Nuclear Monitoring Technicians
19-4051.02Median wage $110,240/yr6,470 employed (US)Rank #705 of 923 scored · top 76% by substitution
Collect and test samples to monitor results of nuclear experiments and contamination of humans, facilities, and environment.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution19
Exposure23
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
16 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
0%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%24
panel mean rating 2.0/5 → substitution pressure 24/100
Technical feasibility todayw 20%20
panel mean rating 1.8/5 → substitution pressure 20/100
Cost vs. human wagew 15%24
panel mean rating 1.9/5 → substitution pressure 24/100
Adoption barriersw 20%inverted — strong barriers lower the score9
panel mean rating 4.7/5 (barrier strength) → substitution pressure 9/100
Sector adoption velocityw 10%10
panel mean rating 1.4/5 → substitution pressure 10/100
Task breakdown (16 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.
Enter data into computers to record characteristics of nuclear events or to locate coordinates of particles.
50CI 32–67 · exposure 58 · augmentation 75 · importance 4.1/5 · click for rater detail
Enter data into computers to record characteristics of nuclear events or to locate coordinates of particles.
50| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Nuclear plants are heavily regulated, risk-averse, and upgrade infrastructure slowly; while automated data logging is standard in modern facilities, full agent-based data entry without human sign-off faces regulatory and cultural resistance typical of high-consequence industries. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Nuclear monitoring is a highly specialized, low-volume field with slow technology adoption cycles due to safety-critical requirements and regulatory scrutiny, unlike fast-moving information sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems significantly assist technicians by auto-populating forms, flagging anomalies, and cross-referencing coordinate data; human technicians remain responsible for validation and interpretation, with AI providing substantial productivity gains on routine data capture and organization. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted data entry, automated logging, and pattern recognition tools can meaningfully speed up and reduce errors in recording nuclear event data while technicians retain oversight. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Data entry and coordinate recording are highly automatable through OCR, sensor integration, and database APIs; however, interpretation of nuclear event characteristics and quality assurance require human validation, preventing a full 5-level automation score. |
| Task automatability | claude-sonnet-5 | 3/5 | Data entry from instrument readings into computers is a structured, repetitive task that AI/automation can handle, but linking it to physical instrument reading and validation of nuclear event characteristics adds complexity beyond pure transcription. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities operate under NRC/international regulatory frameworks that mandate human monitoring, documentation, and sign-off; automated systems must integrate into licensed human-supervised workflows, creating hard legal and compliance barriers to full substitution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Nuclear facilities operate under strict regulatory and safety oversight, and data accuracy for nuclear event characterization may require certified human verification, creating moderate barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automation of routine data entry is cost-effective compared to technician wages; sensor-to-database pipelines and OCR run at minimal marginal cost once infrastructure exists, yielding significant per-task savings. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automation of structured data entry can be cheap once integrated, but specialized nuclear monitoring systems require custom integration, calibration, and validation that raise the effective cost compared to generic tasks. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed systems (SCADA, automated logging, AI-assisted data entry) reliably perform structured data recording in nuclear facilities; however, the safety-critical nature means human oversight is universally retained, limiting full production autonomy. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | While generic data entry and OCR/automation tools are mature, no widely deployed product specifically automates nuclear event data logging and coordinate location in this specialized domain today. |
Calculate safe radiation exposure times for personnel using plant contamination readings and prescribed safe levels of radiation.
43CI 30–55 · exposure 53 · augmentation 75 · importance 4.5/5 · click for rater detail
Calculate safe radiation exposure times for personnel using plant contamination readings and prescribed safe levels of radiation.
43| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear facilities are laggard adopters in AI automation due to regulatory conservatism, strict licensing requirements, and safety-critical culture. While computational aids exist, autonomous agent deployment for exposure-time calculation is rare in production; most facilities retain human oversight. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear plant operations are a highly regulated, safety-critical, low-digitization-of-decision-making sector with very slow AI adoption for safety-determination tasks.__ |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can powerfully augment technicians by rapidly calculating and flagging exposure scenarios, stress-testing shift schedules against contamination maps, and surfacing regulatory thresholds. A human health physicist remains in the loop to validate and authorize, making this a high-value assistive tool. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI/software tools can strongly assist technicians by rapidly computing safe exposure windows from readings and flagging threshold breaches, improving speed and consistency while a human retains final authorization.__ |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | This task is largely computational and rule-based: AI can ingest contamination readings, apply safety formulas, and output exposure times reliably. The core calculation aligns with known regulatory models and could achieve 50%+ time savings with oversight; however, real-world anomalies and contextual judgment may require some human review. |
| Task automatability | claude-sonnet-5 | 3/5 | The calculation itself (dose rate × time vs. exposure limits) is straightforward math AI could perform reliably, but the task requires trusted real-time contamination readings and integration into safety workflows that current off-the-shelf AI is not deployed to handle end-to-end.__ |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Nuclear safety is highly regulated (NRC, IAEA frameworks); exposure calculations often require licensed operator or health physicist sign-off. Liability and error-cost asymmetry are severe (incorrect exposure can harm workers or public), creating strong regulatory and organizational friction against full automation without human authorization. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Radiation safety calculations affecting personnel exposure are subject to strict nuclear regulatory oversight (e.g., NRC rules) requiring qualified, licensed personnel to verify and approve such determinations, creating hard legal/liability barriers.__ |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Once integrated into a facility's system, AI inference cost is minimal (a fraction of a technician's loaded wage per calculation). Maintenance and integration overhead are modest, making the cost ratio heavily in AI's favor—roughly 5-10× cheaper than a human performing the same calculation. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | The arithmetic is cheap for AI to do, but the surrounding validation, sensor integration, and required certification of any calculation tool add cost, keeping overall cost roughly comparable to a trained technician performing it as part of broader duties.__ |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Decision-support tools and calculators exist in nuclear facilities and regulatory frameworks (e.g., dose-limit applications), but few fully autonomous AI systems are deployed in production for independent exposure-time calculation without human sign-off. Feasibility is middling due to oversight requirements and liability concerns. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | No widely deployed nuclear-plant product autonomously performs this exposure-time calculation in production; such systems remain largely rule-based/manual or embedded in specialized, tightly validated software rather than general AI tools.__ |
Prepare reports describing contamination tests, material or equipment decontaminated, or methods used in decontamination processes.
42CI 32–51 · exposure 53 · augmentation 75 · importance 4.0/5 · click for rater detail
Prepare reports describing contamination tests, material or equipment decontaminated, or methods used in decontamination processes.
42| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear power and waste handling sectors move cautiously on automation, with strong regulatory scrutiny and conservative organizational culture. While some facilities may pilot AI-assisted drafting, adoption of AI for report generation remains slow and shallow due to compliance and safety-culture constraints. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Nuclear and radiological safety sectors are typically slow adopters of AI tools due to regulatory conservatism, safety-critical documentation standards, and limited digitization of specialized workflows. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist technicians by auto-populating report templates, organizing measurement data, drafting narrative summaries, and flagging outliers for review. This reduces manual documentation burden substantially while keeping the licensed technician in control of accuracy and regulatory sign-off, raising overall productivity. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist technicians by drafting report structure, summarizing test data, and ensuring consistent terminology, while the technician retains responsibility for accuracy and final certification. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can readily extract data from contamination logs, generate structured reports, and summarize decontamination procedures with high consistency. Report formatting and narrative generation are largely routine, though some domain-specific validation and human review of technical accuracy remain necessary; this still achieves substantial time savings (>50%) for the bulk of the work. |
| Task automatability | claude-sonnet-5 | 3/5 | Report drafting from structured test data and templates can be substantially automated with LLMs, but requires accurate integration of technical measurements and domain-specific compliance language that still needs human verification.4 |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities operate under strict NRC and international regulatory frameworks requiring that contamination reports be prepared, reviewed, and certified by licensed or qualified personnel. Liability and regulatory compliance create hard barriers: a licensed technician or engineer must legally sign off on decontamination records, preventing end-to-end automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Nuclear facilities operate under strict regulatory oversight (NRC and similar bodies) requiring certified technician sign-off on contamination and decontamination records, creating strong liability and compliance barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI inference and integration costs are low, but the requirement for nuclear technician review and regulatory oversight means supervision remains expensive. The cost savings on drafting must be weighed against mandatory expert verification, making the all-in cost roughly comparable to human report writing. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI drafting could cut time spent on report writing significantly, but the need for specialized technician review and regulatory accuracy checks keeps overall cost savings moderate rather than order-of-magnitude. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | LLMs and document-generation products can produce initial draft reports from structured data, but nuclear regulatory compliance demands high accuracy and regulatory review. Deployed systems exist (e.g., enterprise AI writing assistants), but material error rates and the requirement for expert sign-off limit reliable full-stack deployment today. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | General-purpose AI writing tools can draft such reports today, but no widely deployed nuclear-industry-specific product reliably generates compliant contamination/decontamination reports in production. |
Inform supervisors when individual exposures or area radiation levels approach maximum permissible limits.
31CI 25–38 · exposure 42 · augmentation 75 · importance 4.5/5 · click for rater detail
Inform supervisors when individual exposures or area radiation levels approach maximum permissible limits.
31| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear facilities are highly regulated, risk-averse organizations with entrenched licensing and compliance frameworks; automation of critical safety reporting faces strong institutional and legal friction, limiting rapid adoption. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Nuclear and radiation safety sectors are conservative, highly regulated, and slow to adopt new AI-driven monitoring/alerting workflows relative to information-sector benchmarks, despite existing automated instrumentation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted monitoring can significantly augment technician productivity by automatically flagging anomalies, maintaining exposure logs, predicting trend escalations, and drafting supervisor notifications, while the technician retains final judgment and accountability. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled monitoring dashboards, predictive alerts, and automated threshold tracking meaningfully augment technicians by continuously monitoring conditions and flagging approaching limits faster and more consistently than manual checks. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While radiation monitoring sensors can automatically trigger alerts when levels approach thresholds, the task requires human judgment about contextual permissible limits, exposure assessment complexities, and appropriate supervisor notification protocols that current AI systems cannot reliably perform end-to-end. |
| Task automatability | claude-sonnet-5 | 3/5 | Threshold-based alerting on radiation exposure/area levels is a straightforward automatable rule-based function, but the task as stated includes judgment about escalation and communication with supervisors that still benefits from human framing.There is meaningful automatable structure but not full end-to-end replacement of the technician role. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities operate under strict regulatory oversight (NRC, OSHA) that mandates qualified personnel perform exposure monitoring and reporting; supervisory notification of radiation hazards is legally required and typically requires a licensed or certified technician's judgment and accountability. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear safety is heavily regulated (NRC and equivalent bodies), with radiation exposure reporting requiring qualified personnel and strict chain-of-custody/documentation and accountability, making this a high-barrier regulated function. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Automated sensor systems are relatively inexpensive, but the technician's role in interpreting data, cross-referencing limits, and communicating findings to management adds significant human-labor cost that sensor automation alone does not replace. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Sensor/alarm systems are cheap to run, but integration with existing nuclear safety infrastructure, validation, and required human oversight keep all-in costs comparable to human-technician cost rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed monitoring systems exist that automatically log radiation levels and can generate alerts, but the judgment about when and how to inform supervisors—especially regarding individual exposure cumulative history and regulatory interpretation—requires human oversight and is not fully automated in production systems. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Radiation monitoring systems with automated alarms and threshold triggers are deployed in nuclear facilities today, but the specific act of 'informing supervisors' with context and judgment is typically still a human-mediated communication process layered on top of automated sensor alerts. |
Analyze samples, such as air or water samples, for contaminants or other elements.
23CI 20–25 · exposure 25 · augmentation 50 · importance 4.3/5 · click for rater detail
Analyze samples, such as air or water samples, for contaminants or other elements.
23| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear and environmental sectors are highly regulated and risk-averse; adoption of AI in monitoring tasks has been slow and limited to pilot programs. Most organizations rely on traditional instrument operation and human technician expertise, with minimal deployment of autonomous or fully automated systems in production. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Nuclear and environmental monitoring sectors are conservative, safety-critical, and slow to adopt AI-driven analysis tools compared to fast-moving information/finance sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by auto-interpreting spectroscopy or chromatography data, flagging anomalies, and generating preliminary reports, which speeds technician workflows. However, the physical sample work and regulatory sign-off remain human responsibilities, making augmentation useful but not transformative. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with data analysis, pattern recognition in spectral data, and flagging anomalies, improving technician efficiency, though the technician remains essential for validation and judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist in data interpretation and pattern recognition from analytical instrument outputs, the physical task of sample collection, preparation, and running laboratory instruments requires hands-on technician work. AI might automate 20–30% of the analytical data interpretation, but the core sampling and measurement work remains manual. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical sample collection and handling cannot be automated by current AI; lab instruments can automate some detection/quantification but interpretation and regulatory sign-off require human expertise, so overall time savings fall well short of 50% end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Nuclear and environmental monitoring are heavily regulated (NRC, EPA); regulatory frameworks require qualified, licensed technicians to perform or validate sampling and analysis. Legal liability for incorrect contaminant readings, combined with strict chain-of-custody requirements and licensing mandates, creates strong barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear monitoring is heavily regulated (e.g., NRC oversight), requiring certified technicians and documented chain-of-custody/quality assurance, making full automation legally and organizationally very difficult. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Nuclear monitoring technicians earn reasonable wages, and the specialized equipment, compliance overhead, and need for qualified human oversight make AI-assisted analysis still relatively expensive compared to displacing the technician role entirely. The compliance and safety costs remain high. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Specialized lab equipment, calibration, and certified technician oversight keep costs high; AI software assistance provides only marginal cost reduction relative to skilled labor and equipment costs already required. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product fully autonomously handles sample analysis from intake to reporting in nuclear or regulated contexts. Analytical instruments produce data that AI can interpret, but actual sample handling, instrument operation, and quality assurance chains require human oversight and cannot yet be fully automated in production environments. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Analytical instruments (spectrometers, chromatographs) with software assistance exist in production labs, but full AI-driven interpretation and contamination analysis for nuclear safety is not deployed as an autonomous product. |
Determine intensities and types of radiation in work areas, equipment, or materials, using radiation detectors or other instruments.
19CI 9–30 · exposure 17 · augmentation 50 · importance 4.5/5 · click for rater detail
Determine intensities and types of radiation in work areas, equipment, or materials, using radiation detectors or other instruments.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear facilities are highly regulated, capital-intensive, and conservative in operational changes. While automated sensor networks exist, human technicians remain the required decision-makers; adoption of unsupervised AI interpretation is minimal and slow due to regulatory and safety culture barriers. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear monitoring is a highly specialized, safety-critical, low-digitization physical field where AI adoption for core measurement tasks is minimal and slow-moving compared to information-sector work. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems powerfully augment technician capability by providing real-time sensor data integration, trend analysis, anomaly detection, and automated alerts across distributed monitoring points. These tools help technicians detect patterns and respond faster while maintaining required human oversight and decision authority. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help analyze sensor data trends, flag anomalies, or assist in report generation, but the core physical detection and situational judgment task sees limited AI augmentation currently. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While radiation detection instruments can automatically record measurements, determining *types* of radiation and interpreting intensities in context requires human judgment about safety protocols, shielding adequacy, and operational decisions. Current AI cannot autonomously assess the full risk picture or make safety-critical determinations that meet regulatory standards. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical deployment and operation of radiation detection instruments in real work areas, which AI cannot perform without robotic embodiment; the physical measurement act itself is not automatable by current AI systems. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Nuclear facilities operate under strict regulatory frameworks (NRC, OSHA) that mandate licensed or certified personnel to conduct radiation monitoring and interpret results for safety compliance. Legal liability for radiation exposure and contamination creates asymmetric error costs that require human accountability and licensure. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Nuclear safety work is heavily regulated (NRC and similar bodies), often requiring certified technicians to perform and document radiation measurements, creating strong regulatory and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Modern detection equipment is capital-intensive and requires calibration, maintenance, and integration into monitoring networks. While measurements are cheap once deployed, the total system cost (hardware, integration, technician oversight) remains comparable to or higher than direct human labor for routine monitoring tasks. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Radiation detection instruments and sensor networks have upfront and calibration costs, and human technicians remain necessary for interpretation and physical presence, so cost savings from 'AI' specifically are minimal beyond existing instrumentation. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Automated radiation detectors with digital readouts exist and are widely deployed, but systems that autonomously interpret readings, classify radiation types, and make compliance decisions without human oversight are limited to narrow, pre-configured scenarios. Production systems still require technician validation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product independently performs physical radiation surveys; sensor hardware with automated logging exists but the task as a whole (physical monitoring, judgment on intensity/type across contexts) is not handled by AI products in production. |
Monitor personnel to determine the amounts and intensities of radiation exposure.
19CI 18–20 · exposure 25 · augmentation 50 · importance 4.5/5 · click for rater detail
Monitor personnel to determine the amounts and intensities of radiation exposure.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear facilities are heavily regulated, risk-averse, and maintain strict protocols around personnel monitoring; adoption of AI automation is slow and limited to data-logging support roles rather than core decision-making. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear and radiation safety sectors are highly regulated, slow-moving, and prioritize proven hardware/procedural systems over AI adoption, making this a laggard sector for AI deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist technicians by automating data collection, trend analysis, and alert generation for anomalous exposures, improving their efficiency and decision support while the technician retains responsibility for final exposure determinations and regulatory reporting. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by analyzing exposure trends, flagging anomalies in dosimetry data, and supporting record-keeping, providing moderate productivity benefits while humans retain monitoring and compliance responsibilities. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While radiation dosimetry instruments can automatically record exposure data, determining personnel exposure amounts and intensities requires judgment about contextual factors (work duration, shielding effectiveness, individual circumstances) that current AI systems cannot reliably assess without substantial human oversight and calibration. |
| Task automatability | claude-sonnet-5 | 2/5 | While sensor data collection and readings can be automated with dosimetry hardware, interpreting exposure patterns, ensuring correct personnel monitoring protocols, and responding to anomalies requires physical presence, calibration, and judgment that current AI cannot fully replace end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities operate under strict regulatory frameworks (NRC, DOE) that require licensed nuclear technicians to personally monitor and certify radiation exposure; legal accountability and safety-critical nature create hard barriers to substitution with autonomous AI systems. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear safety regulations (NRC, IAEA standards) mandate certified personnel to monitor and interpret radiation exposure, and liability for exposure errors is extremely high, creating strong legal and safety barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-assisted data logging is cheaper than manual recording, but the core task requires licensed technician expertise and oversight that cannot be displaced by automation, so cost savings are marginal relative to the full human labor burden. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Physical monitoring equipment, calibration, and compliance oversight still require significant human involvement and specialized equipment costs, so AI does not yet offer a dramatic cost advantage over trained technicians in this role. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Radiation monitoring instruments produce automated readings, but no deployed AI product independently determines personnel exposure levels and interprets results for regulatory compliance; human technicians remain the essential decision-maker in production nuclear facilities. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Deployed dosimetry and radiation monitoring systems exist and are mature, but they are sensor/hardware-based rather than AI-driven; AI-based interpretation or autonomous decision-making in this safety-critical context is not yet standard in production. |
Set up equipment that automatically detects area radiation deviations and test detection equipment to ensure its accuracy.
16CI 9–23 · exposure 20 · augmentation 38 · importance 4.1/5 · click for rater detail
Set up equipment that automatically detects area radiation deviations and test detection equipment to ensure its accuracy.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear facilities are heavily regulated, risk-averse organizations with strong human-in-the-loop requirements and slow technology adoption cycles. Displacement of technician roles through automation is minimal and constrained by compliance requirements. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear and radiation safety sectors are highly regulated, safety-critical, and slow to adopt autonomous physical systems, showing minimal AI-driven displacement in this domain. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted data analysis and real-time anomaly detection from radiation sensors could support technicians in identifying potential deviations faster, though the core hands-on setup and calibration testing remain human-dependent. Augmentation exists but is moderate rather than transformative. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with data logging, anomaly flagging, or predictive maintenance scheduling for detection equipment, but it offers limited help with the physical setup and calibration itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could assist with some aspects like data monitoring and anomaly detection from sensor streams, the physical setup of radiation detection equipment and hands-on testing in hazardous environments requires human presence and judgment. Only a narrow portion of oversight could be automated, far short of the 50% time-saving bar. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical setup, calibration, and hands-on accuracy testing of radiation detection equipment requires manual manipulation and physical verification that current AI systems cannot perform end-to-end.','rationale2unused' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities operate under stringent regulatory frameworks (NRC, IAEA standards) requiring licensed personnel to perform equipment setup and testing. Physical presence, legal responsibility, and safety certification form hard barriers preventing full automation or substitution by non-licensed systems. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Nuclear facilities operate under strict regulatory oversight (e.g., NRC) requiring certified personnel to verify radiation detection accuracy, creating strong licensing and liability barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialized radiation monitoring technicians earn substantial wages in safety-critical roles. AI systems for anomaly detection are available but do not eliminate the technician's hands-on labor, and the cost of human oversight plus AI infrastructure would likely exceed a technician's marginal cost. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven substitute performing physical equipment setup and calibration, so the human technician remains the only cost-effective option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI system reliably performs end-to-end radiation equipment setup and calibration testing. Some research exists in automated anomaly detection from radiation sensor data, but production systems do not independently set up, test, or validate detection equipment accuracy in real nuclear facilities. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously installs or physically calibrates radiation monitoring equipment; this remains a hands-on technician task with no production-grade robotic substitute. |
Provide initial response to abnormal events or to alarms from radiation monitoring equipment.
13CI 0–25 · exposure 17 · augmentation 63 · importance 4.5/5 · click for rater detail
Provide initial response to abnormal events or to alarms from radiation monitoring equipment.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Nuclear power and monitoring are highly regulated, conservative sectors with slow AI adoption. Most facilities use traditional SCADA and monitoring systems; autonomous agent deployment for safety-critical initial response is extremely limited. Barriers and risk aversion keep velocity low. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear facilities are highly regulated, safety-conscious, and slow to adopt autonomous systems for critical safety response functions, showing minimal AI penetration in this specific function. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered dashboards, anomaly alerts, and data synthesis significantly assist technicians by highlighting abnormal patterns, aggregating sensor data, and reducing response latency. The human remains in the loop and decision-making authority, but their productivity and situational awareness are measurably enhanced. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-enabled monitoring systems can provide real-time alerts, sensor fusion, and decision-support data to help technicians triage and respond faster, though the human must still act on it. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could analyze sensor data and detect anomalies in real-time, the task requires expert judgment about severity, context-specific decision-making (e.g., distinguishing false alarms from genuine hazards), and coordinated emergency response protocols that involve human authority and accountability. No current system reliably handles the full end-to-end response to nuclear alarms at >50% time savings with equal quality. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires immediate physical presence, situational judgment, and safety-critical decision-making in response to unpredictable real-world events, which current AI cannot execute end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities operate under strict NRC, IAEA, and other regulatory frameworks that explicitly require licensed technicians to respond to and assess radiation alarms. Legal liability, chain-of-custody, and safety-critical certification create hard barriers to full automation; a qualified human must make the initial determination. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear safety regulations require certified, licensed personnel to respond to radiation alarms, with strict liability, legal, and regulatory oversight making substitution essentially prohibited. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | While anomaly-detection AI is inexpensive to run, the oversight, validation, and human expert review required to make initial-response decisions mean total cost remains comparable to or higher than a technician's marginal time for this task. The high-stakes nature demands human verification. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot yet perform the physical response and on-site judgment required, so there is no viable substitute cost comparison; a human responder is mandatory. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed anomaly-detection and monitoring systems exist in nuclear facilities and can flag abnormal readings reliably; however, initial response requires human judgment about severity assessment, cause determination, and decision-making under uncertainty. Products assist but do not perform the full task autonomously in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously handles initial physical response to radiation alarms; this remains a research-stage concept at best given the safety-critical physical response requirement. |
Collect samples of air, water, gases, or solids to determine radioactivity levels of contamination.
11CI 0–23 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail
Collect samples of air, water, gases, or solids to determine radioactivity levels of contamination.
11| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear facilities operate in highly regulated, capital-intensive, slow-moving sectors with strong institutional conservatism and safety-first cultures that strongly resist automation of critical monitoring tasks without regulatory approval and extensive validation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear monitoring is a highly specialized, low-digitization physical sector with minimal AI agent deployment for hands-on sampling tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist technicians by processing historical data, predicting contamination hotspots, automating calibration checks, and analyzing radiation detector readings, thereby improving sampling efficiency and decision-making while the technician remains responsible for collection and compliance. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with planning sampling locations, interpreting sensor data, or logging results, but offers little direct assistance to the physical act of sample collection itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist in data analysis and remote monitoring, physical sample collection from potentially hazardous environments requires human presence, dexterity, and real-time decision-making about safe handling. Current robots cannot reliably perform the full task of sample collection in radioactive settings with equivalent safety and accuracy. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical sample collection from air, water, gas, or solid sources requires on-site presence, manual manipulation of equipment, and navigation of contaminated or hazardous environments, which current AI systems cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strict nuclear regulatory requirements (NRC, IAEA) mandate licensed personnel for sample collection and chain-of-custody procedures; liability and safety standards require human accountability and real-time judgment in contaminated environments, creating hard legal and operational barriers. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Nuclear safety regulations, radiation protection protocols, and certification requirements for handling radioactive materials impose strong barriers, alongside liability concerns around contamination sampling accuracy. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic sample collection systems for radioactive environments are expensive to build, maintain, and specialize, making them cost-prohibitive compared to trained human technicians, especially given the relatively modest labor cost for this role. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven substitute for physical sample collection, so cost comparison favors the human worker who performs the actual physical task; robotics for this are expensive, specialized, and not AI-cost-competitive. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Remote monitoring systems and some automated detectors exist for radiation measurement, but routine physical sample collection in contaminated environments still relies on human technicians in deployed settings. No mature automated system has replaced the core collection task across production nuclear facilities. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed products autonomously collect radioactive samples in the field today; this remains a manual technician task supported at most by automated fixed sensors, not mobile sample collection. |
Decontaminate objects by cleaning with soap or solvents or by abrading with wire brushes, buffing wheels, or sandblasting machines.
11CI 0–23 · exposure 13 · augmentation 13 · importance 3.9/5 · click for rater detail
Decontaminate objects by cleaning with soap or solvents or by abrading with wire brushes, buffing wheels, or sandblasting machines.
11| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear facilities are conservative, highly regulated sectors with slow digitization. Adoption of autonomous decontamination robots remains experimental and limited to large facilities; the majority of decontamination work is still manual. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear facility physical maintenance is a highly specialized, low-digitization, safety-regulated sector with minimal AI adoption for physical decontamination tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI and robotics can assist with repetitive abrasive work, reducing technician exposure. However, the human must remain in control and oversight of decontamination validation and safety endpoints, limiting augmentation impact. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers essentially no meaningful assistance to the physical act of scrubbing, sandblasting, or buffing contaminated objects. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While mechanical cleaning processes (sandblasting, wire brushing) are partially automatable with robotics, the task requires judgment about decontamination endpoints, material compatibility, and safety verification. Current AI cannot reliably perform end-to-end decontamination with equal quality without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a hands-on physical decontamination task requiring manual manipulation of tools like wire brushes and sandblasters on contaminated objects; no AI system can perform this physical cleaning work. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear decontamination is heavily regulated (NRC, DOE); a licensed nuclear technician must perform, verify, and sign off on decontamination work. Regulatory mandates and liability asymmetry (contamination release is catastrophic) create hard legal barriers to autonomous substitution. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Nuclear decontamination work involves radiation safety regulations, certification requirements, and strict procedural/safety oversight, creating substantial regulatory and liability barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialized robotic decontamination systems are capital-intensive and require integration, maintenance, and remote operation oversight. The all-in cost often exceeds the loaded wage of a trained decontamination technician, particularly for varied or infrequent tasks. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI software has no direct cost application here since the task is physical manual labor; any automation would require expensive specialized robotics, not cheaper than human labor. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Robotic systems exist for repetitive abrasive cleaning in controlled environments, but deployed solutions are narrowly scoped (specific geometries, materials). No off-the-shelf AI system reliably handles the full task across the variety of nuclear decontamination scenarios without custom engineering and human validation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical decontamination cleaning; this remains entirely a manual/robotic-hardware task outside current AI product scope, though specialized robotics exist separately from 'AI' systems. |
Instruct personnel in radiation safety procedures and demonstrate use of protective clothing and equipment.
9CI 0–18 · exposure 8 · augmentation 38 · importance 4.4/5 · click for rater detail
Instruct personnel in radiation safety procedures and demonstrate use of protective clothing and equipment.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Radiation safety training is heavily regulated and conservative; facilities rely on certified human instructors and legal accountability. Adoption of AI for autonomous safety instruction is effectively zero in this sector. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear facilities are highly regulated, safety-critical, and slow to adopt AI for hands-on safety training compared to fast-moving digital sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist by generating training materials, quizzes, or reference documents, but the core task—live demonstration and real-time safety instruction—remains human-centric with minimal augmentation potential. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help create training materials, simulations, and knowledge checks that support the technician, improving efficiency of the instructional preparation even though delivery remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task fundamentally requires live demonstration and real-time interaction with personnel, including physical modeling of protective equipment use and answering dynamic safety questions. Current AI cannot physically demonstrate donning procedures or provide the embodied, interactive instruction that radiation safety demands. |
| Task automatability | claude-sonnet-5 | 2/5 | Training content generation and quizzes can be AI-assisted, but live instruction and hands-on demonstration of protective equipment use requires physical presence and judgment, limiting end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Regulatory requirements (NRC, OSHA, facility-specific licensing) mandate that qualified, credentialed humans conduct radiation safety instruction and sign off on personnel competency. Liability for radiological incidents creates strong legal barriers to AI substitution. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Nuclear safety training often requires certified personnel and regulatory compliance (NRC/OSHA), with liability for improper training being high, creating strong barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of developing, deploying, and maintaining AI systems to replace live instruction, plus the liability and compliance oversight required, would far exceed the cost of a trained technician delivering the instruction. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply produce training materials, but the physical demonstration and certification portions still require a paid technician on-site, keeping overall cost comparable to human-led training. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can autonomously instruct personnel in safety procedures or demonstrate physical equipment use in production settings. While AI can generate instructional content, actual hands-on demonstration and safety certification remain human-exclusive functions. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product currently delivers hands-on radiation safety training or physically demonstrates equipment use; this remains a human-delivered, in-person task. |
Brief workers on radiation levels in work areas.
6CI 0–13 · exposure 8 · augmentation 38 · importance 4.6/5 · click for rater detail
Brief workers on radiation levels in work areas.
6| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear sector adoption of AI for safety-critical communication is negligible; this task sits in a heavily regulated, compliance-driven industry with strict personnel licensing and minimal digital transformation in worker training protocols. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear facilities are highly regulated, safety-critical, low-digitization-adoption environments where AI deployment for safety briefings is essentially nonexistent. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could potentially assist a technician by preparing summary data or drafting talking points, but the core task—live communication with workers about radiation safety—requires human judgment and presence that AI augmentation cannot substantially enhance. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help aggregate sensor data, generate summaries, and draft briefing materials, improving technician efficiency, though the human must still verify and deliver the briefing. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Briefing workers on radiation levels requires live communication, reading audience comprehension, and making safety judgments tailored to specific personnel and contexts. Current AI cannot reliably deliver nuanced, real-time safety communications that meet regulatory and operational standards. |
| Task automatability | claude-sonnet-5 | 2/5 | Communicating current radiation readings and safety instructions to workers requires real-time site-specific data integration, verbal judgment, and accountability that current AI cannot fully replace end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Hard regulatory and legal barriers exist: nuclear facilities require qualified, licensed personnel to conduct safety briefings, workers must receive instruction from accountable operators, and liability for miscommunication or omission rests with a named human responsible for accuracy. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear safety communications are heavily regulated (NRC and similar bodies) requiring qualified technicians to monitor and communicate radiation hazards, with strict liability and certification requirements. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A nuclear monitoring technician performing this briefing is a high-wage, specialized role. The cost of AI infrastructure, oversight, and potential liability exposure far exceeds the marginal cost of the human performing it themselves. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply generate reports, but the human briefing, liability, and site-specific verification still require a trained technician, keeping overall cost comparable to or exceeding automation savings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs live worker safety briefings on radiation in operational nuclear facilities. This task demands accountability, legal liability attachment to a qualified human, and regulatory compliance that rules out autonomous AI systems. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously briefs workers on radiation hazards in production nuclear settings; this remains a human safety-critical communication role. |
Determine or recommend radioactive decontamination procedures, according to the size and nature of equipment and the degree of contamination.
6CI 0–13 · exposure 8 · augmentation 38 · importance 4.1/5 · click for rater detail
Determine or recommend radioactive decontamination procedures, according to the size and nature of equipment and the degree of contamination.
6| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear facilities are slow-moving, highly regulated, and conservative in automation adoption. This is not a sector showing early AI adoption; compliance and safety culture favor proven human expertise over algorithmic replacement. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear industry is a highly regulated, safety-critical, low-digitization sector with minimal AI agent deployment for safety-critical operational decisions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI might assist a technician by analyzing historical contamination data or providing reference lookups for decontamination methods, but the core judgment—assessing contamination severity and recommending context-specific procedures—remains a human domain requiring domain knowledge, real-time sensor interpretation, and regulatory accountability. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by providing quick access to decontamination protocols, historical data, and decision-support calculations, but the technician retains full responsibility for the determination. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires judgment about contamination severity, equipment characteristics, and procedure selection in complex, safety-critical conditions. Current AI lacks the multimodal sensing, real-time assessment of contamination patterns, and domain expertise to determine appropriate decontamination procedures independently without human expert oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical assessment of contamination levels, equipment-specific judgment, and safety-critical decision-making that current AI cannot perform end-to-end without extensive human verification.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities are heavily regulated (NRC, international standards); decontamination procedures must be recommended or approved by licensed nuclear technicians or health physicists. Legal and liability barriers are absolute—only qualified human professionals can sign off on contamination assessments and decontamination plans. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear safety regulations require certified, trained personnel to make and be accountable for decontamination decisions, with strict regulatory oversight (e.g., NRC) and severe liability for errors. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A licensed nuclear monitoring technician commands a high loaded wage ($60K–$80K+ annually), and the cost of integrating, validating, and maintaining AI systems for this safety-critical task would exceed human labor costs given the liability and regulatory burden. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI could assist in referencing protocols, the human technician's on-site assessment, measurement, and accountability cannot be meaningfully replaced, so cost savings are limited to minor efficiency gains. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can autonomously assess radioactive contamination, evaluate equipment-specific factors, and recommend decontamination procedures in production nuclear environments. This requires human nuclear professionals in licensed facilities. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously determines decontamination procedures in nuclear facilities; this remains a specialized human technical judgment task supported at most by reference databases. |
Calibrate and maintain chemical instrumentation sensing elements and sampling system equipment, using calibration instruments and hand tools.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.0/5 · click for rater detail
Calibrate and maintain chemical instrumentation sensing elements and sampling system equipment, using calibration instruments and hand tools.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear facilities are conservative, highly regulated sectors with slow technology adoption cycles and mandatory human licensing requirements that actively prevent substitution of critical maintenance tasks. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear monitoring is a highly regulated, low-digitization physical sector with minimal AI agent deployment for hands-on equipment maintenance. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with documenting calibration procedures or analyzing sensor data post-collection, but the core hands-on calibration and maintenance work remains fundamentally manual and human-dependent. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with calibration scheduling, data logging analysis, or anomaly detection in sensor readings, but offers little help with the physical calibration and tool-based maintenance itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires precise physical manipulation of delicate instrumentation and sampling equipment in a nuclear environment with strict safety protocols. Current AI systems cannot perform hands-on calibration, maintenance, or equipment handling in the physical world. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation of sensing elements and sampling equipment with hand tools in a nuclear facility, which current AI systems cannot perform end-to-end.hands-on calibration. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear facilities are heavily regulated (NRC oversight), require licensed or certified nuclear technicians to perform critical safety-related tasks, and have strict liability and accountability requirements that legally mandate human expertise and sign-off. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear facility work involves strict regulatory oversight, safety certification, and licensing requirements mandating qualified human technicians for calibration and maintenance tasks. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Specialized robotic systems capable of nuclear environment operation would be prohibitively expensive compared to trained technician labor, with high integration and safety validation costs. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical calibration work, so any AI cost is additive rather than replacing the human labor cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI products can autonomously calibrate chemical instrumentation or maintain sampling equipment. This requires robotic systems with specialized training for nuclear facilities, which are not in general production use. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously calibrates and maintains physical chemical sensing/sampling equipment in nuclear settings; this remains a manual technician task. |
Place radioactive waste, such as sweepings or broken sample bottles, into containers for shipping or disposal.
0CI 0–0 · exposure 0 · augmentation 25 · importance 3.9/5 · click for rater detail
Place radioactive waste, such as sweepings or broken sample bottles, into containers for shipping or disposal.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nuclear facilities are highly conservative, regulated sectors with slow technology adoption cycles. Any automation must pass rigorous safety validation, making adoption velocity minimal in practice despite some remote-handling pilots. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Nuclear facilities are a highly regulated, safety-critical physical sector with minimal AI/robotic adoption for hands-on hazardous material handling tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Assistive tools (robotic arms, remote cameras, dose-rate guidance) can marginally support technicians, but the task's physical hazard sensitivity and regulatory requirements limit meaningful productivity augmentation compared to a human in direct control. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with logistics, labeling, tracking, and documentation of waste containers, but offers little assistance for the core physical placement and handling task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time physical manipulation in a radioactive environment with strict safety protocols. Current AI systems cannot reliably handle physical waste placement, container preparation, and contamination control—core requirements that demand embodied presence and real-time hazard response. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manipulation task requiring handling of hazardous radioactive materials with specialized shielding, tools, and protective equipment; current AI systems cannot perform this physical handling end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Nuclear waste handling is heavily regulated under NRC, EPA, and DOE authority; human technicians must be licensed and trained; liability for contamination or improper disposal is severe; and regulations explicitly require human oversight and certification—creating hard legal barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Nuclear material handling is heavily regulated (NRC and radiation safety protocols) requiring certified, trained personnel with specific authorization to handle and dispose of radioactive waste, creating hard legal and safety barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying certified autonomous systems (custom robotics, remote handling equipment, validation, compliance infrastructure) vastly exceeds the loaded wage of a trained nuclear technician performing this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Specialized radiation-safe robotics or automation would require far more capital investment than a trained human technician performing this task manually, making AI/robotic substitution costlier today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial product reliably performs autonomous radioactive waste handling at production scale. Robotics exist for some industrial tasks, but nuclear waste management demands certified, validated systems that meet regulatory standards—well beyond current general-purpose AI or robotics deployments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed robotic or AI product autonomously performs radioactive waste sorting and containment in production; radiation-hardened robotics exist for narrow remote-handling tasks but not for general sweeping/bottle disposal workflows. |
Related occupations — Life, Physical & Social Science
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.