Biological Technicians

19-4021.00
Median wage $57,510/yr69,620 employed (US)Rank #186 of 923 scored · top 20% by substitution

Assist biological and medical scientists. Set up, operate, and maintain laboratory instruments and equipment, monitor experiments, collect data and samples, make observations, and calculate and record results. May analyze organic substances, such as blood, food, and drugs.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution40
Exposure38
Augmentation58

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

17 rated tasks, binned by substitution score.

Position among all scored occupations

Distribution of 923 occupation scores; the marker is this occupation.

Tasks with substitution ≥ 70

12%

Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.

Why this score

The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.

Task automatabilityw 35%40

panel mean rating 2.6/5 → substitution pressure 40/100

Technical feasibility todayw 20%36

panel mean rating 2.4/5 → substitution pressure 36/100

Cost vs. human wagew 15%36

panel mean rating 2.5/5 → substitution pressure 36/100

Adoption barriersw 20%inverted — strong barriers lower the score48

panel mean rating 3.1/5 (barrier strength) → substitution pressure 48/100

Sector adoption velocityw 10%35

panel mean rating 2.4/5 → substitution pressure 35/100

Task breakdown (17 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.

Input data into databases.

91

CI 85–97 · exposure 95 · augmentation 63 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Life sciences, pharmaceutical, and healthcare organizations—where biological technicians work—are among the fastest adopters of RPA and data automation. Widespread adoption in R&D and clinical settings is already documented.
Sector adoption velocityclaude-sonnet-53/5Lab and scientific settings are adopting digital data pipelines and LIMS systems but often lag behind information/finance sectors due to legacy instruments, regulatory documentation needs, and smaller IT budgets.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by validating data quality, suggesting corrections, or highlighting anomalies before submission, improving technician productivity, though the core task is primarily automatable rather than augmentable.
Augmentation potentialclaude-sonnet-54/5AI-assisted transcription, OCR, and auto-fill tools significantly speed up data entry and reduce errors while technicians remain responsible for verifying accuracy.
Task automatabilityclaude-haiku-4-5-202510015/5Data entry into databases is a core RPA and AI automation task. Current systems can reliably extract structured or semi-structured data from documents and input it into databases at >50% time savings compared to manual entry, with error rates often lower than humans.
Task automatabilityclaude-sonnet-55/5Data entry into structured databases is a well-defined, repetitive digital task that off-the-shelf AI/automation (OCR, form parsers, RPA, LLM-based extraction) can handle end-to-end with substantial time savings at equal or better quality.
Adoption barriersclaude-haiku-4-5-202510011/5No licensing, regulatory, or human-contact requirement exists for database input; organizations face minimal friction in automating this task beyond standard IT governance.
Adoption barriersclaude-sonnet-52/5No licensing or legal requirement mandates a human perform data entry; the main friction is ensuring data integrity/validation in scientific contexts, which is a moderate but not hard barrier.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated data entry costs pennies per record after initial setup, while manual entry by a technician at loaded cost ($20–30/hour) runs to dollars per record; cost advantage is typically 10–100×.
Cost vs. human wageclaude-sonnet-55/5Automated data entry via software is extremely cheap per record compared to a technician's hourly wage, often costing a small fraction of a cent per entry versus minutes of paid labor.
Technical feasibility todayclaude-haiku-4-5-202510015/5Mature, production-grade automation tools (RPA platforms like UiPath, Automation Anywhere, and AI-powered OCR + database integration) are widely deployed across life sciences and healthcare organizations performing this exact task at scale.
Technical feasibility todayclaude-sonnet-54/5Data entry automation tools, lab information management systems (LIMS) integrations, and OCR/ETL pipelines are widely deployed in production across labs and industries, though scientific data entry sometimes requires validation against unusual formats or handwritten notes.

Keep detailed logs of all work-related activities.

76

CI 65–87 · exposure 78 · augmentation 75 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Research and biotech sectors are high-digitization environments with active adoption of laboratory information management systems (LIMS) and automated logging. The information-intensive nature of these sectors drives rapid pilot and production deployment.
Sector adoption velocityclaude-sonnet-52/5Lab and life sciences settings adopt digital tools unevenly; many biological technician roles remain in academic or field settings with lower digitization and slower AI tool adoption.
Augmentation potentialclaude-haiku-4-5-202510014/5AI significantly assists logging by auto-capturing work context, suggesting structured summaries, and flagging anomalies, substantially reducing manual documentation burden while the technician retains oversight and quality control.
Augmentation potentialclaude-sonnet-54/5AI-assisted transcription, template generation, and data organization tools can meaningfully speed up and standardize log-keeping while the technician remains responsible for accuracy.
Task automatabilityclaude-haiku-4-5-202510015/5Logging work activities is a purely documentary task that AI can fully automate: capturing timestamps, summarizing completed actions, and structuring entries into formatted logs. This easily meets the 50% time-saving threshold and requires minimal human judgment.
Task automatabilityclaude-sonnet-54/5Logging work activities is largely a structured documentation task; AI (voice-to-text, templated entry, LIMS integration) can capture and organize this with substantial time savings, though some manual data entry from instruments may remain.
Adoption barriersclaude-haiku-4-5-202510012/5Minimal barriers exist: no licensing or human sign-off requirement for maintaining logs themselves, though some organizations may prefer human oversight for compliance documentation. Adoption friction is low in digitized labs.
Adoption barriersclaude-sonnet-52/5No licensing requirement for logging itself, but accuracy/compliance needs (e.g., GLP, audit trails) mean some human verification is expected, creating mild friction.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated logging via software agents costs essentially zero per log entry once deployed, versus minutes of technician time per entry. The cost ratio is orders of magnitude in favor of automation.
Cost vs. human wageclaude-sonnet-54/5Automated logging via LIMS or transcription software is inexpensive per entry compared to technician time spent manually writing logs, though integration and oversight add some cost.
Technical feasibility todayclaude-haiku-4-5-202510014/5Automated logging systems, process mining tools, and AI-powered documentation assistants are deployed in production across research and biotech organizations. Performance is highly reliable for structured logging, though edge cases around unusual activity descriptions may require minor human review.
Technical feasibility todayclaude-sonnet-53/5Lab information management systems (LIMS) and dictation/transcription tools are deployed in many labs, but full automated logging tailored to biological technician workflows still requires human review and varies in adoption maturity.

Place orders for laboratory equipment and supplies.

69

CI 60–79 · exposure 70 · augmentation 75 · importance 3.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Biotech, pharmaceutical, and research institutions have rapidly adopted e-procurement and automated ordering systems over the past decade. Adoption is particularly deep in well-resourced sectors (academic research, large biotech firms) and continues to accelerate as supply chain digitization advances.
Sector adoption velocityclaude-sonnet-52/5Life science and biological research labs, especially smaller academic or clinical settings, have historically slow digitization of procurement processes compared to finance or professional services sectors.
Augmentation potentialclaude-haiku-4-5-202510014/5AI tools assist technicians by recommending suppliers, suggesting equivalent products at lower cost, flagging inventory levels, and pre-filling order forms based on historical patterns. These augmentations meaningfully reduce search and comparison time while the technician retains approval authority and domain judgment.
Augmentation potentialclaude-sonnet-54/5AI/automation tools can effectively assist by tracking inventory levels, suggesting reorders, and pre-filling purchase requests, meaningfully speeding up the technician's ordering workflow while they retain final approval.
Task automatabilityclaude-haiku-4-5-202510014/5Placing orders for lab equipment and supplies is highly structured and repetitive, involving vendor selection, price comparison, inventory checks, and order form submission—all amenable to automation. Current AI systems can reliably execute procurement workflows, parse catalogs, and populate order systems, achieving well over 50% time savings with existing tools.
Task automatabilityclaude-sonnet-54/5Ordering supplies is a structured procurement task (checking inventory, comparing catalog items, submitting purchase orders) that AI-driven procurement/inventory systems can largely handle with existing integrations.5This is not full end to end when contextual scientific judgment on brand/spec is needed, but most of the mechanical process qualifies for the 50% time-saving threshold.
Adoption barriersclaude-haiku-4-5-202510013/5Many organizations require human approval for purchase orders above threshold amounts and maintain vendor authorization policies for compliance and budget control. Integration with institutional procurement systems and approval workflows adds friction, though the ordering task itself faces no legal licensing barrier.
Adoption barriersclaude-sonnet-52/5No licensing requirement dictates that only a certified person can place equipment orders; some organizational approval workflows and budget authorization add mild friction but not a hard barrier.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated procurement through integrated ordering systems costs a fraction of human labor once implemented. The marginal cost per order (API calls, database lookups, form submission) is orders of magnitude cheaper than technician time to research vendors, compare quotes, and manually enter orders.
Cost vs. human wageclaude-sonnet-54/5Automated reordering/procurement software costs relatively little compared to technician time spent tracking inventory and placing orders, giving a substantial cost advantage once set up.
Technical feasibility todayclaude-haiku-4-5-202510014/5Procurement automation platforms and e-procurement systems with AI-assisted order placement are deployed in many organizations today. While some domain-specific lab suppliers or custom requisition processes may introduce friction, the core task of order placement is consistently performed by deployed systems in production across research institutions and biotech firms.
Technical feasibility todayclaude-sonnet-53/5Procurement automation and lab inventory management software (e.g., LIMS with reorder triggers, e-procurement platforms) exist and are used in some labs, but many biological technicians still order manually via vendor portals or email without full automation deployed.

Measure or weigh compounds and solutions for use in testing or animal feed.

50

CI 30–70 · exposure 50 · augmentation 50 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Pharmaceutical, biotech, and CRO sectors show strong, measurable adoption of liquid handling and weighing automation; large organizations have deployed these systems for years. Smaller labs lag, but the technology is mainstream in regulated environments.
Sector adoption velocityclaude-sonnet-52/5Physical lab work in biological technician roles sees slower AI/robotic adoption compared to purely digital/information-processing sectors, with automation concentrated in large pharma/biotech firms.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted systems can flag anomalies, suggest protocol adjustments, and integrate results into LIMS, improving technician productivity and reducing manual data entry—useful but incremental compared to full automation.
Augmentation potentialclaude-sonnet-53/5AI-driven inventory systems, digital scales with software logging, and protocol assistants can improve accuracy and efficiency, but the core physical act still requires human execution.
Task automatabilityclaude-haiku-4-5-202510014/5Current robotic systems and automated liquid handlers can reliably measure and weigh compounds and solutions with high precision and reproducibility, easily achieving >50% time savings over manual measurement by a human technician. This is a quantitative, repetitive task with well-defined inputs and outputs suited to automation.
Task automatabilityclaude-sonnet-52/5Precise physical measuring/weighing of compounds requires manual manipulation of lab equipment and samples, which current AI cannot perform without robotic hardware; software alone cannot execute this physical task.
Adoption barriersclaude-haiku-4-5-202510013/5No legal requirement mandates human sign-off on weighing/measuring itself, but quality assurance standards (GMP, ISO) and internal protocols often require human verification of results and method compliance, creating procedural friction.
Adoption barriersclaude-sonnet-53/5Accuracy and safety requirements in testing and feed preparation mean quality control and protocol compliance create friction, though not strict licensing barriers akin to clinical tasks.
Cost vs. human wageclaude-haiku-4-5-202510014/5Capital investment in automated weighing/dispensing systems amortizes quickly in high-throughput labs; running costs (reagents, maintenance) are substantially lower per sample than a technician's loaded wage, especially at scale.
Cost vs. human wageclaude-sonnet-52/5Lab automation equipment for precise measuring is capital-intensive and requires calibration/maintenance, often costing more than a technician's wage unless run at very high volume.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed laboratory automation platforms (automated weighers, liquid dispensers, balance systems integrated with LIMS) are in production use across biotech and pharmaceutical facilities. Error rates are low for standardized protocols, though setup and calibration require oversight.
Technical feasibility todayclaude-sonnet-52/5Automated liquid handlers and lab robots exist for some weighing/dispensing tasks, but they are specialized hardware systems, not general AI, and adoption is limited to well-funded labs with standardized protocols.

Analyze experimental data and interpret results to write reports and summaries of findings.

49

CI 44–55 · exposure 50 · augmentation 75 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Life sciences and biotech sectors are moderately digitized but historically cautious adopters of unvalidated automation in critical analysis; while data analysis tools are common, end-to-end AI report generation remains in pilot phase in most organizations rather than production deployment at scale.
Sector adoption velocityclaude-sonnet-53/5Life sciences and lab settings are adopting AI writing and analysis tools at a moderate pace, with pilots common but full production integration for interpretive reporting still limited.
Augmentation potentialclaude-haiku-4-5-202510014/5AI assistants substantially augment technician productivity by automating data wrangling, generating statistical summaries, and drafting initial report text; a human expert using these tools can iterate and interpret findings much faster than manual analysis, with the human retaining full analytical judgment.
Augmentation potentialclaude-sonnet-54/5AI substantially speeds up drafting, summarizing, and organizing findings, letting technicians focus on judgment-heavy interpretation while AI handles structure and language.
Task automatabilityclaude-haiku-4-5-202510013/5AI can automate significant portions—data analysis, statistical summaries, and initial report drafting—but interpretation of experimental results often requires domain expertise and judgment about what findings mean in context, limiting full end-to-end automation to roughly half the task.
Task automatabilityclaude-sonnet-53/5AI can draft summaries and identify patterns in structured data, but interpreting novel experimental results requires domain judgment and validation that current systems cannot reliably perform end-to-end.','
Adoption barriersclaude-haiku-4-5-202510013/5Regulatory frameworks in biotech and pharmaceuticals often require that a qualified human review, interpret, and sign off on experimental findings; organizational and legal liability for incorrect interpretation creates friction that prevents full substitution without human validation.
Adoption barriersclaude-sonnet-52/5No strict licensing requirement for technicians' report writing, but organizational quality control and scientific accountability create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI systems require significant setup, domain-specific configuration, and human expert oversight to validate interpretations; loaded human technician wage for skilled analysis is often comparable to or lower than the true all-in cost of AI plus required human review and quality assurance.
Cost vs. human wageclaude-sonnet-53/5AI drafting reduces some writing time cheaply, but the need for expert oversight and data validation keeps overall cost roughly comparable to human-only work for accurate results.
Technical feasibility todayclaude-haiku-4-5-202510013/5Multiple products (data analysis platforms, scientific writing assistants, statistical software with AI features) can perform parts of this task in production, but material limitations exist: they often require human validation of interpretations, struggle with novel experimental designs, and may introduce errors in statistical reasoning that require expert review.
Technical feasibility todayclaude-sonnet-53/5LLM-based tools are used in labs to assist with drafting reports and summarizing datasets, but reliability on scientific interpretation is inconsistent and requires expert review.

Use computers, computer-interfaced equipment, robotics or high-technology industrial applications to perform work duties.

48

CI 30–66 · exposure 50 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Pharma, biotech, and clinical diagnostic labs are adopting automation at significant pace; robotic systems and integrated platforms are standard in many facilities. Adoption is faster in larger, well-resourced organizations than in smaller academic or contract research settings.
Sector adoption velocityclaude-sonnet-52/5Life sciences and lab-based industries have moderate digitization but physical lab automation adoption is slower and more capital-intensive than in pure information-processing sectors.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-assisted data analysis, real-time monitoring dashboards, and predictive maintenance tools substantially enhance technician productivity by reducing manual data review and instrument downtime. The technician remains central to quality control and decision-making while AI handles routine monitoring and analysis.
Augmentation potentialclaude-sonnet-53/5AI-driven software and computer-interfaced instruments can assist technicians with data analysis, instrument control, and workflow optimization, improving efficiency while the human remains essential for physical tasks.
Task automatabilityclaude-haiku-4-5-202510014/5Much of the operational component—data logging, equipment control, and routine laboratory automation—can be automated with minimal human intervention, meeting the 50% time-saving threshold. However, some judgment-based oversight and troubleshooting typically require human presence, preventing a rating of 5.
Task automatabilityclaude-sonnet-52/5This task describes operating existing lab equipment and computer interfaces, which is fundamentally a physical/manual and equipment-operation activity requiring hands-on presence; AI can assist software interfaces but cannot replace the physical operation and judgment involved.
Adoption barriersclaude-haiku-4-5-202510013/5Regulatory requirements (GLP, FDA validation of automated methods) and safety protocols in handling biological materials create friction, though no single hard legal requirement mandates human performance. Organizational inertia and the cost of validated system changeover also slow adoption.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement for equipment operation itself, but lab safety protocols, quality control standards, and institutional procurement processes create meaningful friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510013/5Automated laboratory equipment is capital-intensive and requires IT maintenance, offsetting labor savings. For routine, high-volume tasks the ratio improves, but for diverse, lower-volume work the comparison with a technician's loaded wage is roughly equivalent.
Cost vs. human wageclaude-sonnet-52/5Deploying robotics/automated lab systems requires significant capital investment, integration, and maintenance costs that often exceed the cost of a technician for moderate-volume work, though it can be cheaper at very high scale.
Technical feasibility todayclaude-haiku-4-5-202510014/5Laboratory automation platforms, robotic liquid handlers, and computer control systems are deployed routinely in pharmaceutical and biotech production environments. Error rates on routine tasks are low, though complex instrument integration still involves significant manual configuration and oversight.
Technical feasibility todayclaude-sonnet-52/5Lab automation and robotics exist in some high-throughput settings, but most biological technician work still requires human operation, calibration, and troubleshooting of equipment in real-world lab conditions.

Monitor and observe experiments, recording production and test data for evaluation by research personnel.

38

CI 30–46 · exposure 38 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Automation of monitoring in research labs is progressing moderately, with high-throughput facilities and well-funded sectors (pharma, biotech) adopting automated data logging, while smaller academic labs lag. Overall adoption is in the pilot-to-early-production phase rather than widespread replacement.
Sector adoption velocityclaude-sonnet-52/5Life science labs are adopting digital lab tools and automation gradually, but adoption of AI-driven autonomous experiment monitoring remains limited to well-funded pharma/biotech settings, not broad or fast.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-powered monitoring assistants (anomaly detection, real-time alerts, automated data logging) substantially augment technician productivity by handling continuous observation and flagging deviations, freeing the technician to focus on troubleshooting and qualitative assessment while remaining in the loop.
Augmentation potentialclaude-sonnet-53/5AI-assisted data logging, anomaly flagging, and automated data capture tools help technicians track experiments more efficiently, though human observation remains central.
Task automatabilityclaude-haiku-4-5-202510013/5Monitoring and recording structured experimental data can be partially automated through sensors, data logging systems, and automated image/signal capture, achieving meaningful time savings on routine observation. However, identifying anomalies, contextual judgment about when to intervene, and qualitative assessment of experimental conditions typically require human oversight, preventing full end-to-end automation at 50%+ time savings.
Task automatabilityclaude-sonnet-52/5Sensor logging and data capture can be automated, but the observational judgment and recognition of anomalies during live experiments still require human presence and expertise for most lab settings today.
Adoption barriersclaude-haiku-4-5-202510013/5Laboratory work often requires human sign-off and approval of data quality and interpretation by research supervisors; institutional protocols may mandate human observation and documentation for compliance and reproducibility. Some regulatory environments (e.g., GLP, GMP) require human verification, though these requirements vary by field and jurisdiction.
Adoption barriersclaude-sonnet-53/5No licensing requirement per se, but liability for data integrity, chain-of-custody, and quality control in regulated research settings creates moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Automated monitoring systems require significant upfront capital investment in sensors and software integration specific to each experimental setup, plus ongoing maintenance and calibration. For many academic or smaller lab settings, the loaded cost of deploying automated systems may still exceed the cost of a technician's hourly wage per task.
Cost vs. human wageclaude-sonnet-52/5Automated sensors/dataloggers can be cheap for narrow repetitive measurements, but broader experiment monitoring still needs technician oversight, keeping blended cost comparable to or above human labor in many labs.
Technical feasibility todayclaude-haiku-4-5-202510013/5Deployed data acquisition and monitoring systems (automated lab instruments, LIMS software, sensor networks) reliably perform data collection and logging in production research settings. However, integration varies by experimental type, and comprehensive automated interpretation of observations across diverse experiments remains inconsistent, limiting full production reliability.
Technical feasibility todayclaude-sonnet-52/5Lab automation and IoT sensor systems exist for specific standardized assays, but general-purpose 'monitor and observe experiments' across biological contexts is not reliably handled by deployed products.

Set up, adjust, calibrate, clean, maintain, and troubleshoot laboratory and field equipment.

31

CI 10–52 · exposure 33 · augmentation 50 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of robotic automation in laboratory settings is growing in high-throughput biotech and pharma, but uptake remains limited in smaller labs and field work. Most biological technician roles still rely on human labor; adoption lags behind information and finance sectors.
Sector adoption velocityclaude-sonnet-51/5Physical equipment maintenance in lab/field settings is a low-digitization, hands-on task with minimal AI/robotics adoption in this sector to date.
Augmentation potentialclaude-haiku-4-5-202510014/5AI and automation can significantly assist technicians by automating routine setup and maintenance steps, flagging calibration drift, predicting equipment failures, and guiding troubleshooting logic, allowing humans to focus on complex diagnostic judgment and field adaptation.
Augmentation potentialclaude-sonnet-52/5AI can assist with diagnostic troubleshooting guides, predictive maintenance alerts, or documentation, but offers limited direct help with the physical calibration and hands-on maintenance work itself.
Task automatabilityclaude-haiku-4-5-202510014/5Most setup, calibration, and cleaning steps for laboratory equipment are highly procedural and repetitive—amenable to robotic automation or AI-guided workflows. However, troubleshooting complex field equipment often requires adaptive diagnosis in varied conditions, so end-to-end automation at equal quality may not exceed 50% time savings in all scenarios, particularly for novel failures.
Task automatabilityclaude-sonnet-51/5This requires physical manipulation of lab and field equipment (calibration, cleaning, hands-on troubleshooting) which current AI systems cannot perform without robotics, and no off-the-shelf system exists to do this end-to-end.
Adoption barriersclaude-haiku-4-5-202510013/5Calibration of precision instruments and maintenance of biological equipment often require certification or regulatory sign-off in healthcare/pharma contexts. Equipment varies widely across labs and field sites, necessitating customization; human technician oversight and sign-off for QA are typically mandated, creating meaningful friction.
Adoption barriersclaude-sonnet-53/5While not formally licensed work, equipment calibration often has quality/compliance implications (e.g., lab accreditation standards) and requires trained personnel physically present, creating moderate organizational and safety barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5Robotic laboratory automation and specialized field diagnostic tools carry high capital and integration costs. While per-task inference is cheap, the overhead of setup, training, and continuous oversight makes the all-in cost comparable to or exceeding that of a technician for most field and troubleshooting work.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing the physical labor involved, so the human remains the only cost-effective option; robotic solutions would be far more expensive than technician wages.
Technical feasibility todayclaude-haiku-4-5-202510013/5Laboratory automation platforms and robotic arms can handle routine setup and maintenance tasks in controlled environments, and some deployed systems manage predictive maintenance. However, field troubleshooting remains inconsistent and error-prone; production systems exist for narrow use cases but not reliably across diverse equipment types.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously sets up, calibrates, or physically maintains biological lab/field equipment; this remains a manual, hands-on task performed by technicians.

Examine animals and specimens to detect the presence of disease or other problems.

31

CI 25–37 · exposure 30 · augmentation 63 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Hospitals, research labs, and diagnostic centers are piloting AI-assisted pathology and specimen analysis, but production deployment remains piecemeal; adoption accelerates in high-volume imaging but lags in small labs and specialized animal health settings.
Sector adoption velocityclaude-sonnet-52/5Biological/life sciences lab settings are adopting AI tools slowly compared to digital-native sectors, with pilots for image-based diagnostics but limited widespread production deployment for this task.
Augmentation potentialclaude-haiku-4-5-202510014/5AI excels at augmenting technicians by rapidly screening images, flagging abnormalities, and highlighting regions of interest, allowing humans to focus on confirmation and complex cases; this assistant role is already demonstrable and materially improves technician throughput.
Augmentation potentialclaude-sonnet-53/5AI-assisted image analysis and pattern recognition tools can help technicians flag abnormalities or prioritize specimens for closer review, improving throughput while humans retain final judgment.
Task automatabilityclaude-haiku-4-5-202510012/5Current AI can assist with image-based disease detection (pathology slides, imaging) but typically requires human confirmation and context; end-to-end autonomous disease detection in live animals or complex specimens falls short of the 50% time-saving bar due to edge cases, specimen variation, and need for human expertise.
Task automatabilityclaude-sonnet-52/5Visual/physical examination of live animals or biological specimens for disease signs requires hands-on manipulation, sensory judgment, and often lab testing that current AI cannot fully replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Clinical and regulatory requirements often mandate licensed personnel (pathologists, veterinarians, or certified technicians) to validate findings and sign off on disease diagnoses; quality and liability standards create friction against full automation even where technical capability exists.
Adoption barriersclaude-sonnet-53/5No strict licensing mandate for this specific task, but liability for misdiagnosis, need for physical specimen handling, and quality control practices create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI image analysis tools have moderate licensing costs and require integration and human oversight; when factoring in the technician's loaded wage plus the still-necessary human review time, cost savings are limited and variable across specimen types.
Cost vs. human wageclaude-sonnet-52/5Specialized imaging AI can be cheap per-analysis, but integration, specimen handling, and physical examination still require paid technician labor, keeping overall cost comparable to human-only workflows.
Technical feasibility todayclaude-haiku-4-5-202510013/5AI products exist for histopathology and radiology image analysis (e.g., pathology AI platforms), but they operate as decision-support tools with material error rates; deployment in production labs is growing but usually with mandatory human review rather than autonomous operation.
Technical feasibility todayclaude-sonnet-52/5AI image analysis tools exist for specific pathology or microscopy tasks, but broad deployed products reliably performing general animal/specimen disease detection in production are limited and narrow in scope.

Isolate, identify and prepare specimens for examination.

28

CI 25–30 · exposure 25 · augmentation 63 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of full specimen automation remains slow and piecemeal. Most labs use manual or semi-automated workflows; high-touch standardization required for each specimen type and lab context limits rapid rollout. Pilot systems exist but production-scale autonomous specimen isolation is rare.
Sector adoption velocityclaude-sonnet-52/5Life science and lab-based sectors adopt AI more slowly than digital-native industries, with automation concentrated in analysis/imaging rather than physical specimen handling.
Augmentation potentialclaude-haiku-4-5-202510013/5AI augmentation is modest but growing: image-based identification assists technicians in species/pathogen recognition, and digital logging systems reduce documentation time. However, the physical isolation and preparation steps remain largely manual, limiting overall productivity gains.
Augmentation potentialclaude-sonnet-54/5AI-powered image analysis and identification tools meaningfully speed up species/sample identification and documentation, while the technician still performs physical isolation and preparation.
Task automatabilityclaude-haiku-4-5-202510012/5Specimen isolation and preparation involve substantial physical handling, dexterity, and real-time decision-making in wet-lab environments. While AI can assist with identification (image analysis), the isolation and physical preparation steps require robotic manipulation that current general-purpose systems cannot reliably perform end-to-end at the required speed and quality.
Task automatabilityclaude-sonnet-52/5This requires physical manipulation of biological specimens (dissection, mounting, staining, culturing) which current AI cannot perform end-to-end; only the identification/classification sub-step is partially automatable via image recognition tools.'
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: lab biosafety regulations (BSL levels) require human oversight and sign-off; specimen handling involves biohazard liability; quality assurance in clinical/research labs mandates human certification and accountability for specimen integrity and chain of custody.
Adoption barriersclaude-sonnet-53/5Lab protocols, chain-of-custody, and quality control standards create moderate friction, though no formal licensing mandates a human perform this specific task.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current lab automation systems are capital-intensive and require significant integration costs, making them more expensive than a trained technician for most specimen preparation workflows. Even simple image-based identification systems require oversight and validation by human experts.
Cost vs. human wageclaude-sonnet-52/5The physical isolation and preparation steps still require human labor, lab equipment, and technician time, so AI only reduces cost for the identification portion, not the whole task.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs the full sequence of isolating, identifying, and preparing biological specimens autonomously. Lab robotics exist for narrow, highly structured tasks (e.g., liquid handling), but integration across the heterogeneous specimen types and preparation methods found in real labs remains research-stage or prototypical.
Technical feasibility todayclaude-sonnet-52/5AI-based species/microorganism identification tools exist and are used as aids, but no deployed product isolates and physically prepares specimens for examination in production lab settings.

Feed livestock or laboratory animals.

27

CI 16–37 · exposure 22 · augmentation 38 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Large-scale commercial farming has adopted automated feeders, but adoption is slower and more patchy in smaller research labs and specialized animal care settings. Regulatory caution and the capital barrier slow velocity outside industrial agriculture.
Sector adoption velocityclaude-sonnet-52/5Agriculture and animal research are traditionally slow-adopting, physical, low-digitization sectors; automated feeding systems have existed for decades but adoption is uneven and largely mechanical rather than AI-driven.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-driven monitoring systems can assist technicians by tracking feed consumption patterns, flagging animals with unusual intake, and predicting health issues, raising monitoring efficiency. However, the physical act of feeding and hands-on animal assessment remain human-dependent tasks where AI plays a supporting rather than transformative role.
Augmentation potentialclaude-sonnet-52/5AI can help optimize feeding schedules, monitor consumption patterns, and flag anomalies via sensor data, but does not materially change the physical act of feeding itself, which remains a hands-on task.
Task automatabilityclaude-haiku-4-5-202510012/5While automated feeding systems exist for livestock, they require significant setup and handle only routine feeding schedules. Most systems cannot adapt to individual animal health needs, behavioral changes, or emergency conditions that a technician must monitor, limiting time savings to perhaps 20–30% in practice.
Task automatabilityclaude-sonnet-51/5Physically feeding livestock or laboratory animals requires manual handling, mobility, and dexterity that current AI systems (software-based) cannot perform; this is a robotics/physical task, not amenable to LLM-style automation.
Adoption barriersclaude-haiku-4-5-202510014/5Animal welfare regulations, food safety compliance (especially for laboratory animals), and liability requirements often mandate human inspection and sign-off on feeding operations. Insurance and regulatory bodies frequently require documented human oversight of feeding adequacy and animal condition.
Adoption barriersclaude-sonnet-53/5Animal welfare regulations, IACUC oversight for lab animals, and biosecurity protocols create moderate barriers, though feeding itself isn't a licensed-professional-only task; the barrier stems more from need for careful observation of animal health during feeding rather than legal restriction.
Cost vs. human wageclaude-haiku-4-5-202510012/5Automated feeding systems require substantial upfront capital investment, ongoing maintenance, software licensing, and integration costs. For many smaller biological tech settings, the all-in cost per feeding cycle exceeds the loaded wage of a technician, especially when oversight and troubleshooting are factored in.
Cost vs. human wageclaude-sonnet-52/5Automated feeders can be cost-effective at scale for livestock but require significant capital investment; for lab animal care requiring individualized attention and health monitoring, human labor remains necessary and cost-competitive when factoring in oversight and animal welfare needs.
Technical feasibility todayclaude-haiku-4-5-202510013/5Automated feeders are deployed in commercial farming and some lab facilities, but error rates in portion control, ingredient mixing, and handling sick/stressed animals remain material. Most systems are narrowly scoped (grain-fed cattle) and lack robustness for diverse animal welfare scenarios.
Technical feasibility todayclaude-sonnet-52/5Automated feeding systems (timed dispensers, precision livestock feeders) exist and are deployed in some agricultural and lab settings, but these are specialized hardware solutions, not general AI, and many facilities still rely on manual feeding, especially for lab animals requiring monitoring.

Clean, maintain and prepare supplies and work areas.

25

CI 15–35 · exposure 13 · augmentation 25 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of robotic cleaning in biological labs is limited and mostly in large research institutions or specialized facilities; most smaller labs and clinics rely on manual preparation and maintenance due to cost and task variability.
Sector adoption velocityclaude-sonnet-51/5Physical lab maintenance tasks in biological/scientific settings show minimal AI or robotic adoption; this is a low-digitization, physically embodied task category.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-powered scheduling and inventory management systems can assist technicians in prioritizing which areas and supplies need attention, and basic alerting systems exist; however, the core physical work remains minimally augmented by current AI.
Augmentation potentialclaude-sonnet-51/5AI offers essentially no meaningful assistance for physical cleaning, organizing, and supply preparation tasks in a lab environment.
Task automatabilityclaude-haiku-4-5-202510012/5Physical cleaning and maintenance of lab work areas requires dexterous manipulation, spatial reasoning, and context-aware judgment about which supplies need attention. Current robotic systems can handle narrow, repetitive cleaning tasks but cannot reliably manage the varied, semi-structured environments of a typical lab with satisfactory time savings over human work.
Task automatabilityclaude-sonnet-51/5This is a physical task requiring hands-on cleaning of lab equipment and organizing physical workspaces, which current AI systems cannot perform without robotic embodiment far beyond off-the-shelf availability.
Adoption barriersclaude-haiku-4-5-202510012/5Lab environments often have biosafety and regulatory requirements (e.g., sterility protocols, chemical handling) that create some friction toward automation, though these are primarily oversight requirements rather than hard legal prohibitions on machine performance.
Adoption barriersclaude-sonnet-52/5No licensing requirement specifically for cleaning/prep, but lab safety protocols and contamination control create some procedural friction, though these are not legal barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Cleaning robots and specialized maintenance systems remain relatively expensive to acquire and integrate compared to the modest wages of technician-level cleaning labor, especially when factoring in overhead and reliability gaps.
Cost vs. human wageclaude-sonnet-51/5AI has no viable mechanism to perform this physical task, so any hypothetical robotic solution would be far more costly than a human technician's labor for this routine work.
Technical feasibility todayclaude-haiku-4-5-202510012/5While some specialized robotic cleaning systems exist in controlled environments, deployed products do not reliably perform general lab cleaning, maintenance, and supply preparation at production scale. Most real-world lab maintenance remains human-driven due to variability and the need for judgment.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical lab cleaning, sterilization, or supply preparation in production settings; this remains a manual task performed by human technicians.

Conduct standardized biological, microbiological or biochemical tests and laboratory analyses to evaluate the quantity or quality of physical or chemical substances in food or other products.

25

CI 25–25 · exposure 25 · augmentation 50 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While routine assays in high-volume labs (clinical, food safety) see partial automation, most smaller research and QA labs still rely heavily on manual technician work; sector adoption remains moderate and slow due to capital and regulatory friction.
Sector adoption velocityclaude-sonnet-52/5Lab and food-testing sectors adopt automation slowly due to capital costs, validation requirements, and small-lab prevalence, with AI-driven data analysis pilots more common than full task automation.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist in data interpretation, flagging anomalies, and generating preliminary reports, but the technician remains the primary executor of physical testing and final validation, offering useful but not transformative productivity gain.
Augmentation potentialclaude-sonnet-53/5AI can assist with data logging, anomaly detection, and interpreting results faster, improving technician throughput, though it doesn't replace the physical testing steps.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in data analysis and result interpretation, the core task involves hands-on laboratory procedures (sample preparation, instrument operation, microscopy) that require physical manipulation and real-time decision-making in a wet lab environment—capabilities current AI systems lack end-to-end.
Task automatabilityclaude-sonnet-52/5Physical sample handling, running assays, and operating lab instruments require hands-on manipulation that current AI cannot perform end-to-end; AI can assist with data interpretation but not the core wet-lab work.'
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory compliance (FDA, ISO standards, GxP) typically requires a licensed or certified technician to perform or validate the test; data integrity and traceability rules mandate human sign-off on critical testing steps, creating hard procedural barriers.
Adoption barriersclaude-sonnet-54/5Many biological/chemical tests are subject to regulatory quality standards (e.g., FDA, ISO, GLP) requiring certified personnel and documented chain-of-custody, creating strong compliance-driven barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Specialized lab automation is capital-intensive and typically requires parallel human technician oversight for quality assurance and troubleshooting, making the all-in cost comparable to or higher than skilled technician labor.
Cost vs. human wageclaude-sonnet-52/5Lab automation equipment and robotics require substantial capital investment, calibration, and maintenance, often costing more than technician wages for lower-volume or non-standardized testing.
Technical feasibility todayclaude-haiku-4-5-202510012/5Laboratory robots and automated analyzers exist for specific assays, but they require human setup, calibration, and oversight; no general AI system today reliably performs the full scope of standardized testing independently across diverse biological matrices.
Technical feasibility todayclaude-sonnet-52/5Lab automation robots and analyzers exist for narrow, standardized assays in some high-throughput settings, but most biological technician work still requires manual sample prep and judgment not covered by deployed AI products.

Provide technical support and services for scientists and engineers working in fields such as agriculture, environmental science, resource management, biology, and health sciences.

25

CI 20–30 · exposure 20 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Biology and life-science labs are digitizing slowly compared to finance or software sectors; they retain strong preference for hands-on technical expertise. Pilot AI deployments exist but production adoption of AI for core technical support remains limited, with most organizations still relying on human technicians.
Sector adoption velocityclaude-sonnet-52/5Life sciences and environmental research labs adopt AI tools for data analysis but physical technician work remains largely unautomated, placing this in slower-adopting scientific/lab environments.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist technicians through rapid access to protocol documentation, literature summaries, equipment manuals, and troubleshooting decision trees, raising individual productivity modestly. However, the task's requirement for adaptive judgment and hands-on lab work limits the magnitude of augmentation gains compared to more routine knowledge work.
Augmentation potentialclaude-sonnet-54/5AI tools substantially assist with data analysis, literature review, protocol documentation, and troubleshooting guidance, meaningfully raising technician productivity even though hands-on tasks remain human-performed.
Task automatabilityclaude-haiku-4-5-202510012/5Technical support work involves highly variable, context-dependent problem-solving, troubleshooting, and real-time collaboration with scientists. While AI can assist with documentation, scheduling, and routine inquiries, end-to-end automation of the full technical support role—including hands-on lab work, equipment diagnostics, and adaptive guidance—falls well short of the 50% time-saving threshold with current systems.
Task automatabilityclaude-sonnet-52/5This task encompasses broad, physically situated laboratory and field support (sample handling, equipment maintenance, protocol execution) that current AI cannot perform end-to-end, though data analysis and documentation sub-components are automatable.rating.rating
Adoption barriersclaude-haiku-4-5-202510014/5Scientific and health-science labs operate under regulatory oversight (GLP, HIPAA, safety standards) and organizational culture strongly favors human expertise and accountability. Institutions typically require a qualified human technician to sign off on technical work and troubleshooting, creating meaningful legal and procedural friction against full automation.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement, but lab safety protocols, chain-of-custody, and physical dexterity requirements create real organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI inference and integration costs remain modest, but oversight of AI-generated technical advice in high-stakes lab environments is expensive and ongoing. The all-in cost per resolved technical support interaction likely approaches or exceeds the loaded wage of a technician, particularly when factoring in liability and error correction.
Cost vs. human wageclaude-sonnet-52/5Physical lab/field work still requires human labor and equipment interaction, so AI cannot substitute cheaply for the bulk of this task; only narrow digital sub-tasks see cost advantage.
Technical feasibility todayclaude-haiku-4-5-202510012/5Deployed AI products can handle narrow facets (FAQ answering, ticketing) but lack the domain expertise, real-time Lab equipment integration, and problem-solving depth needed for reliable technical support across diverse scientific fields. No production system demonstrably performs this role end-to-end at acceptable error rates.
Technical feasibility todayclaude-sonnet-51/5No deployed product provides general 'technical support' spanning hands-on lab work, equipment operation, and interdisciplinary scientific assistance; this remains far from a productized capability.

Participate in the research, development, or manufacturing of medicinal and pharmaceutical preparations.

24

CI 20–28 · exposure 25 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI in pharma R&D is emerging (computational chemistry, molecule screening) but real manufacturing and technician roles remain heavily human-driven due to regulatory lock-in and risk aversion. Most pharmaceutical companies run pilots in silico; production-line displacement is minimal and slow.
Sector adoption velocityclaude-sonnet-53/5placeholder
Augmentation potentialclaude-haiku-4-5-202510013/5AI assists pharmaceutical technicians meaningfully in data analysis, literature review, compound database searching, and experimental design optimization. However, hands-on synthesis, real-time quality checks, and safety validation remain human-led, so augmentation is partial rather than transformative.
Augmentation potentialclaude-sonnet-54/5placeholder
Task automatabilityclaude-haiku-4-5-202510012/5While some discrete sub-tasks like data logging, sample tracking, and basic compound analysis can be automated, the task inherently requires human oversight of complex chemical synthesis, quality control decisions, and safety protocols. Current AI cannot reliably conduct end-to-end pharmaceutical R&D or manufacturing with the safety and regulatory compliance standards required, so the ≥50% time-saving threshold is not met.
Task automatabilityclaude-sonnet-52/5This task involves hands-on lab work, equipment operation, and physical manufacturing/QC steps that current AI cannot perform end-to-end; AI can assist with data analysis or documentation but not the physical execution."},"feasibility":{"rating":2,"rationale":"No deployed AI product autonomously runs pharmaceutical R&D or manufacturing lab work; AI is used for auxiliary data analysis and literature review but not as a substitute technician."},"cost_ratio":{"rating":2,"rationale":"Physical lab and manufacturing tasks still require human labor, equipment, and compliance oversight, so AI cost savings are limited to peripheral analytical tasks rather than the whole task."},"barriers":{"rating":4,"rationale":"Pharmaceutical R&D and manufacturing are heavily regulated (GMP, FDA), requiring qualified human personnel, documentation, and accountability, creating strong barriers to full automation."},"adoption_velocity":{"rating":3,"rationale":"Biotech and pharma sectors are adopting AI for data analysis, drug discovery modeling, and lab automation robotics, but adoption for core hands-on manufacturing tasks remains moderate and uneven."},"augmentation":{"rating":4,"rationale":"AI tools significantly aid in data analysis, experimental design, literature synthesis, and quality documentation, meaningfully boosting technician productivity while humans still perform physical tasks.
Adoption barriersclaude-haiku-4-5-202510015/5Pharmaceutical manufacturing and drug development are among the most heavily regulated domains (FDA, EMA, cGMP, IND/NDA requirements). A licensed pharmacist or chemist must sign off on formulations, testing protocols, and manufacturing batches; liability for contamination, efficacy failure, or safety defects is severe and cannot be delegated to AI agents.
Adoption barriersclaude-sonnet-54/5placeholder
Cost vs. human wageclaude-haiku-4-5-202510012/5Pharmaceutical technician labor is relatively inexpensive per hour (USD 30–40k annually loaded), while integrating AI systems for compliance-heavy work, validation, and oversight adds significant engineering and legal costs. The cost-benefit favors human technicians for most pharmaceutical tasks.
Cost vs. human wageclaude-sonnet-52/5placeholder
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product can autonomously conduct pharmaceutical R&D or manufacturing end-to-end. AI tools exist for molecular modeling and drug screening (research-stage), but production pharmaceutical work requires human chemists and technicians to make real-time decisions, validate results, and ensure GMP compliance—only narrow, routine subtasks have reliable automation in production.
Technical feasibility todayclaude-sonnet-52/5placeholder

Monitor laboratory work to ensure compliance with set standards.

23

CI 20–25 · exposure 20 · augmentation 50 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Laboratories are moderately digitized but conservative in compliance oversight; adoption of autonomous monitoring is slow and limited to specialized niches (pharma, large biotech). Most labs still rely heavily on manual checklists and human technician oversight.
Sector adoption velocityclaude-sonnet-52/5Lab science and biotech sectors are adopting digital LIMS and IoT sensors gradually, but physical compliance monitoring by technicians is still largely manual with slow AI penetration compared to information-sector tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by flagging anomalies, maintaining audit trails, and summarizing deviations for human review, raising efficiency in data handling and alerting. However, the human technician remains the decision-maker and validator, so the augmentation is meaningful but bounded.
Augmentation potentialclaude-sonnet-53/5AI-enabled sensors, anomaly detection, and automated logging can meaningfully assist technicians in tracking parameters and flagging deviations, improving efficiency while the technician remains responsible for judgment and sign-off.
Task automatabilityclaude-haiku-4-5-202510012/5Monitoring for compliance requires judgment about context, deviations, and risk assessment. AI can flag anomalies in logs or images but cannot reliably assess the full compliance posture without human judgment, and current systems lack the contextual understanding to independently validate adherence to laboratory standards at equal quality.
Task automatabilityclaude-sonnet-52/5This task requires physical presence, judgment about real-time lab conditions, and hands-on verification of procedures that current AI cannot perform end-to-end; AI can support documentation but not the core monitoring function.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory bodies (FDA, EPA, OSHA) typically require documented human oversight of laboratory compliance, and liability for missed deviations falls on the organization and often a responsible person. Human sign-off on compliance is often mandated, creating a hard barrier to full automation.
Adoption barriersclaude-sonnet-54/5Compliance monitoring in biological labs is often tied to regulatory frameworks (e.g., GLP, biosafety) requiring qualified personnel accountability, creating strong institutional and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI monitoring systems (sensors, cameras, alerts) require significant infrastructure, integration, and human oversight to validate findings. This can approach but rarely undershoots the cost of a technician's partial time on monitoring, especially at typical lab scales.
Cost vs. human wageclaude-sonnet-52/5Sensors and LIMS systems can reduce some manual logging costs, but the physical inspection and judgment-based compliance monitoring still requires a trained technician, so full AI substitution costs remain high relative to labor savings.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some monitoring tools (video analysis, automated alerts on parameter drift) exist in labs, but they are narrow and typically require human review. No deployed end-to-end system reliably monitors compliance with all relevant standards without material gaps or false positives.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously monitors physical laboratory compliance in real time; this remains a human oversight function with only ancillary digital tools (LIMS alerts, checklists) in production.

Conduct research, or assist in the conduct of research, including the collection of information and samples, such as blood, water, soil, plants and animals.

13

CI 5–21 · exposure 8 · augmentation 50 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Biological research and field sampling remain labor-intensive, low-digitization sectors with strong preferences for trained human technicians. AI adoption in this domain is minimal; automation focuses on analysis, not collection.
Sector adoption velocityclaude-sonnet-52/5Life sciences and field research sectors have low digitization of physical fieldwork; AI adoption here lags far behind information-based occupations, with automation limited to some lab instrumentation, not field collection.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist with data logging, sample tracking, environmental parameter recording, and post-collection analysis recommendations, moderately improving technician workflow but not transforming the core manual task of collection itself.
Augmentation potentialclaude-sonnet-53/5AI can assist with research planning, data logging, protocol generation, and sample tracking/analysis, but does not materially change the physical collection process itself.
Task automatabilityclaude-haiku-4-5-202510012/5Physical sample collection (blood, water, soil, plants, animals) requires hands-on laboratory or field manipulation that current robotics cannot reliably perform end-to-end. Information collection and record-keeping could be partially automated, but the primary bottleneck—actual specimen handling and environmental sampling—remains a human task.
Task automatabilityclaude-sonnet-51/5This task requires physical fieldwork and lab collection of samples (blood, water, soil, plants, animals) that current AI systems cannot physically perform; it is fundamentally manual and embodied work.
Adoption barriersclaude-haiku-4-5-202510014/5Chain-of-custody requirements, regulatory compliance for specimen handling (especially blood and biological materials), safety protocols, and quality assurance in research contexts create strong legal and procedural barriers to full automation. Human sign-off on sample integrity is typically mandatory.
Adoption barriersclaude-sonnet-53/5While not always requiring licensure, sample collection often demands chain-of-custody protocols, safety training, fieldwork conditions, and physical presence, creating substantial practical barriers to remote or AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current automation for sample collection is expensive (specialized robotics, infrastructure, maintenance) and still requires significant human oversight, making the all-in cost exceed that of a biological technician performing the work directly.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical collection labor, so no cost comparison favors AI; any automation (e.g., robotic samplers) would require expensive specialized hardware exceeding human labor costs for this task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed commercial systems reliably conduct independent sample collection across the diverse matrices mentioned (biological, aquatic, terrestrial, botanical, zoological). Research robots exist but are narrowly specialized and require extensive human supervision; this is not a production-ready automation.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product collects biological or environmental samples in the field or lab; this remains entirely a human physical activity, sometimes supported by lab robotics for narrow subtasks but not general sample collection.

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.