Photonics Technicians
17-3029.08Median wage $78,350/yr59,930 employed (US)Rank #515 of 923 scored · top 56% by substitution
Build, install, test, or maintain optical or fiber optic equipment, such as lasers, lenses, or mirrors, using spectrometers, interferometers, or related equipment.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution27
Exposure21
Augmentation48
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
24 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
4%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%23
panel mean rating 1.9/5 → substitution pressure 23/100
Technical feasibility todayw 20%19
panel mean rating 1.8/5 → substitution pressure 19/100
Cost vs. human wagew 15%21
panel mean rating 1.8/5 → substitution pressure 21/100
Adoption barriersw 20%inverted — strong barriers lower the score46
panel mean rating 3.1/5 (barrier strength) → substitution pressure 46/100
Sector adoption velocityw 10%24
panel mean rating 1.9/5 → substitution pressure 24/100
Task breakdown (24 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.
Monitor inventory levels and order supplies as necessary.
81CI 72–89 · exposure 80 · augmentation 75 · importance 3.0/5 · click for rater detail
Monitor inventory levels and order supplies as necessary.
81| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Manufacturing and technical supply chain operations have rapidly adopted automated inventory management systems over the past 15 years; photonics labs and technician roles are typically embedded in digitized manufacturing or research environments where such automation is already standard practice. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Manufacturing and specialized technical fields like photonics adopt inventory software steadily but not as fast as pure information/finance sectors, often lagging due to smaller scale operations. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Inventory automation substantially augments technician productivity by eliminating manual stock checks and routine ordering, freeing time for higher-value procurement decisions, supplier relationship management, and hands-on technical work while the system handles threshold-based alerts and basic reordering. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled inventory systems can flag low stock, predict usage patterns, and auto-generate purchase orders, significantly easing the technician's workload while they retain oversight. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Inventory monitoring and supply ordering is largely routine data entry and threshold-based decision-making. Current AI systems (including inventory management software with agent capabilities) can track stock levels, generate purchase orders, and flag reorder points automatically, achieving significant time savings with minimal human intervention. |
| Task automatability | claude-sonnet-5 | 4/5 | Inventory monitoring and reordering follows structured, rule-based logic (thresholds, reorder points) that off-the-shelf inventory management and procurement software already automates effectively. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some organizations may require human authorization for purchase orders above certain thresholds, no legal licensing or strict regulatory requirement mandates human involvement in inventory monitoring itself, and automated systems are widely normalized in the sector. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, regulatory, or liability barriers prevent automating inventory tracking and ordering; it's a routine administrative task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated inventory systems operate on marginal inference costs (API calls, database queries) with minimal oversight, making them at least an order of magnitude cheaper than human labor for continuous monitoring and routine purchase decisions. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated inventory software subscriptions cost far less than technician time spent manually checking and ordering supplies, offering substantial savings at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature enterprise resource planning (ERP) and inventory management systems with automated reordering capabilities are widely deployed in production across manufacturing and technical sectors, reliably executing this task at scale with established workflows. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Mature ERP/inventory management systems (SAP, Oracle, dedicated inventory tools) are widely deployed in manufacturing and lab environments to track stock and trigger automatic reordering. |
Compute or record photonic test data.
52CI 37–67 · exposure 50 · augmentation 88 · importance 4.3/5 · click for rater detail
Compute or record photonic test data.
52| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Manufacturing and research labs increasingly use automated data acquisition and LIMS systems, but adoption of AI-driven computation and validation of photonic measurements remains in pilot/early production phase rather than widespread deployment. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Manufacturing and photonics testing environments adopt automation steadily but not as fast as pure information-sector work, with many small/specialized labs still using manual or semi-manual processes. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can substantially assist technicians by automating data entry, flagging anomalies in test results, suggesting diagnostic insights, and generating summary reports—enabling faster interpretation and decision-making while the technician remains responsible for validation and next steps. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI and automated data systems substantially speed up computation, trend analysis, and documentation for technicians, letting them focus on setup, troubleshooting, and interpretation. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Recording photonic test data is straightforward and automatable via standard data logging systems, but computing derived metrics from raw photonic measurements often requires domain expertise to interpret complex physical phenomena and validate anomalies—making full end-to-end automation without significant customization unlikely. |
| Task automatability | claude-sonnet-5 | 4/5 | Recording and computing test data from photonics equipment is largely structured numerical work—data logging, unit conversion, and calculation—that AI/software can perform with high reliability, though initial sensor/instrument integration requires setup. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Integration with proprietary laboratory equipment, quality-control standards, and traceability requirements for test data create moderate friction; regulatory compliance in photonics/optics sectors may require human verification and sign-off on measurements. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks automation of data recording/computation, though quality control and calibration verification in some regulated manufacturing contexts add minor friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Automated data logging hardware and basic recording systems are cost-effective, but integrating AI for computation of derived photonic metrics, validation, and error handling still requires expert setup and oversight, keeping all-in costs comparable to or exceeding skilled technician wages. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Once integrated, automated data logging and computation software is far cheaper per test than paying a technician to manually record and compute data, though initial integration costs offset some savings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Data logging and recording infrastructure exists and is deployed in labs and manufacturing, but general-purpose AI systems cannot reliably compute or validate photonic-specific measurements without specialized domain models and integration with laboratory equipment. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Lab information systems and instrument software already automate data capture and computation in many photonics/optics labs, but full end-to-end automation across varied instruments and custom test setups is inconsistent and often still requires technician oversight. |
Terminate, cure, polish, or test fiber cables with mechanical connectors.
49CI 15–84 · exposure 45 · augmentation 38 · importance 3.7/5 · click for rater detail
Terminate, cure, polish, or test fiber cables with mechanical connectors.
49| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | High-volume telecommunications and data center fiber cable manufacturing has rapidly adopted automated termination and testing systems. Companies in information/telecom sectors are actively deploying robots for this work, though smaller custom shops may lag. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Photonics/telecom manufacturing and field technician work is a low-digitization, physical-labor sector with minimal AI/robotic adoption for hands-on fiber optic termination tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted vision systems can help technicians inspect and diagnose fiber quality or connector defects, and robotic guidance can assist manual termination in mixed-automation settings. However, augmentation is moderate since the task is already well-suited to full automation rather than human-AI collaboration. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with quality inspection or test result analysis (e.g., automated OTDR trace interpretation) but offers little help with the physical curing, polishing, and connector termination steps themselves. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Terminating, curing, polishing, and testing fiber cables with mechanical connectors are largely procedural, repetitive operations that can be performed by robotic systems with vision and force control. Current industrial robots and vision-guided automation can execute these tasks with high precision and consistency, easily meeting the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on manual fabrication and testing task involving fine motor manipulation of fragile fiber components, curing epoxies, and physical polishing—none of which current AI systems can perform without robotic embodiment specifically engineered for this niche. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing requirements or legal barriers prevent automation of these tasks; they are primarily mechanical and process-driven. Organizational friction may exist due to upfront capital requirements and technician workforce displacement, but these are not insurmountable barriers to adoption. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists, but the task demands specialized physical dexterity and quality-critical craftsmanship, creating practical (not regulatory) barriers to substitution by generic automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Fiber termination and polishing automation can be capital-intensive, but once deployed, per-unit costs are substantially lower than skilled technician labor, particularly at scale. The equipment cost amortized over high-volume production yields favorable ROI compared to hourly technician wages. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven automation solution for this physical task, so cost comparison favors the human technician entirely; any robotic alternative would require expensive specialized hardware exceeding technician wages. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Automated fiber optic processing systems exist in production environments, though they typically handle specific connector types and require some manual setup or changeover. Products are mature in high-volume manufacturing but may struggle with diverse connector variants or edge cases that require human judgment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI/robotic product performs fiber termination, curing, and polishing end-to-end in production; this remains a manual technician task in industry. |
Document procedures, such as calibration of optical or fiber optic equipment.
45CI 34–56 · exposure 38 · augmentation 63 · importance 3.9/5 · click for rater detail
Document procedures, such as calibration of optical or fiber optic equipment.
45| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics and optical equipment sectors are specialized and slower to digitize than mainstream IT or finance. While documentation tools are available, actual adoption of AI for procedure documentation in precision optics remains limited, with most organizations relying on manual expert authoring. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing and technical trades are moderate-to-low digitization sectors where AI-assisted documentation tools are only beginning to see pilot use, not widespread deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by drafting initial procedure templates, formatting, and generating common sections (safety warnings, tool lists), reducing manual writing burden. However, the technician must validate technical accuracy and equipment-specific details, making it genuinely assistive but not transformative. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can significantly speed up drafting, formatting, and standardizing calibration procedure write-ups, letting technicians focus on verification rather than composition. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate documentation from technical specifications or existing procedures, creating accurate calibration documentation requires understanding equipment-specific parameters, measurement standards, and safety protocols. Current systems lack reliable domain expertise to independently write complete, validated calibration procedures without substantial human oversight and verification. |
| Task automatability | claude-sonnet-5 | 3/5 | Drafting and structuring calibration procedure documentation from technician notes or data logs is well within current AI capability, though capturing precise technical steps and equipment-specific nuances still requires human input.,"rating":3, but final text needs a technician's factual verification, so full end-to-end automation is partial."rationale used above. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Calibration procedures often require sign-off by qualified technicians and may need to comply with ISO standards or equipment manufacturer specifications, creating some regulatory and liability friction. However, the documentation itself is not legally restricted from AI assistance, only subject to quality assurance requirements. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human write documentation, though quality/traceability standards (e.g., ISO calibration records) require accurate human verification, creating some friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI-assisted documentation tools are relatively inexpensive to run, but the required expert review and validation to ensure correctness approaches the cost of human-written documentation, making the all-in cost comparable to hiring a technician to write procedures. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | An LLM-assisted drafting process is far cheaper per document than a technician manually writing from scratch, even accounting for review time. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Documentation generation tools exist and can draft technical content, but no deployed product reliably produces calibration procedures that meet precision instrument standards without expert review. Products struggle with equipment-specific details, regulatory compliance, and accuracy verification needed for optical/fiber optic calibration docs. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Generic AI writing assistants and templated documentation tools are used in technical settings, but no specialized deployed product handles photonics-specific calibration documentation reliably at scale. |
Mix, pour, or use processing chemicals or gases according to safety standards or established operating procedures.
44CI 5–83 · exposure 50 · augmentation 38 · importance 3.1/5 · click for rater detail
Mix, pour, or use processing chemicals or gases according to safety standards or established operating procedures.
44| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Semiconductor and pharmaceutical manufacturing, major users of this task, have already adopted automated chemical handling systems at scale; photonics labs and advanced manufacturing facilities are following suit with high digitization and capex investment. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Semiconductor/photonics manufacturing floor tasks involving physical chemical handling are among the slowest to adopt AI/robotic automation due to safety and precision requirements. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted monitoring and alerting (e.g., anomaly detection on chemical batch quality, real-time safety compliance checks) can support technician oversight, though the task itself is primarily manual/procedural and does not inherently benefit from AI assistance when fully automated. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with monitoring, logging, or alerting on procedure adherence, but offers minimal direct enhancement to the physical act of mixing or pouring chemicals. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Precise mixing and pouring of chemicals at measured ratios, guided by procedures, is fully automatable with robotic dispensing systems and automated gas handling equipment. This task has no judgment requirement beyond following established protocols, meeting the ≥50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical handling of chemicals/gases in a lab or fab environment requires manual dexterity, real-time sensory judgment, and physical presence that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety regulations, hazardous material handling licenses, and liability for chemical errors create significant organizational and legal friction; responsibility for errors often remains with a licensed supervisor or technician who must oversee or approve the automated process. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety regulations, hazardous materials handling protocols, and liability for chemical/gas mishandling create strong barriers requiring trained, often certified personnel. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated dispensers, robotic arms, and integrated process control systems cost thousands to tens of thousands, but handle continuous high-volume operations with minimal labor, achieving cost-per-task that is orders of magnitude below human labor when amortized. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Any automation would require specialized robotics and safety engineering far more costly than a technician's wage for this specific task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Automated chemical dispensing, robotic mixing systems, and gas handling equipment are mature, deployed technologies in pharmaceutical, semiconductor, and photonics laboratories and manufacturing facilities operating at scale today. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically mixes or pours chemicals; robotic chemical dispensing exists only in narrow, highly controlled research/industrial contexts, not as general photonics technician replacement. |
Lay out cutting lines for machining, using drafting tools.
34CI 30–38 · exposure 25 · augmentation 63 · importance 3.1/5 · click for rater detail
Lay out cutting lines for machining, using drafting tools.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Manufacturing and photonics sectors have adopted CAD tools widely, but AI-driven autonomous layout remains in pilot/tool-assisted phase rather than full replacement. Adoption of AI-guided layout is moderate, with human technicians still central to verification. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing and precision technician trades adopt digital tools unevenly; many shops still rely on manual or semi-manual layout processes, reflecting slower adoption than software-centric industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered drafting assistants, generative design tools, and ML-based pattern suggestion systems can significantly enhance technician productivity by automating layout suggestions, checking against specs, and highlighting optimizations—keeping the human technician in control while reducing manual drafting time substantially. |
| Augmentation potential | claude-sonnet-5 | 3/5 | CAD/CAM tools substantially assist technicians in planning and visualizing cutting lines, improving speed and accuracy, though the technician still executes and verifies the physical layout. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Laying out cutting lines requires interpreting engineering specifications and translating them to physical materials, which involves spatial reasoning and contextual judgment. Current AI can assist with design layouts and generate cutting patterns, but end-to-end automation would require reliable physical measurement interpretation and material-specific decision-making that typically still requires human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | CAD/CAM software can generate cutting layouts from specifications, but this task as described involves manual drafting tool use and physical layout on stock, which requires physical presence and coordination not achievable end-to-end by current AI. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Engineering liability and quality standards create moderate friction—errors in cutting layouts directly affect manufactured product quality and safety. Regulations around precision manufacturing and design sign-off typically require human accountability, though not necessarily exclusive human execution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement demands a human specifically perform manual layout, but precision machining work often requires quality verification and sign-off, creating some organizational friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | CAD and drafting software requires licensing and skilled operator oversight. The total cost (tool licensing, human time for verification, integration) remains comparable to or exceeds having a skilled technician perform the layout directly. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While CAM software is cheap once implemented, transitioning from manual drafting practices requires software licensing, CAD model creation, and technician oversight, keeping costs comparable to or only modestly below human labor for this specific legacy task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | CAD software and layout automation tools exist, but they require human input on material properties, tolerance requirements, and context-specific constraints. No deployed AI system reliably performs this task independently at production quality without significant human verification and adjustment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | CAM software and CAD tools are mature for generating toolpaths from digital models, but the manual drafting-tool layout task itself is largely replaced by different digital workflows rather than performed reliably by AI systems directly. |
Design, build, or modify fixtures used to assemble parts.
30CI 25–35 · exposure 20 · augmentation 50 · importance 3.4/5 · click for rater detail
Design, build, or modify fixtures used to assemble parts.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics and precision manufacturing are moderately digitized sectors, but fixture design and construction remain largely manual and skilled-craft-oriented. Adoption of generative design tools exists but is not yet deep or widespread in production settings. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing and technician trades adopt AI slowly compared to information-sector fields, with automation concentrated in CAD/simulation tools rather than fixture fabrication itself. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist with CAD generation, design optimization, and technical documentation, helping technicians iterate faster and explore more design variations. However, human judgment on material properties, manufacturability, and real-world constraints remains essential. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted CAD tools, generative design, and simulation can meaningfully speed up the design phase of fixtures, though the physical build/modification still requires human hands-on work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Designing and building physical fixtures requires spatial reasoning, material selection, and hands-on prototyping that current AI struggles with end-to-end. While AI can assist with CAD design and technical documentation, the iterative physical construction and modification phases remain heavily dependent on human judgment and manual work. |
| Task automatability | claude-sonnet-5 | 2/5 | Designing and physically building or modifying custom fixtures requires hands-on fabrication, iterative physical fitting, and tacit mechanical judgment that current AI cannot execute end-to-end; at best CAD-assisted design portions could be sped up. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While there are no strict regulatory barriers, the task requires hands-on physical work and expert judgment that organizations prefer to keep human-supervised. Custom fixture design often involves liability for part assembly quality, creating some organizational friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically, but physical fabrication and precision alignment needs create practical barriers to full automation beyond simple organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI design tools are relatively inexpensive, but the bulk of the task cost lies in physical construction, materials, and skilled labor. Current AI cannot eliminate these costs, making the overall cost-to-human-wage ratio unfavorable for full automation. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While CAD/design assistance can be cheap, the physical build and modification of fixtures still requires skilled labor, machining equipment, and iteration, keeping AI's all-in cost comparable to or only marginally cheaper than a technician's time. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI system reliably designs, builds, or modifies physical fixtures independently. CAD software with some generative features exists, but selecting materials, validating designs under real-world constraints, and constructing prototypes require human expertise and physical interaction that current tools do not automate reliably. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously designs, builds, and modifies physical assembly fixtures for photonics work; this remains a human machining/engineering task with only design-software assistance available. |
Test or perform failure analysis for optomechanical or optoelectrical products, according to test plans.
28CI 25–30 · exposure 25 · augmentation 75 · importance 3.7/5 · click for rater detail
Test or perform failure analysis for optomechanical or optoelectrical products, according to test plans.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics is a specialized, low-volume sector with high precision demands and safety constraints. Adoption of fully autonomous testing automation remains limited; most organizations use augmented testing tools rather than replacement systems. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing and hardware testing sectors are physical, low-digitization environments where AI adoption for hands-on inspection and test execution remains in early pilot stages. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist by automating data logging, pattern recognition in failure signatures, statistical analysis of test results, and report generation, allowing technicians to focus on troubleshooting decisions and equipment calibration rather than manual data handling. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-based data analysis, pattern recognition in failure signatures, and automated report generation can meaningfully speed up interpretation of test data even though the human still performs physical testing and judgment calls. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Testing optomechanical/optoelectrical products requires hands-on manipulation of precision equipment, calibration judgment, and interpretation of nuanced failure signatures that current AI cannot perform end-to-end. While AI can assist in data analysis and report generation, the physical testing and real-time problem-solving remain human-dependent. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical testing and failure analysis of optomechanical/optoelectrical hardware require hands-on manipulation of equipment, physical measurements, and specialized instrumentation that current AI cannot perform end-to-end; only data analysis portions are automatable. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant barriers exist: specialized equipment certification, safety protocols around lasers and optical hazards, reproducibility requirements for compliance documentation, and domain expertise requirements that organizations typically require a trained technician to validate and sign off on results. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement typically applies, but liability for certifying product quality, safety-critical applications, and organizational quality-control sign-off processes create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Automated test equipment and AI analysis tools are expensive to procure, integrate, and maintain for specialized photonics tasks, while skilled technician labor remains the cost baseline. The all-in cost of automation often exceeds human technician wages for this specialized work. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Physical test equipment, fixtures, and hands-on technician labor still dominate the cost structure; AI can only reduce a small analytical slice of the task, so overall cost savings are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI system reliably conducts physical optomechanical testing or failure analysis independently. Vision systems and robotic arms exist but lack the precision calibration, contextual judgment, and adaptive troubleshooting that this specialized domain demands in production settings. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some software tools assist with data logging and statistical analysis of test results, but no deployed product autonomously executes physical test setups or performs full failure analysis on optical hardware. |
Recommend optical or optic equipment design or material changes to reduce costs or processing times.
28CI 25–30 · exposure 20 · augmentation 50 · importance 3.6/5 · click for rater detail
Recommend optical or optic equipment design or material changes to reduce costs or processing times.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics and optics manufacturing remains concentrated in specialized, capital-intensive sectors with slower digital transformation than software or finance. AI adoption in optical design is still largely confined to research labs and early-stage pilots; production-level displacement of recommendation tasks is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics/optics manufacturing is a specialized physical-engineering sector with slower AI adoption compared to information-centric industries; pilots for design optimization exist but are not widespread. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by rapidly surveying materials databases, simulating optical performance under varied parameters, and flagging cost-reduction opportunities from literature—leaving the technician to synthesize and validate recommendations against real-world constraints. This represents useful but not transformative augmentation of the human expert's workflow. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can help analyze cost data, simulate material properties, and summarize literature, providing useful support to technicians formulating recommendations, though human expertise remains central. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task requires deep domain knowledge of optical physics, materials science, and specific manufacturing constraints to propose cost-effective or performance-enhancing design changes. While AI can assist in literature review and parameter optimization, the creative synthesis of material trade-offs and design innovation needed to *recommend* changes remains beyond current autonomous capabilities. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires hands-on knowledge of optical systems, manufacturing constraints, and cost tradeoffs that AI cannot fully assess without physical inspection and domain-specific tacit knowledge; only partial support (data analysis, literature review) is feasible today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Recommendations may directly affect product quality, safety, and liability; manufacturers typically require sign-off by qualified engineering or technical staff before implementation. However, there is no strict licensing barrier to AI suggesting design changes, creating moderate organizational friction rather than hard regulatory barriers. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars AI, but organizational reliance on technician expertise, safety/quality implications of design changes, and validation requirements create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for optical design and materials analysis carry significant integration and validation costs, and still require expert photonics technicians to review, contextualize, and implement recommendations. The labor cost of human expertise in this specialized domain remains lower than the combined cost of AI infrastructure, domain-specific training, and oversight for autonomous performance. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply assist with research and data crunching, but the core recommendation requires expert technician review and validation, keeping overall cost comparable to or only modestly cheaper than human-only work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI systems reliably perform this task end-to-end in production. Research tools exist for optical design optimization and materials prediction, but these operate within narrow, pre-defined parameter spaces and require expert human validation. Real-world recommendations demand tacit knowledge of supplier relationships, manufacturing feasibility, and cost structures that AI systems do not yet reliably capture. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously recommends optical/photonics equipment design or material changes for cost/time reduction in production settings; this remains an expert engineering judgment task. |
Assist scientists or engineers in the conduct of photonic experiments.
26CI 21–30 · exposure 20 · augmentation 50 · importance 3.8/5 · click for rater detail
Assist scientists or engineers in the conduct of photonic experiments.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics laboratories and manufacturing plants are adopting sensor automation and data-logging tools but remain largely craft-based with strong preference for direct technician expertise; adoption is piecemeal and concentrated in large R&D firms and defense/aerospace contractors rather than broad and rapid displacement. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Scientific/technical lab environments adopt AI for data analysis and simulation but physical experimental assistance remains largely unautomated and adoption is slow in this niche. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by automating data logging, suggesting next experimental parameters based on historical results, flagging signal anomalies in real-time, and automating routine quality checks—raising technician productivity on the analytical side while they remain responsible for physical setup, calibration, and judgment on alignment and equipment state. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help with experimental design suggestions, data analysis, anomaly detection, and documentation, meaningfully aiding technicians even though it can't replace the hands-on work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Assisting in photonic experiments involves calibration, measurement recording, equipment adjustment, and data collection—some of which AI could automate (data logging, anomaly detection in real-time streams). However, the task requires in-situ judgment about optical alignment, physical equipment troubleshooting, and interpretation of live experimental phenomena that demand immediate sensory and spatial reasoning currently beyond AI capability; most of the task remains human-dependent. |
| Task automatability | claude-sonnet-5 | 2/5 | This task involves physical lab setup, equipment handling, and hands-on assistance that current AI cannot perform end-to-end; only data logging or analysis subcomponents are automatable. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Photonic work in research and manufacturing operates within institutional quality assurance and safety protocols; experiments often require a credentialed technician to sign off on data integrity and safety compliance. Some automation of routine measurement and logging faces acceptance friction but not hard legal barriers, though safety-critical adjustments remain human-gated. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing barrier exists specifically, but physical equipment operation, safety protocols, and specialized lab environments create strong practical barriers to remote or automated substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | A photonics technician earns $45–65k loaded; AI augmentation tools (image analysis, data management) cost a few thousand per year but do not yet replace the full labor cost. Full autonomous photonic experiment systems (if they existed) would require significant capital investment and bespoke integration, making all-in cost unfavorable versus a single human technician. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical presence and hands-on manipulation required, so there is no viable AI cost comparison for the core task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI systems can log data, flag statistical anomalies, and suggest next steps based on experimental parameters, no mature production system reliably conducts photonic experiments end-to-end or autonomously troubleshoots optical equipment failures in real lab settings. Bench-level automation exists for narrowly scoped tasks, but photonics requires adaptive physical manipulation and interpretive skill not yet reliably deployed. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product provides physical lab assistance for photonics experiments; this remains firmly in the domain of human technicians with manual dexterity and equipment expertise. |
Perform diagnostic analyses of processing steps, using analytical or metrological tools, such as microscopy, profilometry, or ellipsometry devices.
25CI 20–30 · exposure 20 · augmentation 50 · importance 3.5/5 · click for rater detail
Perform diagnostic analyses of processing steps, using analytical or metrological tools, such as microscopy, profilometry, or ellipsometry devices.
25| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics is a specialized, capital-intensive sector with relatively small populations of technicians; adoption of AI for diagnostic analysis remains pilot-stage and has not achieved significant production displacement even in leading organizations. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Semiconductor/photonics manufacturing is a specialized, capital-intensive physical sector with slower AI adoption relative to pure information-work sectors, though some data analytics tools are being piloted. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered image analysis and pattern recognition can meaningfully assist technicians in reviewing microscopy and spectroscopic data, flagging anomalies, and suggesting preliminary interpretations, improving their diagnostic speed and consistency on routine analyses. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist in analyzing imaging or metrology data, flagging anomalies, and pattern recognition, improving technician efficiency in interpreting results even though the physical diagnostic process remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can analyze images from microscopy and interpret spectroscopic data, the full diagnostic workflow requires skilled interpretation of complex multimodal instrument outputs, troubleshooting decision-making, and physical sample preparation that current AI cannot independently execute end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical operation of specialized metrology equipment (microscopy, profilometry, ellipsometry) and hands-on sample handling that current AI cannot perform end-to-end; only data interpretation portions are automatable.'},'`ok`, ,` |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Photonics manufacturing relies on calibrated, traceable measurement standards and often operates under quality assurance and regulatory frameworks (ISO, FDA in medical/semiconductor) that require documented human expertise and sign-off, creating significant legal and compliance friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No explicit licensing requirement, but cleanroom protocols, equipment liability, and the need for physical presence and calibrated judgment create meaningful organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI tools for microscopy and metrological data analysis require domain expert oversight, integration with proprietary instrument software, and validation labor that keeps total automation costs comparable to or higher than paying a technician for the diagnostic work. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Specialized equipment operation and interpretation still require skilled technician oversight, and AI analysis tools add cost on top of existing metrology infrastructure without replacing the technician's physical role. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI-based image analysis tools exist for microscopy data and basic metrological interpretation, but deployed products remain narrow in scope, require significant human setup and validation, and have not demonstrated reliable production-scale autonomous diagnostic analysis across the range of photonics processing steps. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously runs full diagnostic metrology workflows in photonics fabs; AI is at most used for post-hoc data analysis, not physical instrument operation or diagnosis. |
Assist engineers in the development of new products, fixtures, tools, or processes.
23CI 16–30 · exposure 20 · augmentation 50 · importance 3.6/5 · click for rater detail
Assist engineers in the development of new products, fixtures, tools, or processes.
23| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics is a specialized, small sector with primarily hands-on laboratory work, limited digitization, and small firm structures. Adoption of AI in this field remains slow, with most applications still in pilot stages rather than production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing and hardware engineering sectors show slower, more cautious AI adoption compared to software/information sectors, with pilots for design assistance rather than deployed production automation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist photonics technicians with simulation modeling, technical documentation, literature synthesis, and data analysis, moderately enhancing their productivity while the engineer-technician team remains central to development work. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can help with simulation, documentation, literature review, and design calculations, providing moderate productivity boosts to engineers and technicians during development work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task requires close collaboration with engineers on novel product and process development, which involves creative problem-solving, domain expertise interpretation, and real-time iteration. AI can assist with data analysis, documentation, and research compilation, but cannot independently drive the iterative engineering development cycle that defines the core activity. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a hands-on, collaborative support role involving physical prototyping, testing fixtures, and lab work that current AI cannot perform end-to-end; only narrow sub-components like documentation or simple calculations are automatable. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | This work involves direct collaboration with licensed engineers and hands-on manipulation of precision optical equipment in regulated environments. The requirement for human technical judgment, lab safety compliance, and direct engineer oversight creates significant adoption barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically, but organizational and technical barriers are significant: physical lab equipment, safety protocols, and close engineer collaboration limit remote/AI substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Photonics technicians command specialized wages for hands-on expertise in optical systems and laboratory work. AI tools remain expensive to integrate into specialized lab environments relative to the cost of a technician, particularly when accounting for setup, validation, and oversight overhead. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Because AI cannot substitute for the physical and technical assistance work involved, there is no viable AI cost comparison—human technicians remain necessary for hands-on tasks. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI tools can support aspects of engineering work (CAD assistance, literature review, technical writing), no deployed product reliably performs the full assistive engineering role across photonics design and manufacturing challenges. Current systems lack the deep domain expertise and real-time lab integration this role demands. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously assists in physical photonics product/fixture development; this remains a research-stage capability gap requiring physical manipulation and specialized domain judgment. |
Assemble fiber optical, optoelectronic, or free-space optics components, subcomponents, assemblies, or subassemblies.
21CI 7–35 · exposure 13 · augmentation 38 · importance 3.9/5 · click for rater detail
Assemble fiber optical, optoelectronic, or free-space optics components, subcomponents, assemblies, or subassemblies.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics manufacturing remains concentrated in specialized, lower-digitization sectors (small job shops, R&D labs, custom integrators) rather than high-volume consumer electronics. Adoption of AI-driven assembly automation is nascent, with most facilities still relying on skilled manual assembly. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing is a specialized, moderately digitized physical production sector where robotic automation exists but general AI adoption for hands-on assembly remains slow and narrow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered vision systems for component alignment, defect detection, and work-instruction guidance can meaningfully assist technicians by reducing inspection time and guiding precision placement. However, augmentation is confined to specific subtasks rather than transforming the entire assembly workflow. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with process planning, defect detection, or documentation, but offers limited direct help with the physical act of assembling optical components. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Fiber optic and optoelectronic assembly involves fine-pitched, precision mechanical work with strict alignment tolerances that typically require human dexterity and real-time sensory feedback. While AI vision systems can inspect components, end-to-end autonomous assembly of complex optical subassemblies remains beyond current robotics capabilities, and no deployed AI systems achieve the ≥50% time-saving threshold for the full task. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical assembly of precision optical/optoelectronic hardware requires fine motor manipulation, alignment, and hands-on fabrication that current AI systems cannot perform end-to-end without robotics far beyond off-the-shelf capability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Quality assurance, alignment verification, and rework of optical assemblies typically require certified technician sign-off and expert judgment on failure modes. Many contracts and regulatory contexts (aerospace, telecom) demand human accountability for critical optical subassemblies, creating licensing and organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks automation, but the physical nature of the work, need for precision handling, and quality/safety verification create practical barriers to any generic AI substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Precision assembly equipment and specialized robotics capable of handling optical components are capital-intensive and require significant integration overhead. For the specialized, low-volume work typical in photonics, human technician labor often remains cheaper than deploying custom automation. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based substitute performing this physical task, so cost comparison favors the human technician by default; specialized robotic assembly lines exist but are capital-intensive, not cheap AI inference. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI-guided inspection and defect detection exist in production settings, but no mature AI products autonomously perform the core assembly task reliably. Robotic integration in this space remains largely custom, research-backed, or at pilot stage rather than deployed at scale in photonics manufacturing. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed general-purpose AI product performs physical optics assembly; any automation here is specialized robotic manufacturing equipment, not AI systems in the sense assessed by this rubric. |
Optimize photonic process parameters by making prototype or production devices.
21CI 13–30 · exposure 13 · augmentation 50 · importance 3.9/5 · click for rater detail
Optimize photonic process parameters by making prototype or production devices.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics is a specialized, capital-intensive sector with slower digitization than software or finance; most optimization still relies on technician expertise and iterative physical experimentation rather than AI-driven automation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing is a specialized, moderately digitized hardware sector where AI adoption for physical prototyping tasks lags behind software/information sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered simulation, parameter recommendation, and data analysis can meaningfully assist technicians in narrowing design space and predicting outcomes, but the human remains central to device construction, measurement interpretation, and validation. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with simulation, design optimization, and data analysis of test results to guide parameter choices, even though the physical fabrication and testing must be done by humans. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task requires hands-on physical manipulation of equipment, real-time troubleshooting of hardware, and iterative experimental design that combines technical judgment with direct device fabrication and testing—capabilities current AI systems cannot perform autonomously. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires hands-on fabrication, physical testing, and iterative hardware experimentation in cleanroom/lab environments that current AI cannot perform end-to-end without robotic embodiment and physical manipulation capabilities. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Lab safety regulations, equipment operation licensing, and the need for direct accountability in manufacturing quality control create some organizational friction; however, no hard legal requirement mandates human signature on optimization decisions alone. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no explicit licensing requirement exists, the task demands physical dexterity, specialized equipment access, and lab safety protocols that create substantial organizational and physical barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Photonics technicians earn substantial wages, and the cost of computational simulation or AI guidance is relatively low, but the critical human labor (device fabrication, measurement, calibration, safety oversight) still dominates total cost per optimized device. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot replace the physical labor, equipment operation, and hands-on troubleshooting involved, so no meaningful cost substitution exists; humans remain necessary for the physical task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with parameter modeling and simulation, no deployed product reliably optimizes photonic devices end-to-end; actual prototype/production work requires physical lab presence, equipment operation, and empirical validation that remains human-dependent. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously builds and tests prototype photonic devices; this remains firmly in the domain of skilled human technicians operating physical equipment. |
Maintain clean working environments, according to clean room standards.
19CI 7–30 · exposure 13 · augmentation 38 · importance 4.2/5 · click for rater detail
Maintain clean working environments, according to clean room standards.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Cleanroom environments are specialized, capital-heavy, and concentrated in regulated sectors (semiconductors, pharmaceuticals, biotech) with slower organizational adoption of radical automation. Most adoption remains in sensor monitoring and process logging rather than autonomous task execution. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing and semiconductor-adjacent industries have some automation adoption for monitoring, but actual cleanroom maintenance tasks (physical cleaning, gowning procedures) remain manual with slow uptake of robotic solutions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered monitoring dashboards, predictive alerts for contamination risk, and automated logging of environmental parameters usefully assist technicians in detecting anomalies and scheduling maintenance, but the technician remains responsible for hands-on remediation and compliance validation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI-based sensors and monitoring systems can help track particulate levels and flag contamination risks, offering some assistance, but do not meaningfully augment the physical maintenance work itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Maintaining clean environments requires physical presence, inspection judgment, and responsive adjustment of variables (airflow, humidity, particle filtration). While AI could monitor sensors and trigger alerts, the hands-on cleaning, containment protocols, and real-time environmental tuning demand human intervention and cannot achieve the 50% time-saving threshold autonomously. |
| Task automatability | claude-sonnet-5 | 1/5 | Physically cleaning, organizing, and maintaining a cleanroom to strict contamination-control standards requires manual dexterity and physical presence that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Clean room standards (ISO 14644, GMP) mandate specific protocols, certifications, and documented compliance; many require a qualified human to certify conditions and sign off on maintenance logs. Regulatory and liability frameworks create strong barriers to autonomous substitution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no licensing is strictly required, cleanroom protocols often mandate specific certified procedures and human accountability for contamination control, creating moderate organizational and safety-driven friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Cleanroom maintenance automation (environmental control systems, monitoring) is capital-intensive and still requires technician oversight, validation, and periodic manual intervention. The all-in cost of these systems often exceeds the loaded labor cost for skilled technicians who manage the space. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI systems cannot perform this physical maintenance task, so there is no viable AI cost comparison; any robotic solution would require expensive specialized hardware exceeding human labor costs for this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Sensor monitoring systems and environmental logging exist in production cleanrooms, but no deployed AI system can autonomously perform the physical maintenance, validation, and corrective actions required by clean room standards. Current technology is limited to alerting humans to breaches, not remedying them. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | There are no deployed AI products that autonomously maintain cleanroom cleanliness standards; this remains a physical human task with occasional robotic assistance in narrow niches like specialized cleaning robots, not general adoption. |
Assemble components of energy-efficient optical communications systems involving photonic switches, optical backplanes, or optoelectronic interfaces.
19CI 7–30 · exposure 13 · augmentation 38 · importance 3.2/5 · click for rater detail
Assemble components of energy-efficient optical communications systems involving photonic switches, optical backplanes, or optoelectronic interfaces.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics manufacturing remains relatively specialized and consolidated in advanced facilities; while some leading semiconductor and telecom equipment makers adopt robotic assembly, the sector overall has low digital maturity and slow adoption of general-purpose manufacturing automation outside a few high-volume production lines. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing is a specialized, moderately digitized niche within electronics/telecom hardware where automation exists for some processes but general AI adoption for hands-on assembly is slow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered vision systems and automated test fixtures can assist technicians by verifying component alignment, flagging defects, and logging assembly parameters; however, the core task of hands-on precision assembly and troubleshooting still relies heavily on human expertise, limiting the productivity multiplier. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with design simulation, defect detection, or test data analysis adjacent to this task, but offers little direct help with the physical assembly work itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI vision systems can inspect components and guide assembly sequences, the precise alignment and mechanical assembly of delicate photonic components (switches, backplanes, optoelectronic interfaces) requires skilled manipulation and real-time tactile feedback that current robotic systems struggle with reliably at scale. Only isolated sub-steps like parts verification benefit from meaningful automation. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on precision assembly task requiring physical manipulation of delicate optical components, alignment, and calibration that current AI cannot perform end-to-end without robotic hardware not yet deployed for this purpose. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Quality and reliability standards in optical communications systems are stringent; optical alignment errors can cause system failure, creating strong liability and organizational incentives to retain human expertise and sign-off. Safety-critical optical interfaces and regulatory compliance in telecommunications add further friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not formally licensed work, precision optical assembly involves specialized training, quality/reliability requirements, and physical dexterity that create strong practical barriers to automation, though no regulatory sign-off is required. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialized robotic systems and vision inspection infrastructure for photonic assembly carry high capital and integration costs that typically exceed the loaded wage of a skilled photonics technician, especially when accounting for setup, maintenance, and quality oversight requirements. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing physical assembly, so any AI-based approach would require expensive custom robotics with no cost advantage over skilled technician labor. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Robotic assembly systems exist in manufacturing but generally perform simpler, standardized tasks; photonic component assembly involves micron-level precision, fragile optical alignments, and frequent design variations that deployed systems have not reliably mastered in production environments. Specialized equipment exists for specific configurations but not general-purpose automation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously assembles photonic switches or optoelectronic interfaces; this remains a manual/technician-driven process supported at best by specialized fixtures and test equipment, not AI systems. |
Fabricate devices, such as optoelectronic or semiconductor devices.
19CI 7–30 · exposure 13 · augmentation 50 · importance 3.0/5 · click for rater detail
Fabricate devices, such as optoelectronic or semiconductor devices.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While semiconductor and photonics manufacturing has begun piloting AI-driven inspection and process control, adoption of autonomous fabrication is slow. The sector is capital-intensive, serves specialized markets, and maintains conservative risk postures around quality and yield. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Semiconductor manufacturing uses automation and robotics extensively but AI-driven autonomous fabrication replacing technicians is not a common adoption pattern; physical production sectors adopt AI more slowly than information sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist technicians through predictive maintenance alerts, real-time parameter recommendations, and automated defect detection, moderately improving technician productivity. However, augmentation is limited by the physical, hands-on nature of the work. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with process monitoring, defect detection, yield optimization, and equipment diagnostics, improving technician productivity without replacing the physical fabrication work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Photonic device fabrication requires precise physical manipulation in controlled cleanroom environments with real-time quality monitoring. While AI can assist in process control and parameter optimization, the hands-on assembly and quality assurance steps remain heavily manual and difficult to fully automate with current robotics at the speed and precision required. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical fabrication of optoelectronic/semiconductor devices requires hands-on cleanroom operations, equipment handling, and precision manual/mechanical skill that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Photonic device manufacturing operates under strict quality, safety, and regulatory standards (ISO, cleanroom certification); human oversight and sign-off are often mandatory. Many customers and industries require human responsibility for device performance, creating legal and contractual barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed in the legal sense, cleanroom safety protocols, equipment certification, and quality control processes create significant organizational and procedural barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current specialized fabrication equipment and AI vision systems are capital-intensive, and integration costs remain high relative to the cost of skilled technician labor. The relatively low volume of many custom photonic devices limits economies of scale that would drive costs down. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot replace the physical labor and specialized equipment operation involved, so there is no meaningful AI cost basis to compare against human wages for this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI-assisted inspection and defect detection systems exist in semiconductor manufacturing, but end-to-end autonomous photonic device fabrication is not yet reliably deployed at production scale. Most deployed solutions are narrow (e.g., defect classification) rather than handling the full fabrication workflow. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically fabricates devices; robotics and process automation exist but require human technicians for setup, handling, and troubleshooting. |
Set up or operate assembly or processing equipment, such as lasers, cameras, die bonders, wire bonders, dispensers, reflow ovens, soldering irons, die shears, wire pull testers, temperature or humidity chambers, or optical spectrum analyzers.
18CI 5–30 · exposure 13 · augmentation 38 · importance 3.8/5 · click for rater detail
Set up or operate assembly or processing equipment, such as lasers, cameras, die bonders, wire bonders, dispensers, reflow ovens, soldering irons, die shears, wire pull testers, temperature or humidity chambers, or optical spectrum analyzers.
18| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics manufacturing remains capital-intensive and specialized, with adoption concentrated in high-volume semiconductor and optical component fabs that justify custom automation. Broader sectors and smaller manufacturers still rely on skilled technicians; adoption is sector-specific and slow outside large integrated fabs. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Photonics manufacturing and hardware assembly are physical, low-digitization environments where AI/robotic adoption for equipment operation remains slow and pilot-stage at best. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by monitoring equipment parameters, logging data, and suggesting maintenance or process adjustments, raising technician situational awareness. However, augmentation is limited to data and guidance; the human must remain in control of physical setup and critical operational decisions. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with monitoring equipment data, flagging anomalies in spectrum analyzer outputs, or optimizing process parameters, but it does not materially transform the hands-on setup and operation work itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can monitor sensor data and control some equipment parameters via software interfaces, the task requires physical manipulation of diverse specialized equipment, real-time calibration judgment, and handling of delicate optical/electronics components. Current AI lacks the embodied dexterity and adaptive problem-solving for reliable end-to-end setup and operation of multiple heterogeneous devices. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation of specialized hardware, precise mechanical setup, and hands-on calibration that current AI systems cannot perform without robotic embodiment, which is not standard in this role today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety-critical equipment operation (high-power lasers, thermal chambers, precision bonding) carries significant liability and error costs that create organizational and regulatory friction. Equipment suppliers often require certified human operators for warranty and safety compliance, and specialized knowledge gates adoption. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not formally licensed, this involves specialized technical skill, safety protocols (lasers, soldering), and equipment liability that create organizational and safety-driven friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI integration into existing photonics assembly lines requires custom hardware interfaces, software engineering, and continuous maintenance. The equipment itself is expensive and specialized; full automation would require substantial capital investment and per-task oversight costs that currently exceed the loaded wage of a trained technician. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical equipment operation, so AI cost comparison is essentially moot—human labor remains the only functional option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some individual equipment types have partial automation (e.g., reflow ovens with preset profiles, automated wire bonders in high-volume production), but no general-purpose AI system reliably sets up and operates the full range of photonics assembly equipment listed. Deployed solutions are narrow, equipment-specific, and typically require extensive human oversight. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously sets up or operates photonics assembly equipment like die bonders or wire pull testers; this remains a manual technician task in production environments. |
Adjust or maintain equipment, such as lasers, laser systems, microscopes, oscilloscopes, pulse generators, power meters, beam analyzers, or energy measurement devices.
16CI 5–26 · exposure 13 · augmentation 38 · importance 4.2/5 · click for rater detail
Adjust or maintain equipment, such as lasers, laser systems, microscopes, oscilloscopes, pulse generators, power meters, beam analyzers, or energy measurement devices.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics technician roles are concentrated in specialized research, manufacturing, and professional service sectors with lower overall digital transformation velocity; adoption of autonomous maintenance systems remains minimal and largely experimental. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Photonics manufacturing and lab technician roles are physical, low-digitization environments with minimal AI/robotic adoption for hands-on equipment maintenance. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered diagnostics and predictive maintenance alerts can assist technicians in identifying faults and planning interventions, moderately improving productivity on fault detection and documentation phases of the task. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with diagnostic data interpretation, predictive maintenance alerts, or documentation, but offers little help with the core physical adjustment and calibration work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could diagnose some equipment faults through sensor data analysis, the task requires hands-on physical adjustment and maintenance of precision optical equipment—soldering, alignment, component replacement—which current robotic systems cannot perform reliably without extensive domain-specific setup and supervision. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires hands-on physical manipulation, alignment, and calibration of precision optical/laser hardware, which current AI systems cannot physically perform. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety regulations, equipment-specific certifications, liability for damage to expensive precision instruments, and the need for a licensed technician to authorize and sign off on maintenance create substantial legal and organizational barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement per se, but specialized technical training, safety protocols around lasers, and equipment liability create meaningful practical barriers to any automation attempt. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Specialized robotic systems capable of precise equipment adjustment and maintenance would require significant capital investment and integration costs that vastly exceed the loaded wage of a skilled photonics technician. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing the physical adjustment/maintenance, so the human cost is the only real option, making AI cost comparison moot or unfavorable if forced robotics were used. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably performs the full suite of laser and precision optical equipment maintenance independently; diagnostic AI exists but must be paired with human technicians for actual physical adjustments and calibration. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed products autonomously adjust or maintain photonics lab equipment; this remains a manual technician task requiring physical dexterity and sensory judgment. |
Assemble or adjust parts or related electrical units of prototypes to prepare for testing.
16CI 5–26 · exposure 13 · augmentation 38 · importance 3.4/5 · click for rater detail
Assemble or adjust parts or related electrical units of prototypes to prepare for testing.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Photonics is a specialized, relatively low-volume sector dominated by research labs and small to mid-sized firms with limited digitization. Adoption of advanced automation remains slow; most prototype assembly continues to rely on skilled human technicians rather than deployed AI or robotic systems. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Photonics manufacturing and prototyping is a low-volume, highly manual, physical hardware sector with minimal AI/robotic adoption for fine assembly tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered vision systems could assist in inspection and defect detection during assembly, and generative design tools could suggest assembly sequences. However, the hands-on manual skill and real-time problem-solving required limit the scope of meaningful augmentation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with documentation, testing protocol suggestions, or diagnostics, but offers little direct help with the physical assembly and adjustment process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While some mechanical assembly steps could theoretically be automated, prototypes require frequent adjustments, problem-solving, and precise manual dexterity that current robotic and AI systems struggle with at scale. The variability of prototype designs and the need for real-time visual inspection and tactile feedback make consistent 50% time savings unlikely with today's general-purpose systems. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on physical assembly and precision alignment task involving optical/electrical components that requires manual dexterity and fine-tuning—current AI systems cannot perform this physical manipulation end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Quality assurance and liability concerns are significant—errors in prototype assembly can compromise safety-critical testing. Many organizations require human oversight and sign-off on prototype preparation, and customer expectations for precision and accountability create organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars automation, but the need for precise physical dexterity, specialized tooling, and quality assurance on prototypes creates strong practical barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Custom robotic assembly solutions carry high capital and integration costs that far exceed the wage of a skilled technician, especially for low-volume prototype work. The setup and customization required per prototype design makes AI/automation prohibitively expensive compared to direct human labor. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI/robotic substitute performing this task, so any hypothetical automation would require expensive custom robotics far exceeding technician wages for low-volume prototype work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Specialized robotic assembly systems exist in controlled manufacturing environments, but they are typically task-specific and not deployed for prototype adjustment work. General-purpose AI systems lack the fine-motor control and adaptive problem-solving needed to handle the unpredictability of prototype assembly without significant human intervention. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously assembles or adjusts photonics prototype hardware; robotic assembly for such delicate, low-volume prototyping work remains research-stage at best. |
Set up or operate prototype or test apparatus, such as control consoles, collimators, recording equipment, or cables.
13CI 5–21 · exposure 8 · augmentation 50 · importance 3.5/5 · click for rater detail
Set up or operate prototype or test apparatus, such as control consoles, collimators, recording equipment, or cables.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Photonics technician roles remain concentrated in specialized research and manufacturing environments with low digitization rates and strong institutional preference for certified human expertise; adoption of autonomous AI systems is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Photonics manufacturing and lab environments are physical, specialized settings with low general AI adoption for hands-on hardware tasks, though some automation exists in adjacent digital workflows. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could usefully assist with procedure documentation lookup, sensor data interpretation, or automated diagnostic logging, but the core manual setup and equipment operation tasks require sustained human presence and judgment. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with test planning, data logging, anomaly detection in recorded data, and guidance documentation, though the physical setup itself remains manual. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could potentially assist with routine setup procedures through documentation parsing or automated diagnostics, the task requires physical manipulation, real-time calibration, equipment-specific troubleshooting, and context-dependent decision-making that current autonomous systems cannot perform reliably end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation of optical hardware, cabling, and alignment of precision equipment, which current AI systems cannot perform without robotic embodiment far beyond deployed capability.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety protocols, equipment liability, validation requirements for test apparatus, and regulatory compliance in photonics/physics labs create strong organizational and procedural barriers to full automation; human technicians are typically required for sign-off on sensitive measurements. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement mandates a human specifically, but safety, calibration precision, and equipment liability create meaningful organizational friction against unsupervised automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Integration and oversight costs for any AI assistance (human monitoring, error correction, re-training) would exceed the cost of a photonics technician performing the task directly, given the specialized knowledge and low error tolerance required. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical setup task, so the all-in AI cost is not comparable or is effectively infinite relative to a technician's wage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI systems can independently set up or operate complex photonics test apparatus; this work remains manual and requires skilled technicians. Current automation addresses narrow subtasks (e.g., data logging) rather than full setup or operation workflows. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously sets up or operates photonics test apparatus in production; this remains a manual, hands-on lab task. |
Splice fibers, using fusion splicing or other techniques.
10CI 10–10 · exposure 0 · augmentation 25 · importance 3.8/5 · click for rater detail
Splice fibers, using fusion splicing or other techniques.
10| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Photonics and telecommunications sectors have not adopted automation for fiber splicing despite decades of opportunity. Human technicians remain the norm; digitization pressure is low and the physical, precision-dependent nature of the task keeps it in laggard adoption categories. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Photonics/telecom field technician work is a physical, low-digitization trade with minimal AI agent adoption for hands-on installation tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI might assist in splice-quality monitoring or predictive analytics on fiber performance, but such tools do not materially augment the core manual splicing task itself. The technician remains almost entirely dependent on their own sensorimotor skills and experience. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Modern fusion splicers include automated alignment and loss-estimation software that assists technicians, but this is embedded firmware rather than general AI, offering only incremental assistance. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Fiber splicing requires precise 3D positioning, sub-micrometer alignment, and real-time visual/spectroscopic feedback in a physical environment. Current AI lacks embodied dexterity and the sensorimotor control to position fusion splicers or align fibers reliably at the speeds required; this remains fundamentally a manual skilled task. |
| Task automatability | claude-sonnet-5 | 1/5 | Fiber splicing is a hands-on physical task requiring precise manual manipulation, alignment, and machine operation with fiber optic strands; no AI system can perform this physical manipulation today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While not legally licensed, fiber splicing in critical infrastructure (telecom, data centers) often requires certification and sign-off by qualified technicians. Customer and regulatory expectations for human verification, combined with high cost of splice failures, create moderate organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing law mandates a human specifically for splicing, but the physical dexterity requirement, cleanroom-like precision, and reliance on specialized hardware create strong practical barriers to automation via AI/robotics. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Developing and maintaining a specialized robotic fiber-splicing system would cost orders of magnitude more than training and deploying skilled human technicians. The hardware, calibration, maintenance, and oversight would exceed the loaded wage of experienced photonics technicians. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based substitute performing this physical task, so AI cost is not comparable; a human technician with a splicing machine remains the only viable cost path. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed automated system performs fiber splicing end-to-end at production scale. Experimental robotic splicers exist in research settings but do not achieve the quality, speed, and reliability of human technicians in operational telecommunications or manufacturing environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical fiber splicing; fusion splicers are semi-automated mechanical/optical alignment tools operated by a human technician, not AI-driven robotic systems in production. |
Build prototype optomechanical devices for use in equipment such as aerial cameras, gun sights, or telescopes.
7CI 5–10 · exposure 0 · augmentation 38 · importance 3.7/5 · click for rater detail
Build prototype optomechanical devices for use in equipment such as aerial cameras, gun sights, or telescopes.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Photonics technician roles remain concentrated in specialized, low-digitization sectors (optics manufacturing, aerospace, defense); adoption of AI or robotics for prototype assembly is minimal, with most work still performed by human technicians in legacy facilities. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Precision manufacturing and photonics fabrication are physical, low-digitization sectors with minimal AI-driven displacement of hands-on prototyping work to date. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can provide limited assistance in design optimization, optical simulation, and documentation, but offers minimal real-time support for the hands-on assembly, alignment, and testing work that dominates this task, and cannot meaningfully raise technician productivity on the core work itself. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can assist with design simulation, CAD optimization, and documentation supporting the build process, but the physical assembly itself sees limited direct AI augmentation. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Building prototype optomechanical devices requires precise physical assembly, alignment, and testing of complex optical and mechanical components that current AI cannot perform end-to-end. This task involves hands-on fabrication, calibration, and iterative refinement in the physical domain, which is beyond the scope of deployed AI systems today. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical prototyping of optomechanical devices requires manual assembly, alignment, and fine manipulation of hardware that current AI systems cannot perform end-to-end without robotic embodiment far beyond deployed capability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant organizational and technical barriers exist: prototype work requires specialized equipment, clean room environments, and human judgment on quality and alignment tolerances. Customer requirements and regulatory standards for optical/mechanical systems often mandate human verification and sign-off, creating friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars automation, but precision hardware assembly, safety-critical optics (e.g., gun sights, aerial cameras) and quality control create practical organizational and technical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying robotic systems capable of optomechanical assembly, combined with the specialized hardware, integration, and oversight required, far exceeds the cost of employing a skilled photonics technician for prototype work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for physical prototype construction, so cost comparison favors the human technician entirely; any automation would require expensive custom robotics exceeding labor costs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI products can autonomously build physical optomechanical prototypes. While AI can assist in design simulation and planning, the actual construction, alignment of optical elements, and mechanical integration require robotic systems (which are specialized and not general off-the-shelf solutions) and human technician expertise. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously builds physical optomechanical prototypes; this remains a hands-on fabrication and assembly task performed by skilled technicians. |
Repair or calibrate products, such as surgical lasers.
3CI 0–5 · exposure 0 · augmentation 38 · importance 3.4/5 · click for rater detail
Repair or calibrate products, such as surgical lasers.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Photonics technician roles are in specialized, regulated sectors (medical devices, high-precision manufacturing) with slow digital transformation and strong reliance on skilled human labor. Adoption of AI automation in this domain is negligible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Hands-on hardware repair in medical device manufacturing/servicing is a low-digitization, physical-world sector with minimal AI/robotic adoption for this specific task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide diagnostic support or troubleshooting guides to assist technicians, but current systems offer limited augmentation for the hands-on calibration and repair work that defines this task. The potential for AI-assisted decision-making is modest. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted diagnostic software, predictive maintenance analytics, and guided troubleshooting can help technicians identify issues faster, though the physical calibration work remains manual. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Repair and calibration of surgical lasers requires physical manipulation in tight tolerances, contextual diagnosis of hardware faults, and domain expertise that current AI systems cannot perform end-to-end. While AI might assist in diagnostics, the hands-on repair work remains entirely manual. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical repair and calibration of precision optical/laser hardware requires hands-on manipulation, diagnostics, and sensor alignment that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Medical device repair is heavily regulated (FDA, ISO 13485); calibration of surgical equipment typically requires certification and vendor authorization. Liability for failed repairs on surgical equipment is severe, and manufacturers often legally mandate authorized technicians perform these tasks. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Surgical laser equipment is subject to medical device regulations, safety certifications, and liability concerns that require qualified human technicians to perform calibration and repair. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Surgical laser repair requires a skilled technician with years of training and certification; the loaded labor cost is high. Current AI has no substitute for the hands-on work, making AI-based alternatives either non-existent or far more expensive than deploying human technicians. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for the physical labor involved, so AI cost comparison is not applicable and human labor remains necessary. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product reliably performs surgical laser repair or calibration autonomously in production. This is a specialized technical task requiring embodied robotics and real-time physical adaptation that does not exist at scale today. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously repairs or calibrates surgical lasers; this remains a manual technician task with specialized test equipment. |
Related occupations — Architecture & Engineering
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.