, but they struggle with edge cases, non-standard configurations, and the interactive troubleshooting needed to diagnose complex problems reliably in production settings. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | AI-powered support chatbots and virtual assistants are deployed in tech support for equipment guidance, but accuracy on complex/custom equipment issues remains inconsistent and often escalates to humans. |
Converse with customers to determine details of equipment problems.
43CI 35–50 · exposure 30 · augmentation 63 · importance 4.4/5 · click for rater detail
Converse with customers to determine details of equipment problems.
43| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Many repair service companies have piloted chatbots and phone IVR systems for initial troubleshooting, but full replacement of human diagnostic conversation remains uncommon in production. Adoption is middling—pilots exist widely, but sustained production use at scale is still limited. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | placeholder |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can effectively assist technicians by transcribing customer descriptions, suggesting likely issues based on symptoms, generating initial diagnostic checklists, and organizing information—all while the technician remains in control of the conversation. This augmentation meaningfully raises technician productivity in the diagnostic phase. |
| Augmentation potential | claude-sonnet-5 | 3/5 | placeholder |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can engage in basic conversational exchanges and gather information through dialogue, reliably extracting nuanced equipment problem details from customers requires understanding context, recognizing vague descriptions, and disambiguating technical issues—tasks where AI often misses critical details or makes unfounded assumptions. Current AI falls short of the 50% time-saving threshold for this customer-facing diagnostic conversation. |
| Task automatability | claude-sonnet-5 | 2/5 | While AI can conduct diagnostic conversations via chatbots, this task is embedded in an in-person field repair workflow where verbal triage often happens on-site or requires quick back-and-forth tied to physical inspection, limiting full automation.", "rating adjusted low.":true}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are minimal regulatory or legal barriers to automating customer conversations; most repair services can operate chatbots without licensing restrictions. However, customers often prefer speaking to a human technician, and organizational friction around customer satisfaction may slow adoption. |
| Adoption barriers | claude-sonnet-5 | 2/5 | placeholder |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference and integration costs for a chatbot or agent are very low compared to the loaded wage of a repair technician having a diagnostic conversation. Even accounting for oversight and integration, the cost per customer interaction is substantially below human labor cost. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | placeholder |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Chatbots and AI assistants can handle routine customer inquiries and basic troubleshooting scripts in production, but they struggle with edge cases, unusual problems, and customers who lack technical vocabulary. Products exist but lack the reliability to fully replace human technicians for the diagnostic conversation phase. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | placeholder |
Test components or circuits of faulty equipment to locate defects, using oscilloscopes, signal generators, ammeters, voltmeters, or special diagnostic software programs.
30CI 25–35 · exposure 25 · augmentation 75 · importance 3.7/5 · click for rater detail
Test components or circuits of faulty equipment to locate defects, using oscilloscopes, signal generators, ammeters, voltmeters, or special diagnostic software programs.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Equipment repair remains a fragmented, largely small-business sector with low digital infrastructure maturity. While some large service providers use diagnostic software, the sector has been slow to adopt AI-driven automation compared to information or financial services, and technician employment remains stable. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field repair and hardware maintenance sectors show slower AI adoption compared to information/professional services, with diagnostic AI tools used as aids rather than replacements. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI diagnostic software can substantially assist technicians by analyzing signal data, suggesting fault hypotheses, and highlighting anomalies—allowing them to work faster and more accurately while they interpret results and make physical measurements. This is a strong use case for AI-assisted problem-solving within a human-led workflow. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered diagnostic software can significantly speed up defect localization by analyzing sensor data, error logs, and historical failure patterns, meaningfully augmenting a technician's troubleshooting process. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI-powered diagnostic software can suggest potential faults by analyzing signals, a human technician must physically handle oscilloscopes, connect test equipment, and interpret nuanced waveforms in context—tasks requiring hand-eye coordination and spatial reasoning that current AI cannot perform end-to-end. The diagnostic reasoning is partially automatable, but the physical testing and equipment manipulation remain fundamentally manual. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical manipulation of test equipment (probes, meters) on real hardware, which current AI systems cannot perform; software-based diagnostics assist but full end-to-end automation isn't achievable today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Repair work on customer equipment carries liability and warranty risks; errors can damage hardware or create safety hazards. Organizational practices and customer expectations typically require a qualified technician to perform and sign off on diagnostics and repairs, creating meaningful friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing requirement typically gates this repair work, but physical access to equipment and liability for improperly diagnosed repairs create some friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI diagnostic tools can reduce analysis time, but the technician's labor for physical testing, equipment setup, and verification remains substantial. The combined cost of software licensing, integration, and human oversight does not achieve cost parity with traditional diagnostic labor, let alone advantage. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI diagnostic software is cheap to run, but since it cannot physically operate test equipment or replace the technician's hands-on work, the human labor cost remains largely unavoidable, making all-in AI substitution costly relative to partial benefit. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some diagnostic software exists that analyzes circuit signals and suggests fault locations, but these are narrow-scope tools requiring human direction and verification. No deployed system reliably performs the full testing workflow (equipment setup, measurement, interpretation, and decision-making) without significant human oversight and physical manipulation. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Diagnostic software exists and is used to flag error codes or suggest likely faults, but the physical testing with oscilloscopes/meters and interpretation of readings in context still requires a human technician in production settings. |
Train new repairers.
30CI 25–35 · exposure 25 · augmentation 63 · importance 3.1/5 · click for rater detail
Train new repairers.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Technical repair training remains dominated by human craftspeople and formal apprenticeships. While some manufacturers experiment with digital aids, deep production adoption of AI-led technician training is rare; the sector values proven human expertise over automated instruction due to safety and equipment variation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | This is a niche physical repair trade with low digitization and slow AI adoption relative to information-sector occupations. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist trainers by generating customized video tutorials, creating adaptive practice modules, tracking learner progress, and producing documentation—raising trainer efficiency. However, the core value (hands-on feedback, judgment calls, safety sign-off) remains human-dependent. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist with creating training manuals, quizzes, simulations, and answering technical questions, enhancing trainer efficiency and consistency. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Training new repairers involves hands-on skill transfer, assessment of competency, and adaptive feedback based on individual learner progress. While AI can generate training materials and video content, it cannot reliably supervise physical practice, evaluate practical proficiency in real-time, or provide the mentoring relationship required for this role, falling well short of 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Training new repairers involves hands-on demonstration, mentorship, and adaptive coaching based on trainee performance, which current AI cannot fully replicate end-to-end.produce |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory requirements often mandate that trainers hold industry certifications (vendor-specific, electrical safety). Organizations have strong preference for human mentorship in technical trades due to safety liability and the tacit knowledge transfer required for equipment variation. Institutional inertia in apprenticeship models provides substantial adoption friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically for training delivery, but organizational reliance on experienced technicians and hands-on skill transfer creates moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI training systems require significant setup and customization for specific equipment types, plus human oversight to verify learning outcomes and safety compliance. The total cost remains comparable to or higher than employing experienced repairers as trainers, especially when quality assurance is included. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI can cheaply generate training materials, the hands-on supervision and correction needed still requires human trainers, keeping overall cost comparable to human-led training. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI-powered training platforms and CBT systems exist, but they struggle with assessing practical hands-on skills and real equipment troubleshooting in field conditions. No deployed product reliably replicates the full scope of technician trainer work including hands-on guidance, safety oversight, and adaptive re-training. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some AI-based training content generation and e-learning modules exist, but no deployed product reliably substitutes for hands-on mentorship and skills assessment in this trade. |
Operate machines to test functioning of parts or mechanisms.
28CI 21–35 · exposure 20 · augmentation 50 · importance 4.2/5 · click for rater detail
Operate machines to test functioning of parts or mechanisms.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption is slow in repair and maintenance sectors, which remain highly fragmented among small firms and service technicians. Most organizations still rely on human expertise rather than automated testing systems, even in moderately digitized contexts. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Repair trades are a low-digitization, physically embodied sector where AI adoption for hands-on diagnostic tasks remains minimal and mostly limited to software-based diagnostic aids. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Diagnostic software and automated test logs can assist technicians by flagging anomalies and guiding them through testing procedures. AI can augment human decision-making without replacing the core manual and interpretive work required to operate and evaluate machine functionality. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered diagnostic software and expert systems can help interpret test results, suggest likely faults, or guide troubleshooting steps, improving technician efficiency even though the physical operation itself isn't automated. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Testing machine functioning requires physical operation of hardware and sensory interpretation of results. While AI vision systems can assess outputs and logs, the hands-on mechanical operation and real-time troubleshooting of complex mechanisms remains difficult to automate end-to-end without significant human intervention. |
| Task automatability | claude-sonnet-5 | 2/5 | Operating physical machines to test functioning requires physical manipulation, plugging in, running diagnostics, and observing real-world hardware behavior that current AI cannot perform without robotic embodiment.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | ATM and office machine repair often requires on-site work with customer equipment and legal responsibility for proper diagnosis. Some regulatory or contractual requirements exist that favor human accountability, though these are not absolute legal barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement typically applies, but the physical nature of the task and need for hands-on diagnostic judgment create practical barriers to remote or software-only substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI systems for machine testing still require significant human oversight, sensor setup, and interpretation. The all-in cost of AI-driven testing equipment remains comparable to or often exceeds the cost of a trained technician performing the work directly. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical labor involved, so a human technician remains necessary and cost comparison favors the human by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Computer vision and diagnostic software exist to aid testing, but no mature product reliably performs full end-to-end mechanical testing and diagnostics at scale in production. Most deployed systems assist rather than replace human technicians who must physically operate machines and interpret nuanced failure modes. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically operates ATMs, copiers, or office machines to test parts today; this remains firmly in the physical/robotics research stage. |
Calibrate testing instruments.
28CI 21–35 · exposure 20 · augmentation 50 · importance 3.2/5 · click for rater detail
Calibrate testing instruments.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of AI-assisted calibration tools is emerging in high-tech sectors (semiconductors, aerospace) but remains slow in broader service and maintenance industries. Many repair shops still rely on traditional manual calibration methods with minimal automation integration. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field repair and hardware maintenance sectors show slow AI adoption for physical tasks, with automation efforts concentrated on diagnostics and scheduling rather than hands-on calibration. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by automating diagnostics, recommending calibration settings, documenting results, and flagging outliers, helping technicians work faster and with greater accuracy. However, the human technician remains essential for executing the physical adjustments and final sign-off. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by providing calibration guidelines, interpreting sensor readings, logging results, and flagging deviations, improving technician efficiency even though it doesn't perform the physical calibration itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Calibration requires precise physical manipulation and sensor adjustment, often in situ, combined with measurement interpretation. While AI can assist in diagnosis and recommend calibration parameters, current systems cannot reliably perform the hands-on adjustment and verification steps autonomously. |
| Task automatability | claude-sonnet-5 | 2/5 | Calibrating physical testing instruments requires hands-on manipulation, adjustment, and verification against physical standards that current AI systems cannot perform without robotic embodiment, which is not standard in this field today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Calibration often requires domain-specific certification or compliance with standards (ISO, regulatory bodies), and liability for instrument accuracy creates organizational friction. Some sectors enforce that certified humans must validate critical calibrations, though barriers are not absolute. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing mandates a human specifically calibrate instruments, but the physical nature of the task and need for tactile verification create practical (not legal) barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-assisted calibration tools reduce overhead through diagnostics and documentation, but the physical manipulation and verification still require trained human technicians. Total cost savings are modest compared to the technician's loaded wage when accounting for equipment and liability. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Without a viable AI-driven physical solution, the human technician remains the only cost-effective option; any AI-assisted approach would require expensive robotics/sensors exceeding the cost of human labor for this narrow task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some specialized calibration software and diagnostic tools exist, but end-to-end autonomous calibration of diverse testing instruments is not deployed in production. Most real-world calibration still requires human technicians to perform critical physical adjustments and final verification. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | There are no deployed AI products that autonomously calibrate physical test equipment for ATM/office machine repair; this remains a manual technician task requiring physical dexterity and judgment. |
Install and configure new equipment, including operating software or peripheral equipment.
26CI 21–30 · exposure 20 · augmentation 63 · importance 4.0/5 · click for rater detail
Install and configure new equipment, including operating software or peripheral equipment.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Equipment vendors offer some automated deployment tools and configuration management platforms, but adoption remains slow in legacy sectors (banking, ATM networks). Most organizations still rely on field technicians rather than AI-driven setup. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field service and repair trades are slow to adopt AI-driven automation for hands-on installation work compared to office/information-sector tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistants can substantially boost technician productivity by automating configuration templates, checking compatibility, generating installation checklists, and providing real-time troubleshooting guidance—allowing humans to focus on physical placement and testing. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with configuration scripts, software setup guidance, diagnostics, and troubleshooting documentation, improving efficiency on the software/configuration portion of the task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with configuration steps and documentation, physical installation requires manual labor and real-time adaptation to site conditions. Software setup can be partially automated, but hardware placement, cabling, and hardware-software integration verification typically need human hands and contextual judgment. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a physical installation and hardware configuration task requiring on-site presence, cabling, mounting, and hands-on setup that current AI systems cannot perform end-to-end.}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Some installations are regulated (financial ATMs, medical devices), and client sites often require licensed technicians for sign-off on hardware setup. However, configuration documentation and software setup face fewer strict barriers. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing barrier per se, but physical access, security protocols for financial equipment (ATMs), and vendor-specific hardware knowledge create real friction to remote or automated substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI can reduce time on documentation and configuration scripting, but the physical labor and on-site validation remain expensive human work. Overall cost savings are modest because the most time-consuming bottleneck—physical installation—cannot yet be automated. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical labor involved, so the all-in cost of an AI-only solution is not comparable; a human technician remains the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably handles end-to-end equipment installation and configuration autonomously; AI excels at guiding installation steps or troubleshooting software, but actual physical setup and full commissioning remain manual tasks in production environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product physically installs and configures ATMs, office machines or peripheral hardware; this remains a manual field-service task. |
Reinstall software programs or adjust settings on existing software to fix machine malfunctions.
25CI 20–30 · exposure 20 · augmentation 63 · importance 4.2/5 · click for rater detail
Reinstall software programs or adjust settings on existing software to fix machine malfunctions.
25| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of autonomous repair agents in ATM and office machine service remains in pilot stages; most field service organizations still rely on human technicians with remote diagnostic support, reflecting slow penetration of true autonomous repair automation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field service and hardware repair sectors show slower AI adoption compared to information/professional services, with AI mainly used for remote diagnostics and knowledge support rather than replacing technicians. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered diagnostic tools and remote monitoring systems can assist technicians by suggesting likely causes and recommended fixes, streamlining their troubleshooting workflow. However, the human technician remains essential for verification, physical intervention, and handling unexpected edge cases. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered diagnostic assistants, troubleshooting chatbots, and knowledge bases can meaningfully speed up a technician's identification of software issues and recommended fixes, improving productivity while the human still performs the physical reinstall/adjustment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Reinstalling software and adjusting settings can be partially automated for straightforward, well-documented cases, but diagnosis of the underlying malfunction and context-specific troubleshooting typically require human judgment. Current AI lacks reliable ability to fully autonomously resolve diverse machine malfunctions without human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | This task requires physical presence at the machine, diagnostic judgment, and hands-on adjustment, which current AI cannot perform end-to-end despite being able to assist with diagnostic steps or documentation lookup.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Repair of financial machines (ATMs) and business-critical equipment typically involves vendor support contracts, liability and warranty considerations, and customer trust requirements that legally or contractually bind tasks to certified human technicians. Regulatory and organizational friction against full automation is substantial. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement, but liability concerns around malfunctioning ATMs (financial transactions) and organizational reliance on trained technicians create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI-assisted tools and remote monitoring systems require significant infrastructure integration and oversight costs, while technician wages remain relatively modest. The all-in cost of automating this task does not yet undercut human labor materially. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could reduce diagnostic time via troubleshooting guidance, but the physical repair labor and on-site presence still require a human, so all-in cost savings are modest rather than order-of-magnitude. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While basic software reinstallation and some settings adjustments are documented in support tools, deployed AI systems rarely perform this task end-to-end reliably in production environments. Narrow automation exists for specific known issues, but real-world deployment for general ATM and office machine repair is minimal. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously reinstalls software or adjusts settings on physical ATMs/office machines in production; remote IT support tools exist for general computers but not for this specialized hardware-software repair context. |
Test new systems to ensure that they are in working order.
24CI 16–31 · exposure 20 · augmentation 50 · importance 4.1/5 · click for rater detail
Test new systems to ensure that they are in working order.
24| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of AI for hardware system testing remains slow; most organizations still rely on human technician validation despite digitization trends. Production deployment of autonomous testing agents for ATMs and office machines is rare, with adoption concentrated in narrow software testing domains rather than equipment commissioning. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Repair and field service sectors have historically slow AI adoption due to the physical nature of work, though diagnostic software aids are increasingly used. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist technicians by automating test plan generation, analyzing system logs for anomalies, and flagging potential failures—improving productivity on diagnostic and documentation portions. However, the human technician remains essential for physical verification and final sign-off, so augmentation is meaningful but partial. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-based diagnostic software and troubleshooting guides can assist technicians in identifying issues faster, though the physical testing and verification remain human-driven. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Testing new systems requires observing physical hardware operation, identifying edge cases, and diagnosing failures—tasks that often demand manual interaction and real-world validation. While AI can assist with test design and log analysis, end-to-end automated testing of ATMs and office machines still requires human technicians to physically verify functionality across varied hardware configurations, making consistent 50% time savings at equal quality unlikely today. |
| Task automatability | claude-sonnet-5 | 2/5 | Testing hardware systems like ATMs or office machines requires physical interaction, diagnostic equipment, and manual verification that AI cannot perform end-to-end; software-only test automation is not the bulk of this task.rateur |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory and liability barriers are substantial: equipment manufacturers typically require certified technicians to validate systems before deployment, and errors in testing can result in ATM outages or safety hazards. Warranty and compliance requirements create strong incentives to maintain human sign-off on critical test results. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing requirement typically governs this testing, but physical access, liability for equipment functioning, and specialized diagnostic tools create moderate friction against remote AI substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The AI inference, integration setup, and human oversight for testing complex hardware systems still exceeds the cost of deploying a skilled technician to perform direct functional tests, given the capital cost of test rigs and the liability of missed failures. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical labor and equipment access needed, so a human technician remains necessary and cost-comparable or cheaper than any AI-robotic hybrid solution. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Automated testing frameworks exist for software components, but testing physical machinery (ATMs, copiers, etc.) in working order requires hands-on hardware verification that current AI systems cannot reliably perform independently. No deployed product systematically tests entire new ATM or office machine systems autonomously at scale with acceptable error rates. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously physically tests hardware systems like ATMs or office equipment in the field today; this remains a hands-on technician task. |
Align, adjust, or calibrate equipment according to specifications.
22CI 14–30 · exposure 20 · augmentation 50 · importance 4.2/5 · click for rater detail
Align, adjust, or calibrate equipment according to specifications.
22| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Adoption of full automation in ATM and office machine repair remains minimal; these sectors are geographically dispersed, serve varied equipment, and maintain strong technician-based service models. Adoption velocity is slow outside high-volume, standardized repair centers. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Field service and repair trades adopt AI slowly, mostly for diagnostics or scheduling support rather than replacing physical calibration work. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by interpreting calibration specifications, suggesting adjustment sequences, or flagging out-of-spec conditions via diagnostic analysis, helping technicians work more efficiently while they perform the hands-on adjustment. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered diagnostic tools and manuals/chatbots can guide technicians on calibration steps and specifications, improving efficiency without performing the physical task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can interpret specifications and guide alignment procedures, physical adjustment and calibration require dexterous manipulation and real-time sensory feedback that current robotic systems handle only in narrow, heavily structured scenarios. The task is not yet automatable end-to-end at 50% time savings with general-purpose systems. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a physical hands-on calibration task requiring manual manipulation of hardware and sensors, which current AI cannot perform end-to-end; at most AI can guide the technician via diagnostics.rt |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory requirements, warranty liability, and customer trust create meaningful barriers: manufacturers often mandate certification or authorized technician sign-offs on calibration, and misalignment can cause costly downtime or safety issues, raising error-cost asymmetry. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement, but physical access, liability for miscalibration causing equipment failure, and lack of robotic actuation create practical friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current robotic systems capable of precision alignment are capital-intensive and require specialized integration. For typical ATM or office machine repairs, the all-in cost of a robotic solution would significantly exceed the loaded wage of a technician performing the work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI systems cannot substitute for the physical labor and dexterity involved, so there is no viable AI cost comparison—the human remains necessary. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed products reliably perform full equipment alignment and calibration autonomously in production. Specialized industrial robots exist for specific machines, but they are narrowly scoped, require extensive setup, and lack the adaptive capability to handle the variety of ATM, office, and computer equipment this role covers. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed products physically align or calibrate ATM/office machine hardware autonomously today; this remains a manual technician task. |
Update existing equipment, performing tasks such as installing updated circuit boards or additional memory.
21CI 15–26 · exposure 8 · augmentation 50 · importance 3.7/5 · click for rater detail
Update existing equipment, performing tasks such as installing updated circuit boards or additional memory.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Equipment repair remains heavily manual and concentrated in small-to-mid-sized service shops with low technology adoption rates. Large enterprises may use AI diagnostics to decide *what* to upgrade, but actual installation remains performed by technicians. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field equipment repair is a low-digitization, physical-labor sector with minimal AI/robotic adoption for actual hands-on hardware installation tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by providing step-by-step installation guides, compatibility checking, and post-installation diagnostics, improving efficiency and error-checking. However, the physical installation task itself is still performed by the human technician with AI providing informational support. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist technicians via diagnostic guidance, documentation lookup, and troubleshooting recommendations, though the physical installation itself remains unassisted by AI. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can theoretically guide hardware installation steps, the task requires precise physical manipulation in real equipment environments. Current AI systems lack embodied robotics integration for reliable circuit board or memory installation, and domain-specific troubleshooting before/after installation remains human-dependent. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical hands-on task requiring opening equipment, handling components, and physically installing hardware, which current AI systems cannot perform without robotic embodiment far beyond deployed capability.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While equipment manufacturers may have warranty and liability concerns about non-authorized repairs, the task itself has few regulatory/licensure barriers. Equipment OEMs sometimes restrict authorized technician pools, but these are business rather than legal barriers. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing typically required for this repair work, but physical access, liability for equipment damage, and lack of robotic actuation create practical barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A trained human technician performs this task in under an hour for most upgrades. The capital cost of robotic systems capable of safe hardware installation, plus software integration and maintenance, far exceeds the technician's loaded hourly wage. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing the physical labor, so any AI cost comparison is moot—human technicians remain the only cost-effective option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed products reliably perform physical hardware updates (circuit board/memory installation) end-to-end in production repair settings. Robotic manipulation for delicate electronics repair exists only in research or highly specialized, bespoke scenarios, not as general commercial offerings. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically installs circuit boards or memory upgrades in field equipment; this remains entirely a research/robotics frontier for unstructured physical tasks. |
Assemble machines according to specifications, using hand or power tools and measuring devices.
20CI 10–30 · exposure 8 · augmentation 38 · importance 3.7/5 · click for rater detail
Assemble machines according to specifications, using hand or power tools and measuring devices.
20| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Assembly automation exists primarily in high-volume manufacturing (automotive, electronics); specialized machine repair and assembly remains dominated by small repair shops and field technicians with limited adoption of advanced robotics. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | This occupation involves hands-on physical repair work in a sector with low digitization and automation penetration; robotic adoption for such tasks remains minimal and slow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted tools (AR-guided assembly instructions, specification parsing, tool recommendations) can improve technician productivity on interpretation and planning tasks, though the core assembly work remains human-driven. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can provide some assistance via digital manuals, diagnostic guidance, or AR overlays, but it doesn't materially transform the physical assembly process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Assembly of physical machines requires dexterous manipulation, spatial reasoning, and real-time adaptation to tolerance variations that current AI systems cannot reliably perform end-to-end. While AI can interpret specifications and guide humans, autonomous robots capable of general-purpose assembly at equal quality to humans remain research-stage. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical assembly of machines using hand/power tools requires manipulation and fine motor skills that current AI and robotics cannot perform reliably outside narrow factory settings, especially for varied repair/assembly contexts. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Assembly work typically requires physical presence and certification for certain equipment (e.g., medical devices, safety-critical machines), creating moderate friction. However, no strict licensing requirement uniformly prevents automation, though quality standards and liability concerns slow adoption. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement dictates a human must assemble machines, but physical dexterity, judgment on varied equipment, and lack of standardized environments create practical barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialized assembly robots for narrow tasks can be cost-competitive with humans in high-volume production, but for varied machine types and specifications, integration, programming, and maintenance costs exceed the loaded wage of a skilled technician. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Deploying robotics/automation for this variable, low-volume assembly work would be far more expensive than a technician's wage due to lack of standardization and need for custom tooling. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs complete machine assembly from specifications in production environments. Robotic arms exist for narrow, highly repetitive assembly tasks, but generalist assembly requiring interpretation of technical drawings, tool selection, and quality judgment lacks mature commercial solutions. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product exists that autonomously assembles machines to spec in field repair or general shop settings; robotic assembly is confined to fixed, high-volume manufacturing lines, not this occupation's variable tasks. |
Fill machines with toners, inks, or other duplicating fluids.
17CI 10–24 · exposure 8 · augmentation 13 · importance 3.1/5 · click for rater detail
Fill machines with toners, inks, or other duplicating fluids.
17| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task occurs in dispersed office and field environments with low digital integration; repair and maintenance sectors have historically lagged in automation adoption, and the physical, on-site nature of the work limits the applicability of current AI and robotic systems. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Repair and maintenance of office equipment is a physical, low-digitization trade with minimal AI/robotic adoption for hands-on consumable replacement tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers minimal assistance for the core mechanical act of filling machines with fluids; a technician performing this task would not benefit meaningfully from AI guidance, vision assistance, or decision support in routine refill operations. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help via diagnostic apps or manuals guiding technicians on machine-specific fluid types and procedures, but this offers only marginal assistance to the physical act itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task involves physical manipulation—opening machines, inserting fluids, securing components—that requires dexterity and spatial reasoning in unstructured environments. Current robotics can handle highly controlled refilling in factories, but general-purpose machines in diverse office settings remain out of reach for end-to-end automation at 50% time savings. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manipulation task requiring hands to open machines, locate reservoirs, and pour or insert consumables; no current AI system can perform physical actions without embodiment, and robotics for this specific task is not deployed. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While there is no explicit licensing requirement for fluid refilling itself, organizational and safety preferences (ensuring proper handling of potentially hazardous fluids, warranty concerns, liability for machine damage) create moderate friction against autonomous substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for refilling consumables, but the task requires physical presence and manual dexterity, creating a natural embodiment barrier rather than a regulatory one. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Custom robotic systems capable of this task are prohibitively expensive compared to the modest loaded wage of a technician performing routine fluid refills, and integration costs remain high for variable machine types. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven substitute for this physical task, so any AI cost comparison is moot; a human technician remains the only cost-effective option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs autonomous machine refilling in production office environments. Specialized industrial robots exist for controlled settings, but they do not generalize to the variety of machine types and configurations encountered by repair technicians in the field. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical refilling of toner or ink in office/ATM equipment; this remains firmly a human manual task with no robotic automation in production. |
Disassemble machines to examine parts, such as wires, gears, or bearings for wear or defects, using hand or power tools and measuring devices.
14CI 5–24 · exposure 8 · augmentation 25 · importance 4.3/5 · click for rater detail
Disassemble machines to examine parts, such as wires, gears, or bearings for wear or defects, using hand or power tools and measuring devices.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task is performed primarily by field service technicians in small to medium firms with low digital infrastructure; adoption of robotic disassembly and inspection remains minimal outside research labs. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field repair of physical office/ATM equipment is a low-digitization, physical-labor sector with minimal AI/robotics adoption for hands-on disassembly tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI-powered vision systems could assist by analyzing images of parts post-disassembly to flag defects, but the core physical disassembly and measurement process must remain human-driven, limiting the scope of meaningful augmentation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with diagnostic guidance, defect pattern recognition from images, or repair documentation lookup, but offers little help with the physical disassembly and tactile inspection itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While visual inspection of machine parts for wear could theoretically be automated with computer vision, the task requires hands-on disassembly using hand and power tools, physical manipulation in varied environments, and expert judgment about defect severity—capabilities current AI systems cannot perform end-to-end without substantial human intervention. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical disassembly and manual inspection of hardware components using hand/power tools, which is a physical manipulation task current AI systems cannot perform end-to-end without embodiment.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong adoption barriers exist due to equipment liability (tools and machinery damage), safety-critical nature (risk of injury from machinery), and regulatory requirements in many sectors for certified technicians to certify repairs and machine condition. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing barrier specifically prevents automation, but physical dexterity requirements, variable machine types, and lack of mature robotic manipulation create strong practical barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Building and maintaining a robotic system capable of safe disassembly, parts handling, measurement, and defect classification would vastly exceed the labor cost of a trained technician performing this work manually. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven robotic system for this task today, so the human technician remains the only cost-effective option; deploying custom robotics would be far more expensive than the technician's wage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously disassemble machines, handle tools, and diagnose wear or defects in production settings; this task remains firmly in the domain of skilled human technicians with specialized equipment and domain knowledge. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs autonomous physical disassembly and hands-on wear inspection of ATM/office machine internals; this remains far outside current robotics deployment for this niche. |
Clean, oil, or adjust mechanical parts to maintain machines' operating efficiency and to prevent breakdowns.
14CI 5–24 · exposure 8 · augmentation 25 · importance 4.2/5 · click for rater detail
Clean, oil, or adjust mechanical parts to maintain machines' operating efficiency and to prevent breakdowns.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This trade remains dominated by small shops and service firms with low digitization; adoption of automation in physical repair work is slow and limited to highly specific manufacturing contexts, not field or general machine repair. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field service and hardware repair are physical, low-digitization sectors with minimal AI agent deployment for hands-on mechanical maintenance tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can assist with diagnostics (identifying what needs maintenance) or documentation, but offers limited augmentation for the core physical work of cleaning, oiling, and adjusting mechanical parts, which requires human hands-on execution. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with diagnostic guidance, maintenance scheduling, or documentation lookup, but offers little direct help with the physical acts of cleaning, oiling, and adjusting parts. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Physical manipulation of mechanical parts (cleaning, oiling, adjusting) requires dexterity, spatial reasoning, and tactile feedback that current AI lacks; robotic arms exist but are task-specific and lack the generalization needed for diverse machines. While diagnostic elements could be partially automated, the core maintenance work remains fundamentally manual. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a manual, physical maintenance task requiring hands-on cleaning, lubrication, and mechanical adjustment of physical components inside machines like ATMs and office equipment; no current AI system can physically perform this work. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Substantial barriers exist: liability for equipment damage, warranty implications, regulatory compliance in certain industries, and customer preference for certified human technicians. Many organizations require licensed personnel to perform critical maintenance. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates a specific credentialed human for basic maintenance, but the physical nature of the task itself is the real barrier rather than regulation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Current robotic systems capable of performing physical maintenance are expensive, require significant integration, and still need human oversight; a skilled technician's loaded wage remains cheaper for most repair scenarios. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical labor involved, so there is no viable AI cost comparison—only a human technician can perform this work, making AI infeasible regardless of cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system reliably performs physical machine maintenance autonomously today. Robotic solutions exist for narrow, controlled scenarios but not as general-purpose tools for the varied mechanical maintenance described. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical cleaning, oiling, or mechanical adjustment of hardware; this remains entirely a manual dexterity task requiring physical presence and tools. |
Reassemble machines after making repairs or replacing parts.
13CI 10–15 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail
Reassemble machines after making repairs or replacing parts.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This occupation remains in small service shops and maintenance departments with limited IT infrastructure; digital-first adoption patterns are rare, and robotics deployment in field repair is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field repair and hardware maintenance sectors show minimal AI-driven displacement of physical assembly tasks; adoption in this specific hands-on function is essentially nonexistent. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Current AI tools (computer vision for part identification, workflow guidance) offer marginal assistance in parts tracking and procedure confirmation, but the core reassembly task itself sees minimal augmentation from deployed AI systems. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with diagnostic guidance, repair manuals, or troubleshooting steps prior to reassembly, but offers little direct help during the physical reassembly process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Reassembly of machines after repair requires dexterous manipulation in 3D physical space, precise part alignment, and real-time tactile feedback. Current AI systems lack the embodied manipulation capabilities to reliably perform this end-to-end without human intervention. |
| Task automatability | claude-sonnet-5 | 1/5 | Reassembling physical machines requires fine motor manipulation, dexterity, and physical presence that current AI systems cannot perform; no software-based AI can execute this hands-on mechanical task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While not legally restricted, significant organizational friction exists: customers typically expect human technician sign-off on complex repairs, and liability concerns around machine reassembly quality create friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically mandates a human for reassembly, but practical barriers like need for physical dexterity and troubleshooting judgment keep this human-dominated, though not due to formal regulation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Specialized robotic systems capable of reassembly (with heavy human oversight and custom programming per machine type) cost far more than the human technician labor they might replace, making economic substitution implausible today. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for this physical task, so any hypothetical robotic solution would be far more costly than a technician's labor given current robotics costs and lack of generalization. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial products reliably perform autonomous machine reassembly in real repair shops. This task remains in research/prototype territory with limited demonstration of production-scale capability. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical reassembly of ATMs, office machines, or computers; robotic manipulation for such varied, unstructured tasks remains research-stage at best. |
Repair, adjust, or replace electrical or mechanical components or parts, using hand tools, power tools, or soldering or welding equipment.
7CI 5–10 · exposure 0 · augmentation 25 · importance 4.2/5 · click for rater detail
Repair, adjust, or replace electrical or mechanical components or parts, using hand tools, power tools, or soldering or welding equipment.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Repair sectors (ATM servicing, office equipment, electrical systems) are distributed, small-scale operations with low digitization and heavy reliance on human technicians. Adoption of autonomous repair robotics remains minimal and experimental. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field repair of physical machines is a low-digitization, hands-on trade with minimal AI/robotic adoption in production settings currently. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can assist marginally through remote diagnostics or documentation search, but most of the task—physical fault isolation, component replacement, and equipment testing—requires hands-on human presence with limited augmentation potential from current systems. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with diagnostic guidance, troubleshooting documentation, or parts lookup, but offers little help with the physical manipulation and repair actions themselves. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical manipulation of components in real-world equipment using hand tools, power tools, and soldering/welding equipment. Current AI systems cannot operate physical tools or perform hands-on repairs in the field, making end-to-end automation impossible. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on physical repair task requiring manipulation of hardware, tools, and equipment in varied real-world configurations, which current AI systems cannot perform without robotic embodiment far beyond today's capabilities. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety regulations, equipment-specific certifications, liability for equipment damage or personal injury, and the need for technician judgment and accountability create substantial organizational and legal barriers to full automation of repair work. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing typically gates this repair work, but liability for faulty repairs, safety with electrical/soldering work, and practical access constraints create moderate friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The hardware cost of a capable repair robot (dexterous manipulator, mobile platform, tools, sensors) would far exceed the loaded wage of a human technician, with integration and ongoing maintenance adding further cost. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven substitute for physical repair labor, so the cost comparison favors the human technician entirely; any robotic solution would be far more costly than a technician's wage today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously diagnose and physically repair electrical or mechanical components. While robotic arms exist in controlled factory settings, they lack the general-purpose dexterity, tool switching, and adaptive problem-solving needed for field repair work on diverse equipment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously repairs or replaces electrical/mechanical components in ATMs or office machines; robotic manipulation for diverse, unstructured repair tasks remains research-stage. |
Travel to customers' stores or offices to service machines or to provide emergency repair service.
5CI 0–10 · exposure 0 · augmentation 38 · importance 4.2/5 · click for rater detail
Travel to customers' stores or offices to service machines or to provide emergency repair service.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task involves physical presence and hands-on work in customer locations, sectors where automation adoption remains low and technicians are still primarily human-dependent. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field service and repair sectors show minimal AI-driven displacement given the physical nature of the work; this is a low-digitization, low-adoption context for full automation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide minor assistance (e.g., diagnostic decision support, remote guidance via video), but the core task of traveling to and physically servicing machines requires human presence and action, limiting meaningful augmentation. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist via diagnostic support, dispatch optimization, route planning, and remote troubleshooting guidance, improving efficiency even though the physical repair and travel remain human-performed. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical presence at customer locations and hands-on intervention with machinery, which current AI systems cannot perform. No automated system can travel to a site and physically service or repair equipment. |
| Task automatability | claude-sonnet-5 | 1/5 | This task inherently requires physical travel and hands-on manipulation of hardware, which current AI systems cannot perform since they lack embodiment or mobility. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong barriers exist: customer authorization and safety requirements necessitate a qualified human technician on-site, liability concerns are high, and regulatory/warranty obligations typically require certified personnel to perform repairs. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no formal licensing typically restricts this work, physical presence, liability for equipment damage, and customer trust in a human technician create practical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Travel and on-site repair require human technicians; there is no AI system that can perform this work, making comparison to human cost irrelevant and AI substantially more expensive if any capability existed. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for physical travel and repair, so any AI cost comparison is moot; a human technician remains the only functional option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product can travel to customer locations or perform physical repair work. This task is entirely outside the scope of current autonomous or robotic systems in production for this industry. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product exists that can physically travel to a customer site and perform repairs; this remains purely a research-stage robotics challenge, not a commercial reality. |
Lay cable and hook up electrical connections between machines, power sources, and phone lines.
5CI 0–10 · exposure 0 · augmentation 25 · importance 3.5/5 · click for rater detail
Lay cable and hook up electrical connections between machines, power sources, and phone lines.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The repair sector remains labor-intensive and on-site dependent. Adoption of AI robotics for physical installation work is virtually non-existent in production; this remains a laggard sector for automation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Field repair and installation work in this occupation is physical, decentralized, and shows minimal AI/robotics adoption compared to information-sector tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could potentially assist technicians via remote guidance or diagnostic troubleshooting before they arrive, but it offers limited real-time support during the physical task itself of laying cable and making connections. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with diagrams, wiring guides, or troubleshooting documentation beforehand, but offers negligible help during the actual physical cable-laying and connection process. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical manipulation in real-world environments (laying cable, making electrical connections), which current AI cannot perform. Robotics for this level of dexterity and environmental adaptation remain experimental and are not deployed at scale. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical task requiring hands-on cable routing, wiring, and connection work in varied real-world environments, which current AI systems cannot perform without embodiment in advanced robotics. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Electrical work is heavily regulated and typically requires licensed electricians to perform or sign off on connections for safety and code compliance. Physical presence and direct accountability create hard legal barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not formally licensed in most jurisdictions, electrical work often requires code compliance, safety training, and liability considerations that create moderate friction against unqualified automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Robotic systems capable of physical installation and electrical work are prohibitively expensive and require extensive setup compared to a trained technician's hourly rate, making AI substantially more costly today. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Without any AI/robotic system capable of this physical installation work, there is no cost comparison to be made—human labor is the only viable option today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI products perform end-to-end physical cable installation and electrical hookup reliably. This work requires human hands-on presence on-site and real-time problem-solving. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously lays cable and makes electrical hookups between machines and infrastructure; this remains a manual, physically-skilled task performed by technicians. |
Related occupations — Installation, Maintenance & Repair
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.