Locksmiths and Safe Repairers
49-9094.00Median wage $51,320/yr15,040 employed (US)Rank #816 of 923 scored · top 88% by substitution
Repair and open locks, make keys, change locks and safe combinations, and install and repair safes.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution14
Exposure8
Augmentation23
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
14 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
7%
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%9
panel mean rating 1.4/5 → substitution pressure 9/100
Technical feasibility todayw 20%7
panel mean rating 1.3/5 → substitution pressure 7/100
Cost vs. human wagew 15%11
panel mean rating 1.4/5 → substitution pressure 11/100
Adoption barriersw 20%inverted — strong barriers lower the score38
panel mean rating 3.5/5 (barrier strength) → substitution pressure 38/100
Sector adoption velocityw 10%5
panel mean rating 1.2/5 → substitution pressure 5/100
Task breakdown (14 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.
Keep records of company locks and keys.
73CI 65–81 · exposure 70 · augmentation 75 · importance 4.2/5 · click for rater detail
Keep records of company locks and keys.
73| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Locksmith and safe repair firms are typically small, manually-operated businesses with moderate digitization. While some larger security firms use management software, uptake of AI-driven record automation in this sector remains pilot-stage rather than widespread production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Locksmithing is a small-business, physical-trade sector with generally low digitization and slow AI adoption, even for back-office tasks like record-keeping. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can dramatically assist locksmiths by automatically logging and categorizing lock and key records during routine work, flagging duplicates, and generating reports—raising productivity without requiring the human to leave the task. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered inventory and database tools can significantly streamline tracking, searching, and updating lock/key records, improving accuracy and speed for the human record-keeper. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Record-keeping of locks and keys is a structured, repeatable task involving data entry, categorization, and storage—well within current AI capabilities. AI systems can extract key details from images or documents, organize them into databases, and maintain searchable inventories with minimal human intervention, achieving >50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 4/5 | Maintaining lock/key records is a structured data-entry and database management task that off-the-shelf software (spreadsheets, key-management systems, or AI-assisted database tools) can largely automate with modest setup. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing or legal requirement mandates human record-keeping of locks and keys. The primary friction is organizational inertia and the desire to keep sensitive security documentation under human supervision, but neither is a hard adoption barrier. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing or legal requirement that a human locksmith personally maintain these records; it's purely administrative and freely automatable. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automating record-keeping with cloud databases and OCR is extremely cheap compared to the hourly wage of a locksmith or administrative assistant managing the same inventory manually. The per-task cost is orders of magnitude lower. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Digital record-keeping tools and simple database/AI assistants cost very little compared to a technician's time spent manually logging and updating records. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature document management and inventory systems with OCR and database automation are widely deployed in production across retail, security, and facilities management. Systems reliably ingest, classify, and store records with acceptable error rates for standard locksmith inventory tasks. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Key-tracking software products exist and are used in facilities management, but many locksmiths still rely on manual logs or basic spreadsheets rather than AI-driven systems, so reliable AI-specific deployment is moderate. |
Cut new or duplicate keys, using key cutting machines.
23CI 11–35 · exposure 13 · augmentation 38 · importance 4.5/5 · click for rater detail
Cut new or duplicate keys, using key cutting machines.
23| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Locksmith shops are small, distributed, local businesses with limited digital integration; while key machines are standard tools, adoption of autonomous AI cutting systems is minimal because current technology cannot substitute for licensed human judgment on access authorization and liability. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Locksmithing is a low-digitization physical trade; while self-service kiosks have spread in retail settings, true AI-driven automation adoption in this trade is minimal and slow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Modern key-cutting machines already augment locksmith productivity by automating the repetitive cutting operation, though AI-driven improvements in blank selection, pattern matching, or cut-quality verification could further assist without removing the human from security and authorization decisions. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with inventory management, scheduling, or identifying key codes from photos, but offers little direct assistance to the physical act of cutting keys. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Key cutting involves precise mechanical operation of specialized equipment and handling of varied key blanks, which can be partially automated by modern key-cutting machines, but these still require human setup, blank selection, and quality verification. Current AI cannot reliably handle the full sensorimotor loop end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 1/5 | Cutting keys requires physical machine operation, precise manual measurement, and handling of physical stock and equipment, which current AI systems cannot perform without robotic embodiment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Locksmiths often require licensing and must verify customer identity and authorization before cutting keys, and liability for creating unauthorized duplicates creates strong legal and regulatory barriers to full automation. The human-in-the-loop requirement for security and liability reasons is substantial. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing generally required for basic key cutting, though safe work may involve security/liability concerns; the main barrier is physical/mechanical rather than regulatory. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Existing key-cutting machines reduce labor time significantly but still require a locksmith's presence for setup, operation, and quality control, making the combined cost of machine maintenance and human supervision comparable to or higher than the task's market wage for a skilled technician. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Automated kiosk machines exist and are cheap per duplicate for common keys, but they are not 'AI' per se and don't handle non-standard or safe-related work, so overall cost comparison to a human locksmith remains mixed. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Automated key-cutting machines exist and are widely deployed in locksmith shops, but they are tools requiring human operation rather than autonomous AI systems. No current AI agent or system can independently select blanks, load machines, and verify cuts without human intervention and oversight. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | There are automated key-cutting kiosks (e.g., in hardware stores) but these are pre-programmed mechanical devices, not AI-driven products, and no deployed AI system autonomously performs this task end-to-end for arbitrary keys including safes. |
Cut new or duplicate keys, using impressions or code key machines.
18CI 9–28 · exposure 13 · augmentation 25 · importance 4.8/5 · click for rater detail
Cut new or duplicate keys, using impressions or code key machines.
18| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmithing is a small, fragmented, highly localized trade with slow digitization. Adoption of new automated technologies is laggard; the sector remains dominated by independent operators and small firms with limited capital investment in advanced automation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing is a low-digitization, physical trade sector with minimal AI adoption; self-service kiosks exist but are decades-old mechanical automation, not part of a modern AI adoption wave. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Current AI could assist in cataloging key types or suggesting blank selection via image recognition, but the physical cutting task and impression reading remain largely manual. Augmentation potential is limited because the operator-machine interface is already optimized and the task requires precise physical control. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with code lookups, inventory management, or diagnosing key codes from photos, but it offers little direct enhancement to the physical cutting process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While key cutting machines are already semi-automated, the task requires human judgment to read impressions, select correct blanks, and verify fit—steps that current AI cannot reliably perform end-to-end without skilled human oversight. AI vision systems cannot yet reliably interpret key impressions or validate cut quality to automotive or high-security standards. |
| Task automatability | claude-sonnet-5 | 1/5 | Cutting keys requires physical manipulation of a key-cutting machine and handling metal blanks, which is a physical manufacturing task current AI systems (software/LLMs) cannot perform without robotics.9There is no general-purpose AI that operates key-cutting machines end-to-end today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Locksmiths often require licensing and must verify customer identity and ownership before cutting keys; liability and security (preventing duplicate key misuse) create substantial legal and regulatory barriers. Many jurisdictions impose legal requirements for human verification of requests. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing law universally requires a human locksmith for key cutting, but liability, security concerns (unauthorized key duplication), and physical possession of blanks/machines create meaningful friction against pure AI/robotic substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | A key-cutting machine (capital cost amortized) plus machine operator remains the standard; an AI vision system capable of reading impressions and controlling the machine would require significant custom integration and validation. The current human + machine cost structure is likely cheaper than building reliable autonomous cutting. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Automated key-cutting kiosks exist and are cheap per unit, but they are specialized hardware investments, not AI cost savings in the inference sense; integrating true AI decision-making offers little marginal cost benefit over existing mechanical automation. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Specialized key-cutting machines exist and are mature, but they are operator-controlled tools, not autonomous AI systems. Current AI has not been deployed in production to autonomously cut keys from impressions or code specifications without a trained locksmith present. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously cuts keys from impressions or codes; existing automated kiosks (e.g., self-service key duplication machines) use fixed mechanical/optical systems, not general AI, and require physical presence and hardware, not software automation. |
Remove interior and exterior finishes on safes and vaults, and spray on new finishes.
14CI 10–19 · exposure 8 · augmentation 13 · importance 1.8/5 · click for rater detail
Remove interior and exterior finishes on safes and vaults, and spray on new finishes.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmiths operate in a laggard sector—small, geographically distributed businesses with limited digitization and high reliance on on-site manual work. Adoption of automation in finish application remains minimal. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing and safe repair is a small, low-digitization physical trade with minimal AI or robotics adoption reported. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with visual inspection of existing finishes or surface preparation planning, but the core stripping and spraying task offers limited augmentation potential since the human must remain physically present to execute the work. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers essentially no assistance for the physical process of stripping and spraying finishes on safes and vaults. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task requires physical manipulation in real-world environments (removing and applying finishes to safes/vaults), which current AI systems cannot perform end-to-end. While vision systems could identify surfaces and plan spray patterns, the execution—stripping, masking, and precise spray application—requires robotic hardware not yet deployed at scale in this domain. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on physical refinishing task requiring manual sanding, spraying, and finish application on heavy metal objects; no AI system can perform this physical work. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | This task faces moderate barriers: safe and vault work often involves security-sensitive environments and customer preference for licensed professionals, though there is no hard legal requirement that a human must personally perform finish work. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing specifically requires a human for refinishing, but the physical dexterity, specialized equipment, and low economic incentive for robotics create practical barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The capital cost of specialized robotic systems capable of finish removal and spray application would far exceed the loaded labor cost of a locksmith performing this task, particularly given the infrequent and variable nature of safe refinishing work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based alternative to compare costs against; a human locksmith with specialized equipment is the only option, making AI infeasible and thus more 'expensive' by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial products reliably perform finish removal and spray application on safes and vaults. While industrial robots exist for some spray tasks, they lack the flexibility to adapt to the varied geometries, existing finishes, and condition variations of different safe models in production locksmith workflows. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed products perform physical stripping and spray finishing of safes; this remains purely a manual trade skill with no robotic automation in production. |
Install alarm and electronic access systems.
12CI 5–19 · exposure 8 · augmentation 38 · importance 3.9/5 · click for rater detail
Install alarm and electronic access systems.
12| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmiths operate in small, local, and often family-owned or independent service contexts with low digital adoption; physical installation trades show historically slow automation adoption and remain geographically dispersed. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | The locksmith/security installation trade is a physical, low-digitization sector with minimal AI/robotics adoption for hands-on installation work. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist with system design, wiring diagrams, configuration recommendations, and troubleshooting guidance, meaningfully improving planning efficiency, though the hands-on installation itself still requires the technician. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with diagnostics, system configuration software, or documentation, but offers little help with the core physical installation task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with diagnostics and system configuration, physically installing hardware, running wiring, and integrating access systems with site-specific security infrastructure requires hands-on mechanical work that current systems cannot perform end-to-end without human intervention. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical installation of hardware, wiring, drilling, and fitting devices at a customer's premises—tasks requiring manual dexterity and on-site physical presence that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Installation of security systems is often subject to licensing requirements, insurance liability for breach-of-system failures, and regulatory compliance (building codes, fire safety standards), meaning human certification and sign-off are typically mandated or expected by law. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not strictly licensed everywhere, many jurisdictions require certification/licensing for alarm and security system installers, plus liability concerns around security breaches create moderate friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI-assisted design or configuration could reduce some overhead, but the core installation work is labor-intensive and site-specific, making total cost per installation significantly higher for AI + robotics than for a trained locksmith. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for the physical labor involved, so the human locksmith remains the only cost-effective option for installation. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product reliably performs full installation of physical alarm and electronic access systems; this task fundamentally requires embodied robotics or on-site technician presence, which remains research-stage for complex, variable installations. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI/robotic product installs alarm or access control systems autonomously; this remains a manual trade job performed by technicians. |
Insert new or repaired tumblers into locks to change combinations.
7CI 0–15 · exposure 0 · augmentation 0 · importance 4.6/5 · click for rater detail
Insert new or repaired tumblers into locks to change combinations.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmiths operate predominantly in small businesses and field service contexts with low digitization. The physical, hands-on nature of lock service and the distributed, on-site nature of the work means adoption of industrial automation has been minimal. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing is a small-scale, physical trade sector with very low digitization and no evidence of AI or robotic adoption for hands-on lock repair work. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers no meaningful assistance for the core mechanical task of inserting tumblers. While digital tools can help schedule appointments or document combinations, they do not augment the hand-assembly work itself. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers essentially no meaningful assistance for the physical act of inserting tumblers into a lock, though it might help with diagnostics or ordering parts unrelated to this specific manual task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires precise mechanical manipulation of small, delicate components in physical space—inserting tumblers into locks—which current AI systems cannot perform without custom robotics. General-purpose AI and even tool-using agents lack the dexterous manipulation capability and real-time physical feedback needed for reliable, quality-equivalent automation. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manipulation task requiring fine motor dexterity to insert small mechanical tumbler components into a lock cylinder; no AI system can perform this physical manual work end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Locksmiths must be licensed or certified in most jurisdictions, and safe repair is often regulated under security and fraud laws. The combination-change work itself frequently requires human authorization, custody of keys, and physical presence for security and liability purposes. |
| Adoption barriers | claude-sonnet-5 | 2/5 | While not typically licensed at a high bar everywhere, locksmithing often requires certification/licensing in many jurisdictions and demands trust due to security implications, but the core barrier is physical/manual rather than regulatory. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Any hypothetical robotic system capable of this task would require significant capital investment, integration, and maintenance costs that far exceed the modest labor cost of a skilled locksmith performing insertions manually. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical task, so any AI-based approach would require entirely new robotic hardware far exceeding the cost of a human locksmith. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial product performs this mechanical assembly task reliably in production. Specialized robotic systems might exist in research settings, but they are not mature, scalable, or generally available products used by locksmiths in the field. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical lock tumbler installation; this remains entirely in the domain of human manual craft skill with no robotic or AI product addressing it in production. |
Install door hardware, such as locks and closers.
7CI 5–10 · exposure 0 · augmentation 25 · importance 4.3/5 · click for rater detail
Install door hardware, such as locks and closers.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmithing is a small-scale, hands-on trade with limited digitization and no meaningful AI adoption signals; adoption velocity remains negligible and constrained by the inherent physical-world requirements. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Skilled trades like locksmithing show minimal AI/robotic adoption for physical installation work, reflecting low digitization and the hands-on nature of the field. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While AI might assist with diagnostics, scheduling, or inventory management around locksmith work, it offers minimal direct augmentation for the core physical installation task itself, which remains primarily manual. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with scheduling, parts lookup, or diagnostic guidance, but offers little direct assistance during the physical act of installing hardware. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Installing physical hardware requires dexterous manipulation, spatial reasoning, and adaptation to varied door frames and conditions that current robotics and AI cannot reliably perform end-to-end. No current system can autonomously install locks and closers at quality parity with human locksmiths. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manual task requiring measuring, drilling, fitting, and adjusting hardware on doors; no current AI system can perform the physical installation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Locksmithing often requires licensing in many jurisdictions, and liability for security hardware installation (incorrect installation creates security vulnerabilities) creates legal and insurance barriers to automation. Human accountability for lock integrity is a substantial barrier. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing strictly requires a human locksmith in most jurisdictions, but physical access, liability for security failures, and customer trust create real friction against remote or automated substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Robotic installation hardware and AI vision systems capable of this task would cost far more than a locksmith's labor, especially accounting for integration, calibration, and site-specific adaptation overhead. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI/robotic substitute, so any AI-based approach would be far more costly than a human locksmith performing the installation directly. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI or robotic product currently performs door hardware installation reliably in production settings. This remains entirely a human-performed task in all commercial locksmith operations. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed products install door hardware; robotics for this specific unstructured, variable physical task remain research-stage at best. |
Disassemble mechanical or electrical locking devices, and repair or replace worn tumblers, springs, and other parts, using hand tools.
7CI 5–10 · exposure 0 · augmentation 13 · importance 4.1/5 · click for rater detail
Disassemble mechanical or electrical locking devices, and repair or replace worn tumblers, springs, and other parts, using hand tools.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmith services are performed by small, local, craft-based businesses with low digital infrastructure and limited capital for robotics investment. Adoption of AI automation in this sector remains negligible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing is a small-scale, physical trade with minimal digitization and no evidence of AI or robotic adoption for repair tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers minimal assistance for the core physical task of hand-tool disassembly and repair. Diagnostic software might help identify problems, but does not meaningfully augment the hands-on mechanical work itself. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with diagnostic references, parts lookup, or repair manuals/videos, but offers little direct enhancement to the hands-on disassembly and repair process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires precision physical manipulation in three-dimensional space—disassembling delicate mechanical/electrical components and reassembling them with exact tolerances. Current AI cannot operate robotic systems with the dexterity and reliability needed for intricate hand-tool work on varied lock mechanisms. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on physical repair task requiring dexterity, tactile feedback, and manipulation of small mechanical parts with hand tools; no current AI system can perform physical disassembly or repair. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Physical access to customer premises, trust in handling security devices, and liability for damage or security breaches create substantial barriers. Many jurisdictions require licensed locksmiths to handle security devices, adding legal and regulatory friction to any automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not formally licensed everywhere, locksmithing often involves trust, background checks, and security-sensitive access, creating moderate friction against non-human substitution, though not a hard legal barrier in most jurisdictions. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Developing, deploying, and maintaining a robotic system capable of this task would cost far more than the labor of a skilled locksmith performing the work manually. Setup and per-task overhead would exceed human wages for years. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI or robotic substitute for this physical task, so any hypothetical automation would require expensive custom robotics far exceeding human labor costs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can reliably disassemble and repair mechanical locks end-to-end in field conditions. Specialized robotic solutions exist only in research or highly controlled manufacturing environments, not in production locksmith work. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed robotic or AI product performs lock disassembly and tumbler/spring repair; this remains purely a human manual skill with no commercial automation. |
Repair and adjust safes, vault doors, and vault components, using hand tools, lathes, drill presses, and welding and acetylene cutting apparatus.
7CI 5–10 · exposure 0 · augmentation 13 · importance 4.0/5 · click for rater detail
Repair and adjust safes, vault doors, and vault components, using hand tools, lathes, drill presses, and welding and acetylene cutting apparatus.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmithing is a traditional skilled trade with low digitization, dispersed small firms, and strong reliance on on-site physical presence. Adoption of automation in this sector has been minimal and is constrained by the inherent need for physical dexterity and presence. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | The locksmith trade is a small, physical, low-digitization sector with minimal AI adoption for hands-on mechanical repair work. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers no meaningful augmentation to the core task of physically repairing and adjusting safes and vault components. Diagnostics or planning tools might marginally assist a locksmith, but they do not address the hands-on mechanical work that defines the task. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with diagnostics, parts lookup, or documentation, but offers little help with the core physical repair, adjustment, and welding work itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical manipulation of heavy machinery, hand tools, and specialized equipment in three-dimensional space to repair mechanical and structural components. Current AI systems have no capability to perform autonomous physical work, welding, cutting, or precise mechanical adjustment on safes and vault doors. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on physical repair task involving manual dexterity, tool use, welding, and precision fitting on heavy metal components—far beyond current AI or robotic capability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Locksmithing often requires licensing and bonding in many jurisdictions, and safe/vault repair carries significant liability and legal responsibility for security-critical systems. These regulatory and liability barriers substantially protect human employment in this role. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed nationally in most areas, safe and vault work often requires background checks, insurance/bonding, and specialized security clearance due to the sensitive nature of vault security, creating moderate barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | There are no AI systems capable of performing this task, making cost comparison moot. Skilled locksmiths command high hourly rates, and the capital cost of robotics sufficient to perform this work far exceeds the operational savings. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven alternative to perform this physical repair work, so any AI-based approach would be far costlier or simply infeasible compared to a human locksmith. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product performs autonomous safe and vault repair today. This is inherently a physical task requiring embodied robotics at a precision and complexity level not yet in production deployment for locksmithing work. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product exists that can physically repair or adjust safes and vault doors; this remains purely a manual skilled-trade task. |
Unlock cars and other vehicles.
7CI 5–10 · exposure 0 · augmentation 13 · importance 3.3/5 · click for rater detail
Unlock cars and other vehicles.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The locksmith sector remains highly fragmented, locally oriented, and dependent on human-performed manual work. Adoption of automation is minimal; this task occurs in small, often family-run businesses with low digitization and strong reliance on on-site physical presence. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing is a low-digitization, physical trade sector with minimal AI agent deployment or measurable displacement trends. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI cannot meaningfully augment the core physical task of lock manipulation. While AI might assist with scheduling or customer communication, it does not enhance a locksmith's ability to actually unlock a vehicle. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with diagnostics, looking up key codes, or scheduling/dispatch, but offers little assistance for the core physical unlocking action itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Unlocking vehicles requires physical manipulation of locks and security systems in varied environmental conditions. Current AI has no physical embodiment and cannot perform the hands-on mechanical work needed; this task is entirely dependent on robotic systems that do not yet operate reliably in the field. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical, manual task requiring hands-on manipulation of locks, tools, or electronic bypass equipment at a specific vehicle location; no AI system can perform the physical unlocking act. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Vehicle owners and insurers require authorized locksmiths for security and liability reasons, and many jurisdictions regulate locksmith licensing. Unauthorized vehicle entry raises legal barriers, and customers strongly prefer human verification of identity and ownership. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no license is strictly required in most jurisdictions, the task requires physical presence, specialized tools, and trust/liability considerations (theft prevention, insurance verification) that create real friction against remote or automated substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The capital and operational cost of a robot arm capable of lock manipulation, combined with AI integration and safety systems, far exceeds the cost of a human locksmith visit or remote dispatch for the same outcome. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical service, so AI cost cannot undercut human labor cost for this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial product can reliably unlock vehicles end-to-end without human locksmith intervention. Specialized tools and physical access remain required, and no AI system in production performs this autonomously. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical vehicle unlocking; this remains entirely a human/mechanical service, sometimes aided by proprietary dealer key-coding tools operated by a technician. |
Move picklocks in cylinders to open door locks without keys.
5CI 0–10 · exposure 0 · augmentation 0 · importance 4.6/5 · click for rater detail
Move picklocks in cylinders to open door locks without keys.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmith services are primarily small, localized businesses with low digitization and little pressure to automate field lock-opening work. Current adoption of any automation in this sector is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | The locksmith trade is a small-scale, physical, low-digitization sector with essentially no AI/robotic adoption for the core picking task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers no meaningful assistance for the core mechanical task of manipulating picklocks; human locksmiths cannot be productively augmented by current AI systems for the physical act of opening locks. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI provides no meaningful real-time assistance during the physical act of manipulating pins in a lock cylinder. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Operating picklocks requires highly dexterous manipulation of small mechanical components in a confined, varied space. Current AI systems, including robotics, lack the sensory feedback and precise fine-motor control to reliably perform this task across different lock types and conditions. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a fine-motor, tactile physical manipulation task requiring feel for pin tension and mechanical feedback that no current AI or robotic system can perform.PICKING locks requires embodied dexterity far beyond current robotics/AI capability.Not automatable via off-the-shelf AI. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Locksmith work often involves legal authorization and liability; unauthorized lock opening is illegal. Additionally, customers typically contact locksmiths specifically to gain entry in emergency situations requiring human judgment and legal accountability. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing mandates a human specifically, but trust, liability, and physical access/security concerns create moderate friction against any automated substitute, real or hypothetical. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The hardware investment required for a general-purpose robotic system capable of fine lock manipulation vastly exceeds the loaded hourly wage of a locksmith, making automation economically infeasible with current technology. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI system to compare cost against; a human locksmith remains the only functional option, making AI more expensive by default (infinite cost for non-existent capability). |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system or robot can reliably manipulate picklocks in door cylinders in real-world conditions. While some laboratory robotics research explores lock manipulation, no production product performs this task reliably at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs manual lock picking; this remains a specialized human skill with no commercial robotic or AI substitute in production. |
Open safe locks by drilling.
5CI 0–10 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail
Open safe locks by drilling.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmiths operate in small, skilled-trade firms with low digitization. There is no evidence of AI or robotic adoption in this sector; the work remains almost entirely manual and human-dependent. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing is a small-scale, physical, low-digitization trade with essentially no reported AI/robotic adoption for this specific hands-on task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI tools might assist with lock identification or drilling technique documentation, but the physical drilling task itself offers minimal augmentation surface—the locksmith must be present and in control for safety and legal reasons. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help with diagnostics (identifying lock model, drill point references, manuals) but offers little assistance with the actual physical drilling and manipulation. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Opening safe locks by drilling requires precise physical manipulation in real-world environments, real-time problem-solving based on lock type and condition, and judgment about structural integrity—capabilities current AI systems lack. No end-to-end automation exists that could meet the 50% time-saving threshold for this hands-on mechanical task. |
| Task automatability | claude-sonnet-5 | 1/5 | Drilling into a safe requires precise physical manipulation, tactile feedback, and real-time judgment about material and lock mechanism that current AI and robotics cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Locksmithing is a licensed profession in most jurisdictions, and safe opening carries legal, liability, and security implications that typically require a credentialed human professional to perform and certify. Authorization and liability barriers are substantial. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed like medicine, there is significant liability, security sensitivity, and trust requirements (locksmiths are often vetted/bonded) that create moderate friction against any hypothetical automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying a specialized robotic system capable of safe drilling would far exceed the labor cost of a trained locksmith performing the task manually. Current technology remains prohibitively expensive relative to human labor for this application. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system performing this physical task at all, so the human is the only available and cost-effective option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can autonomously perform safe drilling. This task requires robotic systems with sophisticated tactile feedback, lock-specific knowledge, and adaptive control that do not exist in production use for locksmithing. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product or robotic system exists that autonomously locates and drills safe locks in the field; this remains a purely manual, tool-based skilled trade task. |
Install safes, vault doors, and deposit boxes according to blueprints, using equipment such as power drills, taps, dies, truck cranes, and dollies.
5CI 0–10 · exposure 0 · augmentation 25 · importance 3.6/5 · click for rater detail
Install safes, vault doors, and deposit boxes according to blueprints, using equipment such as power drills, taps, dies, truck cranes, and dollies.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Locksmith and safe-repair services remain traditional, largely manual, small-firm operations with low digitization. Adoption of AI-driven or robotic automation in this sector has been negligible, with work continuing to rely on skilled human technicians. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing and physical security installation is a low-digitization trade sector with minimal AI/robotic adoption in the field. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with blueprint interpretation, site measurement documentation, or parts inventory management, but these are peripheral to the core installation task. The primary value in this occupation remains hands-on physical expertise, which AI augmentation addresses only minimally. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help with reading blueprints, planning logistics, or diagnosing issues, but offers little direct assistance during the physical installation process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires precise physical installation of heavy, secure equipment in specific locations according to blueprints, involving equipment operation and spatial calibration that current AI cannot perform autonomously. Physical manipulation, site assessment, and real-time adjustment of placement demand embodied robotic capability far beyond deployed systems today. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical installation task requiring manipulation of heavy equipment (cranes, dollies), precision drilling, and manual fitting to blueprints—far beyond current AI or robotic capability in unstructured field environments. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Installation of safes and vault doors is often subject to regulatory compliance, insurance requirements, and legal liability for proper secure installation. Many jurisdictions require licensed or bonded locksmiths to perform this work, and errors carry high financial and security consequences that mandate human accountability. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed like medicine, safe/vault installation often involves security-sensitive work, insurance requirements, and manufacturer certification, creating moderate procedural friction even though no legal barrier exists against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The capital cost of autonomous robotics capable of this task, plus integration and oversight, would vastly exceed the loaded wage of a locksmith performing the work manually. Current robotic systems lack the dexterity and site-adaptation capability to justify automation cost. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system capable of performing this physical labor, so any AI cost is irrelevant—human labor is the only viable option, making AI comparatively far more expensive (effectively infinite). |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product can autonomously install safes, vault doors, or deposit boxes. This requires physical presence, equipment operation, and structural integration that only specialized human locksmiths or robotic systems (not in production for this use case) can accomplish. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product installs safes or vault doors; this remains entirely a manual trade skill performed by humans on-site. |
Set up and maintain master key systems.
3CI 0–5 · exposure 0 · augmentation 13 · importance 4.3/5 · click for rater detail
Set up and maintain master key systems.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The locksmith trade operates in small, localized service businesses with low digital infrastructure adoption; the physical, on-site nature of the work and regulatory requirements mean AI adoption velocity in this sector remains minimal. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Locksmithing is a small-business, physical-labor trade with minimal digitization or AI adoption in its core hands-on functions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers minimal assistance for the core mechanical and spatial tasks involved in setting up and maintaining master key systems; even advisory tools have limited practical value given the hands-on, context-specific nature of the work. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help with administrative aspects like tracking key hierarchies or documentation via software, but offers little assistance for the physical keying and system design work itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Setting up and maintaining master key systems requires physical manipulation of locks, pins, and hardware, along with spatial reasoning about access hierarchies that demand human judgment and on-site problem-solving. Current AI systems cannot perform the manual and field-based components of this task. |
| Task automatability | claude-sonnet-5 | 1/5 | Designing and maintaining a master key system requires physical assessment of hardware, hands-on pinning/keying of cylinders, and site-specific security judgment that current AI cannot execute end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Locksmiths typically require state licensure and must comply with security regulations and liability standards for handling master key systems; legal and regulatory frameworks mandate human accountability for security systems, creating hard barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Master key systems involve security-sensitive access control often requiring licensed locksmiths, background checks, and liability for security breaches, creating strong barriers to non-human execution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI has no meaningful role in performing this task today, making a cost comparison irrelevant; a human locksmith remains the only feasible option and thus carries the full cost. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so AI cost is effectively infinite relative to a locksmith's labor cost for the same output. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product can independently set up or maintain physical master key systems; the task fundamentally requires hands-on mechanical work that is beyond the scope of current robotic and AI capabilities in real-world deployment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical locksmithing tasks like master key setup; this remains entirely a manual, skilled-trade activity. |
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.