Licensed Practical and Licensed Vocational Nurses

29-2061.00

Median wage $64,400/yr648,410 employed (US)Rank #745 of 923 scored · top 81% by substitution

Care for ill, injured, or convalescing patients or persons with disabilities in hospitals, nursing homes, clinics, private homes, group homes, and similar institutions. May work under the supervision of a registered nurse. Licensing required.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution17

Exposure16

Augmentation41

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

22 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

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%17

panel mean rating 1.7/5 → substitution pressure 17/100

Technical feasibility todayw 20%14

panel mean rating 1.5/5 → substitution pressure 14/100

Cost vs. human wagew 15%16

panel mean rating 1.6/5 → substitution pressure 16/100

Adoption barriersw 20%inverted — strong barriers lower the score21

panel mean rating 4.2/5 (barrier strength) → substitution pressure 21/100

Sector adoption velocityw 10%16

panel mean rating 1.6/5 → substitution pressure 16/100

Task breakdown (22 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.

Inventory and requisition supplies and instruments.

CI 55–79 · exposure 62 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	4/5	Healthcare is a well-digitized, information-intensive sector with strong incentives to reduce non-clinical labor; inventory automation and AI-driven demand forecasting are already widely deployed in hospital systems and pharmacy operations. Adoption is rapid in large health systems, though slower in smaller or rural clinics.
Sector adoption velocity	claude-sonnet-5	3/5	Healthcare facilities are adopting inventory management software and RFID/barcode systems at a moderate pace, with pilots and partial deployments more common than fully mature adoption in smaller LPN/LVN work settings.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI-powered inventory dashboards, automated alerts for low stock, and predictive ordering significantly enhance nurse productivity by reducing manual counting and reorder time, while nurses remain the final check on clinical need and emergency supplies. This is a strong case of augmentation rather than replacement.
Augmentation potential	claude-sonnet-5	4/5	AI-enabled inventory software can significantly streamline tracking, alert on low stock, and auto-generate requisition orders, meaningfully boosting efficiency while a human still oversees ordering and physical restocking.
Task automatability	claude-haiku-4-5-20251001	4/5	Inventory tracking and supply requisitioning are largely transactional and rule-based tasks that can be largely automated through barcode scanning, RFID integration, and inventory management systems with AI-driven demand forecasting. However, some nuanced decisions about non-standard items or emergency protocols may require human oversight, preventing a full 5.
Task automatability	claude-sonnet-5	3/5	Inventory tracking and requisition generation are largely rule-based data tasks that AI/automation software can handle, but physical counting and system integration still require setup and some human verification.
Adoption barriers	claude-haiku-4-5-20251001	2/5	While there is no legal requirement that a licensed nurse personally perform inventory and requisitioning, healthcare organizations typically use inventory staff or delegate to nurses. Integration with existing clinical workflows and the preference to have clinically trained staff verify critical items add some organizational friction, but automation faces no hard legal or licensing barriers.
Adoption barriers	claude-sonnet-5	2/5	No licensure requirement restricts who manages supply inventory, though facility-specific protocols and accountability for medical supplies/controlled items introduce moderate organizational friction.
Cost vs. human wage	claude-haiku-4-5-20251001	5/5	Automated inventory systems and barcode scanning are far cheaper than paying licensed nursing staff to manually count, track, and order supplies. The per-task cost of AI-driven inventory is orders of magnitude lower than human labor, especially when considering nurse wages and opportunity cost.
Cost vs. human wage	claude-sonnet-5	3/5	Automated inventory systems reduce labor time but require software licensing, integration, and maintenance costs that make the savings moderate rather than order-of-magnitude cheaper, especially in smaller facilities.
Technical feasibility today	claude-haiku-4-5-20251001	4/5	Hospital and clinical inventory management systems with barcode/RFID automation and automated reordering are deployed at scale in most major health systems. AI-enhanced demand forecasting and supply chain optimization tools exist in production, though integration with legacy systems and human exception-handling still involve material manual steps.
Technical feasibility today	claude-sonnet-5	3/5	Inventory management systems with automated reordering exist and are used in many healthcare facilities, but full end-to-end automation without human check-ins is not universal, especially for smaller clinics or LPN-staffed settings.

Make appointments, keep records, or perform other clerical duties in doctors' offices or clinics.

CI 54–75 · exposure 62 · augmentation 75 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	4/5	Healthcare organizations, particularly larger clinics and hospital systems, are actively adopting scheduling and EHR automation; vendor products are widely deployed and momentum is strong, though adoption lags in small practices.
Sector adoption velocity	claude-sonnet-5	3/5	Healthcare administrative functions are seeing growing AI adoption (scheduling bots, automated intake) but overall healthcare IT adoption lags leading sectors like finance and tech.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI systems already assist nurses by auto-populating records, flagging scheduling conflicts, and suggesting appointment times, meaningfully reducing clerical burden while the human retains oversight and judgment over patient interactions.
Augmentation potential	claude-sonnet-5	4/5	AI scheduling tools, automated reminders, and EHR autofill meaningfully reduce clerical burden and let LPNs/LVNs focus more on patient-facing duties.
Task automatability	claude-haiku-4-5-20251001	3/5	Appointment scheduling and basic record-keeping can be partially automated with current AI systems (calendar APIs, simple data entry), but the task includes context-dependent judgment (patient needs, provider availability, follow-up logic) that requires human oversight, achieving roughly 50% time savings with proper integration.
Task automatability	claude-sonnet-5	4/5	Scheduling and records-keeping are largely structured, repetitive clerical tasks well within reach of current AI scheduling assistants and EHR automation tools, though some clinic-specific judgment and exception handling remain.
Adoption barriers	claude-haiku-4-5-20251001	3/5	HIPAA compliance and liability concerns require human oversight of patient records and appointment decisions; many organizations prefer human staff for appointment confirmation and patient communication, creating organizational friction despite no hard legal prohibition on automation.
Adoption barriers	claude-sonnet-5	2/5	No licensure requirement for clerical scheduling/record tasks, but HIPAA compliance, data security, and organizational inertia in healthcare settings create moderate friction.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI-driven scheduling and record automation cost significantly less than human clerical labor when amortized across multiple appointments and fully integrated, typically one order of magnitude cheaper for high-volume operations.
Cost vs. human wage	claude-sonnet-5	4/5	Automated scheduling/records software costs a small fraction of a clerical staff wage once implemented, though setup and integration with legacy EHR systems add some cost.
Technical feasibility today	claude-haiku-4-5-20251001	4/5	Deployed products (scheduling software with AI features, EHR systems with automated entry, chatbots for intake) perform parts of this task reliably in production at scale, though integration across legacy clinic systems and handling edge cases remains materially limiting.
Technical feasibility today	claude-sonnet-5	3/5	Deployed products (AI scheduling assistants, EHR chatbots, voice agents) exist and are used in some clinics, but many practices still rely on human staff due to integration gaps and workflow idiosyncrasies.

Prepare or examine food trays for conformance to prescribed diet.

CI 23–43 · exposure 33 · augmentation 50 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare sectors have relatively slow AI adoption overall, and dietary tray preparation remains a manual nursing task with minimal digitization or automation investment in most facilities due to safety criticality and liability concerns.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially bedside nursing tasks, adopts AI slowly due to safety-critical nature, fragmented EHR/dietary systems, and preference for human verification at the point of care.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI-powered visual inspection tools could assist nurses by flagging obvious dietary violations or providing decision support on diet-food matching, improving efficiency and reducing errors while the nurse retains verification authority.
Augmentation potential	claude-sonnet-5	3/5	AI-based dietary management software can flag conflicts between ordered diets and tray contents, helping nurses catch errors faster, though final physical verification remains manual.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI vision systems could theoretically identify foods on a tray, this task requires verification against individualized medical diet orders and handling of food items, which demands physical manipulation and real-time judgment of complex dietary restrictions that current automated systems cannot reliably perform end-to-end without substantial human oversight.
Task automatability	claude-sonnet-5	3/5	Checking a tray against a diet order is a structured, rule-based comparison task that AI/vision systems could largely handle, though physical tray inspection still requires a human or robotic actuator to verify actual food items.,
Adoption barriers	claude-haiku-4-5-20251001	4/5	Licensed nurses are legally responsible for medication and diet administration; regulations typically require a licensed professional to verify dietary compliance before meal service, creating a hard regulatory and liability barrier to full automation.
Adoption barriers	claude-sonnet-5	3/5	Patient safety liability (allergies, choking hazards, diet restrictions like diabetic or renal diets) creates meaningful oversight requirements, though this isn't a strictly licensed-only task by law.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	The infrastructure cost for camera systems, AI models, database integration, and required human verification would likely exceed the wage cost of having a nurse or dietary aide perform the task directly, especially in smaller healthcare facilities.
Cost vs. human wage	claude-sonnet-5	3/5	Software-based diet order cross-checking is cheap, but full automation requiring computer vision hardware and integration with hospital dietary systems narrows the cost advantage over a nurse's brief visual check.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Computer vision can detect and classify foods in images with moderate accuracy, but deployed systems lack reliable integration with dietary requirement databases and cannot handle the full range of medical diet complexities (allergies, drug interactions, texture modifications) that real-world healthcare requires; human review remains necessary.
Technical feasibility today	claude-sonnet-5	2/5	Some hospital dietary systems use software to flag diet order mismatches, but reliable automated visual verification of actual food trays against prescribed diets is not widely deployed in production nursing workflows.

Record food and fluid intake and output.

CI 23–39 · exposure 33 · augmentation 50 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare remains a laggard sector for autonomous AI deployment due to regulatory oversight, patient safety liability, and strong human-contact requirements. No meaningful adoption of AI systems for autonomous intake/output recording has occurred in hospital or clinical settings.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially nursing documentation workflows, adopts AI/automation slowly due to regulatory, safety, and interoperability constraints, with most facilities still relying on manual or semi-manual charting.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI could assist nurses by offering reminder prompts, automated data entry from structured inputs, or pattern detection in intake/output trends to flag abnormalities, moderately raising nurse efficiency while the nurse remains responsible for observation and clinical judgment.
Augmentation potential	claude-sonnet-5	3/5	Voice dictation, EHR templates, and connected monitoring devices can meaningfully speed up documentation and reduce transcription errors, assisting nurses without replacing their oversight.
Task automatability	claude-haiku-4-5-20251001	2/5	Recording intake and output requires observing and documenting patient data, which current AI cannot do independently without human observation and input. While AI could theoretically assist with data entry or flagging patterns, the core task of monitoring patient consumption and elimination requires direct clinical presence and judgment that AI systems lack.
Task automatability	claude-sonnet-5	3/5	Recording intake/output is largely data logging, which AI-enabled sensors and voice/EHR integration could handle, but reliable capture still requires human observation of what a patient actually consumes and physical measurement of output.4-1 combining sensors and manual entry limits full automation today.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Nursing practice acts require licensed nurses to perform or directly oversee patient monitoring and documentation; regulations stipulate that clinical observation and charting are professional nursing responsibilities. Liability and accountability for missed fluid balance issues create strong legal and organizational barriers to full automation.
Adoption barriers	claude-sonnet-5	3/5	While not requiring licensure per se, this data feeds into clinical decisions and liability chains, so there's institutional caution about relying on automated logging without a nurse's verification and signature.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI infrastructure cost for reliable observation and documentation would likely exceed the hourly wage of an LPN/LVN in most settings, especially when accounting for integration, oversight, and error correction needed for clinical safety. The task is relatively inexpensive for a human nurse to perform.
Cost vs. human wage	claude-sonnet-5	2/5	Sensor-based monitoring equipment and integration costs are substantial relative to the marginal cost of a nurse jotting down intake/output during routine care, so cost advantage is not yet clearly favorable to AI.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed AI system independently records patient intake/output in clinical practice today. Electronic health records (EHR) systems can store data that humans enter, but they do not autonomously perform the observation and documentation task. Current systems require a human to observe and manually input the information.
Technical feasibility today	claude-sonnet-5	2/5	Some hospitals use smart scales, connected IV pumps, and voice-to-EHR charting tools, but a fully automated, reliable intake/output tracking system across food and fluids is not yet standard deployed practice.

Measure and record patients' vital signs, such as height, weight, temperature, blood pressure, pulse, or respiration.

CI 25–30 · exposure 25 · augmentation 63 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	While remote monitoring devices see adoption in chronic disease management and some hospital units, the routine measurement and recording of vital signs by nurses remains predominantly manual and human-performed. Adoption of autonomous or AI-driven vital sign capture is still in pilot phases in most healthcare organizations.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially bedside nursing care, is a comparatively slow-adopting sector for AI-driven physical task automation, with automated vitals monitoring more common in ICU/telemetry than routine floor nursing.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI-assisted tools such as electronic vital sign dashboards, automated trending, and decision-support alerts can help nurses interpret and act on vital signs more efficiently. However, the core measurement task itself offers limited augmentation, as the nurse must still perform the physical act of measurement.
Augmentation potential	claude-sonnet-5	4/5	Automated vital sign monitors, wearable sensors, and EHR integration significantly speed up recording and reduce transcription error, letting nurses focus more time on interpretation and patient care while still performing hands-on measurement.
Task automatability	claude-haiku-4-5-20251001	2/5	While vital sign measurement devices (thermometers, blood pressure cuffs, pulse oximeters) exist and can output readings, a human must still perform the physical measurement, positioning, and interpretation. AI cannot autonomously position a patient or operate measurement devices without robotic hardware, which is rare in clinical settings today.
Task automatability	claude-sonnet-5	2/5	Vital sign measurement requires physical contact with the patient and manual manipulation of instruments; while automated sensors and smart devices exist, they don't replace the full task of positioning, applying, and interpreting readings across diverse patient conditions.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Vital sign measurement is a core clinical function that regulations and standard of care expect a licensed nurse to perform directly and be accountable for. Patient safety liability, accuracy requirements, and the legal expectation of licensed professional judgment create significant adoption barriers to full automation.
Adoption barriers	claude-sonnet-5	3/5	While not per se restricted to licensed personnel for the mechanical act of measurement, clinical protocols, liability for missed abnormal readings, and patient contact expectations in most care settings create real institutional friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current remote monitoring solutions and wearables add infrastructure cost and ongoing data management overhead. The loaded cost of these systems plus required clinical oversight often rivals or exceeds the wage cost of a nurse performing manual measurements, especially in high-volume clinical environments.
Cost vs. human wage	claude-sonnet-5	2/5	Automated vital-sign devices reduce marginal labor cost per reading, but they still require nursing staff for setup, patient contact, troubleshooting, and clinical judgment, so overall cost savings versus a nurse performing this task are modest, not order-of-magnitude.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Deployed products can read and record some vital signs (e.g., remote monitoring devices, wearables), but clinical-grade autonomous measurement by current AI systems in hospitals remains limited. Human nurses still perform and verify the vast majority of vital sign measurements in production settings.
Technical feasibility today	claude-sonnet-5	2/5	Automated blood pressure cuffs, digital thermometers, and pulse oximeters are widely deployed and reliable for isolated readings, but no integrated product autonomously performs the full task of measuring, contextualizing, and recording all vital signs for a patient without human involvement.

Answer patients' calls and determine how to assist them.

CI 16–35 · exposure 20 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Healthcare has moderate adoption of AI-assisted triage and call routing in large health systems, but uptake is constrained by liability concerns, regulatory caution, and legacy systems. Pilots are common but production deployment at scale remains limited compared to other sectors.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare bedside care lags in AI adoption due to physical, safety-critical, and regulatory constraints despite some digital triage tools elsewhere in healthcare.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can substantially augment nurses by preparing call summaries, suggesting protocols, retrieving patient history, and flagging urgent cases for immediate human review, allowing nurses to handle more calls and focus on complex clinical decisions rather than data lookup.
Augmentation potential	claude-sonnet-5	2/5	Some call-routing or nurse-call systems with basic triage prompts exist, but they offer limited assistance to the core task of physically responding and assessing needs.
Task automatability	claude-haiku-4-5-20251001	2/5	AI can partially automate initial triage and routing through conversational systems, but determining appropriate assistance often requires clinical judgment, understanding of patient history, and real-time assessment that current AI struggles with reliably. The task requires nuanced human interaction and occasional escalation to licensed staff.
Task automatability	claude-sonnet-5	2/5	Involves physically responding to call bells and triaging in-person patient needs, requiring presence and clinical judgment that current AI cannot perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Strong legal and liability barriers exist: nurses are responsible for patient assessment and safety decisions, and delegation to unmonitored AI creates legal exposure. Regulatory bodies (state nursing boards) and healthcare licensing requirements create hard barriers to full automation without human oversight and sign-off.
Adoption barriers	claude-sonnet-5	4/5	Direct patient care and triage decisions require licensed nursing judgment and physical presence, with liability and regulatory scope-of-practice constraints.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI-driven call triaging infrastructure has very low marginal cost per call compared to the loaded wage of a nurse aide or LPN performing this task, especially when many calls are simple questions answerable by documented protocols.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI substitute performing this physical, in-person task, so cost comparison favors the human by default.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Call handling chatbots exist and can field basic questions, but deployed systems frequently fail at clinical decision-making, patient safety prioritization, and handling atypical or complex presentations. Production systems typically route most calls to humans rather than fully resolving assistance determination.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product physically answers hospital call lights and performs bedside triage; this remains a human nursing function in practice.

Observe patients, charting and reporting changes in patients' conditions, such as adverse reactions to medication or treatment, and taking any necessary action.

CI 20–25 · exposure 25 · augmentation 75 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Healthcare remains cautious in automating clinical assessment tasks despite digitization progress; adoption of AI for observation is largely limited to supplementary monitoring in larger institutions, not mainstream displacement of nursing duties.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially bedside nursing, has been slow to adopt AI compared to information-sector jobs, with adoption concentrated in administrative/documentation support rather than clinical decision-making.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI-powered dashboards, vital sign alerts, and EHR integration substantially assist nurses by flagging anomalies, organizing data, and reducing documentation burden, allowing them to focus on direct patient assessment and intervention. These tools meaningfully enhance productivity while the nurse retains clinical authority.
Augmentation potential	claude-sonnet-5	4/5	AI-powered charting tools, early-warning scoring systems, and decision-support alerts can meaningfully assist nurses in tracking and flagging patient status changes, improving efficiency while the nurse remains responsible for judgment and action.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI systems can assist with parsing clinical data and flagging certain vital sign anomalies, the task requires real-time observation, clinical judgment about severity and context, and immediate interventions that demand human presence and accountability. Current AI cannot reliably replace the situational awareness and adaptive decision-making nurses provide during patient monitoring.
Task automatability	claude-sonnet-5	2/5	Direct patient observation and hands-on assessment (checking vitals, visual/physical cues, taking necessary action) require physical presence and clinical judgment that current AI cannot perform end-to-end; only the charting/documentation portion is automatable.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Strict regulatory and licensing requirements mandate that only licensed nurses perform patient observation and clinical assessment; liability for missed adverse reactions falls on the licensed provider. Legal scope of practice and patient safety regulations create hard barriers to full automation.
Adoption barriers	claude-sonnet-5	4/5	Licensed nurses are legally required to monitor patients and respond to clinical changes; liability, scope-of-practice regulations, and patient safety requirements create strong barriers to full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI monitoring infrastructure has significant upfront and ongoing costs (sensors, integration, oversight), and requires nursing staff to remain present and interpret alerts. The all-in cost is unlikely to undercut the wage of a practical nurse for equivalent safety and clinical outcomes.
Cost vs. human wage	claude-sonnet-5	2/5	AI can cheaply assist with note generation, but the core observation/response task still requires a paid nurse present, so overall cost savings are limited to documentation time.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Remote monitoring and alert systems exist in clinical settings, but they detect specific parameters (e.g., heart rate thresholds) rather than performing the full task of holistic patient observation and contextual reporting. These tools support nurses but do not independently observe patients or determine when to escalate care.
Technical feasibility today	claude-sonnet-5	2/5	Ambient documentation and clinical NLP tools exist and are being piloted in some hospitals, but reliable autonomous detection of adverse reactions and triggering appropriate action is not deployed at scale.

Supervise nurses' aides or assistants.

CI 3–25 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Healthcare adoption of AI for autonomy in high-stakes roles is slow and cautious. Nursing leadership roles remain human-centric; adoption is concentrated in documentation and decision support, not supervisory replacement. Sector-wide, this remains largely pilot or resistance-phase.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially direct patient care supervision, adopts AI slowly due to regulation, liability, and reliance on in-person oversight.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist supervisory work by tracking aide activity, flagging missed tasks, or suggesting scheduling improvements. However, augmentation is modest—AI does not transform the core judgment and relationship-building required to effectively supervise clinical staff, only reduces some administrative overhead.
Augmentation potential	claude-sonnet-5	2/5	AI can help with scheduling, documentation, or training aids that support supervisory tasks, but it doesn't meaningfully transform the interpersonal supervisory function itself.
Task automatability	claude-haiku-4-5-20251001	2/5	Supervising nursing aides requires real-time judgment, responsiveness to human behavior, and authority to delegate or redirect work. While AI could log tasks or flag deviations from protocol, actual supervision—holding accountability, adapting to staff needs, coaching—demands human authority and presence that AI cannot meaningfully replace at ≥50% time saving.
Task automatability	claude-sonnet-5	1/5	Supervising other staff requires real-time judgment, interpersonal management, accountability, and physical presence that current AI cannot replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Supervision of clinical staff in healthcare is often a legally and professionally mandated role of a licensed nurse. State nursing boards and hospital liability frameworks require a licensed nurse to be accountable for aide oversight, creating hard barriers to AI substitution without human sign-off and accountability.
Adoption barriers	claude-sonnet-5	5/5	Supervision of aides is tied to licensure, legal accountability for patient care, and regulatory scope-of-practice rules requiring a licensed nurse to oversee unlicensed staff.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	The cost of AI monitoring and integration infrastructure, plus human oversight of AI decisions, likely approaches or exceeds the cost of periodic human supervision. Nursing is labor-intensive in settings where supervisory spans are already optimized; full replacement would require re-architecture, raising total cost.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this supervisory role, so cost comparison favors the human entirely.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably performs real-time supervision of clinical staff in production. AI systems can assist with scheduling or monitoring task completion, but they cannot make binding supervisory decisions, handle conflicts, or take responsibility for oversight in healthcare settings where human accountability is mandated.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product performs supervisory management of aides/assistants in clinical settings; this remains a human management function.

Sterilize equipment and supplies, using germicides, sterilizer, or autoclave.

CI 5–23 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Adoption of autonomous sterilization automation in production healthcare is minimal; most clinical environments still rely on traditional sterilizer operation by licensed staff with few pilot programs or measurable displacement.
Sector adoption velocity	claude-sonnet-5	1/5	Physical, hands-on clinical support tasks in healthcare facilities show minimal AI adoption; this is a low-digitization, equipment-handling task with no momentum toward AI automation.
Augmentation potential	claude-haiku-4-5-20251001	3/5	Modern autoclaves and sterilization tracking systems (RFID, software dashboards) assist nurses by automating documentation and alerting them to cycles and failures, meaningfully improving workflow efficiency while the nurse retains oversight.
Augmentation potential	claude-sonnet-5	2/5	AI could help track sterilization logs, schedules, or compliance documentation, but offers no direct assistance with the physical sterilization process itself.
Task automatability	claude-haiku-4-5-20251001	2/5	Sterilization requires loading equipment, setting parameters, and monitoring processes—most repetitive steps are procedural, but verification of sterility, handling of delicate instruments, and responding to equipment failures need human judgment. Current systems lack the dexterity and adaptability to fully replace the human in a way that meets the ≥50% time-saving threshold.
Task automatability	claude-sonnet-5	1/5	This is a hands-on physical task requiring manual handling of instruments, loading autoclaves, and verifying sterility; no current AI system can perform the physical manipulation involved.'
Adoption barriers	claude-haiku-4-5-20251001	4/5	Sterilization is a critical safety function in healthcare; regulatory bodies (FDA, state licensing boards) typically require documented oversight and sign-off by licensed personnel, creating a hard legal and liability barrier to full automation without human verification.
Adoption barriers	claude-sonnet-5	4/5	Sterilization is governed by strict infection-control protocols and regulatory standards, requiring trained personnel to verify and document proper sterilization, creating strong procedural and liability barriers.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current robotic and automation solutions for sterilization are capital-intensive and still require significant oversight, making them costlier than a licensed nurse's labor when amortized across actual workload in most healthcare settings.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute, so any 'AI' cost comparison is moot; existing autoclave automation is mechanical, not AI-driven, and doesn't replace the labor of loading/monitoring/verification.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While autoclaves are highly automated machines, end-to-end sterilization automation (loading, parameter verification, unloading, validation checks) is not yet a deployed product category in clinical settings. Robotic arms exist but are not standard in production healthcare environments for this task.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI/robotic products perform clinical equipment sterilization in healthcare settings today; sterilization remains a manual or fixed-automation (non-AI) process.

Evaluate nursing intervention outcomes, conferring with other healthcare team members as necessary.

CI 3–25 · exposure 13 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Healthcare organizations are adopting AI for data flagging and decision support, but adoption of autonomous outcome evaluation is slow due to regulatory constraints, liability concerns, and the requirement for human clinical judgment in patient care decisions.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially bedside nursing, adopts autonomous AI slowly due to regulatory, safety, and liability constraints despite growing use of documentation and decision-support tools.
Augmentation potential	claude-haiku-4-5-20251001	4/5	Current AI systems can significantly augment nurse productivity by summarizing patient data, highlighting deviations from expected outcomes, and flagging trends—allowing nurses to focus their evaluation time on interpretation and conferencing rather than raw data gathering.
Augmentation potential	claude-sonnet-5	3/5	AI can help by summarizing patient data, flagging abnormal trends, or drafting documentation to support the nurse's evaluation and communication, though the core judgment remains human-driven.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist in analyzing vital signs and clinical data, evaluating intervention outcomes requires contextual clinical judgment and often involves conferring with other healthcare professionals—a nuanced collaborative activity that current systems cannot perform end-to-end reliably without substantial human oversight.
Task automatability	claude-sonnet-5	1/5	This requires hands-on patient assessment, clinical judgment integrating physical findings with context, and real-time interprofessional communication—none of which current AI can perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	State licensure laws and clinical liability requirements mandate that a licensed nurse (LPN/LVN) evaluate and sign off on nursing intervention outcomes; automation of the entire evaluation function would likely violate scope-of-practice regulations and create unacceptable liability risk.
Adoption barriers	claude-sonnet-5	5/5	Licensure, scope-of-practice regulation, and legal accountability for clinical judgment and patient safety mandate a licensed nurse perform this evaluation and communication.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI solutions for clinical outcome assessment require integration, validation, and ongoing human review, making the all-in cost comparable to or exceeding the cost of a licensed nurse performing the evaluation directly.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this full task, so cost comparison favors the human nurse who must remain accountable and present.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably evaluates nursing intervention outcomes independently; systems exist for data analysis and alerts but they require significant human interpretation and cannot substitute for the clinical decision-making required in production nursing environments.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product autonomously evaluates nursing intervention outcomes or confers with care teams; this remains firmly a human clinical responsibility.

Set up equipment and prepare medical treatment rooms.

CI 5–19 · exposure 8 · augmentation 25 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare remains a laggard in physical automation; nursing tasks are highly dependent on human presence, regulatory approval, and clinical risk tolerance, with minimal displacement by AI in this domain today.
Sector adoption velocity	claude-sonnet-5	1/5	Healthcare facility physical operations show minimal AI/robotic adoption for routine room and equipment preparation tasks.
Augmentation potential	claude-haiku-4-5-20251001	2/5	Digital inventory systems and checklists can help nurses organize preparation tasks, but current AI offers minimal cognitive or physical assistance in the core work of arranging, testing, and validating medical equipment in treatment rooms.
Augmentation potential	claude-sonnet-5	2/5	AI could assist with checklists, inventory tracking, or scheduling reminders for room setup, but offers little direct help with the physical preparation itself.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI could assist with inventory tracking and equipment checklists, the task requires physical manipulation of sensitive medical equipment, spatial arrangement in real rooms, and validation that setups meet sterility and safety standards—tasks beyond current robotic or AI automation deployed in clinical settings.
Task automatability	claude-sonnet-5	1/5	This is a physical task requiring manual handling and positioning of equipment, sterile setup, and room arrangement that current AI systems cannot perform without robotic embodiment, which is not deployed for this purpose.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Medical licensing requirements, liability asymmetry (errors in equipment setup can cause patient harm), regulatory oversight (FDA, state nursing boards), and the requirement that a licensed professional verify and take responsibility for room preparation create strong legal and organizational barriers.
Adoption barriers	claude-sonnet-5	3/5	While not always requiring a specific license to move equipment, clinical safety, infection control protocols, and facility policy create meaningful organizational friction against non-human execution.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The physical robotics and specialized medical equipment handling required would be far more expensive than a licensed nurse's labor, with no mature commercial solution available.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this physical task, so any hypothetical robotic solution would be far more expensive than a nurse's labor for this purpose.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No production AI or robotic systems reliably set up and prepare medical treatment rooms end-to-end in hospital environments today; this remains entirely human-performed despite some digital inventory tools.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI or robotic product performs equipment setup and treatment room preparation in clinical settings today; this remains manual human labor.

Work as part of a healthcare team to assess patient needs, plan and modify care, and implement interventions.

CI 3–20 · exposure 13 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Healthcare adoption of AI agents for autonomous care planning remains pilot-stage; regulatory caution, liability concerns, and the necessity of human licensure have slowed production deployment despite high digitization in health IT infrastructure.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare is a historically slow-adopting sector for full task automation due to regulatory, safety, and liability constraints, though administrative AI tools are spreading.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can meaningfully assist nurses by automating documentation, flagging vital-sign anomalies, suggesting evidence-based interventions, and organizing patient data, allowing nurses to focus on clinical judgment and interpersonal care while retaining full responsibility.
Augmentation potential	claude-sonnet-5	3/5	AI can assist with documentation, care plan templates, clinical decision support, and flagging risks, improving efficiency while the nurse retains responsibility for assessment and intervention.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist with data aggregation and documentation, the task requires real-time clinical judgment, direct patient interaction, and dynamic care modification that depends on nuanced human assessment and responsiveness. Current systems cannot reliably perform the full cycle of assessment-planning-implementation end-to-end.
Task automatability	claude-sonnet-5	1/5	This requires hands-on physical care, direct patient assessment, and real-time clinical judgment integrated with team collaboration, none of which current AI can perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Strong legal and regulatory barriers exist: nursing licensure is legally mandated, liability falls directly on the licensed nurse, and healthcare regulations (including state boards and federal law) require a licensed practitioner to assess, plan, and be accountable for patient care decisions.
Adoption barriers	claude-sonnet-5	5/5	Nursing practice is heavily regulated, requires licensure, direct patient contact, and legal accountability for care decisions, creating hard barriers to automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI infrastructure and oversight for clinical tasks remains costly and requires human nurses to verify and act on recommendations, making the all-in cost per completed care cycle comparable to or exceeding direct nurse labor for equivalent outcomes.
Cost vs. human wage	claude-sonnet-5	1/5	AI cannot substitute for the physical presence and licensed judgment required, so there is no viable cost comparison for full task replacement.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably performs this complete clinical task independently; AI tools exist for narrow sub-tasks (vital monitoring, documentation drafting) but lack the embodied presence, clinical reasoning, and accountability required for integrated care planning and team coordination in production settings.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product autonomously assesses patients, plans care, and implements interventions; AI clinical tools remain decision-support aids, not autonomous care providers.

Clean rooms and make beds.

CI 5–18 · exposure 0 · augmentation 13 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Healthcare is a traditionally risk-averse, highly regulated sector with slow technology adoption outside of diagnostics and imaging. Although robotic cleaning is discussed, actual deployment in hospitals remains minimal and experimental.
Sector adoption velocity	claude-sonnet-5	1/5	Healthcare housekeeping and bedside physical tasks show minimal AI/robotic adoption; this is a highly manual, low-digitization task within a sector that adopts automation slowly for physical labor.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI can assist with work scheduling and room-status tracking, but offers limited augmentation for the physical act of cleaning and bed-making itself; human nurses remain essential for the hands-on work and patient care judgment.
Augmentation potential	claude-sonnet-5	1/5	Current AI tools offer no meaningful assistance for the physical acts of cleaning rooms or making beds.
Task automatability	claude-haiku-4-5-20251001	1/5	Cleaning rooms and making beds require physical manipulation in unstructured, cluttered environments with fragile medical equipment and patient-specific needs. Current AI robots lack the dexterity, environmental understanding, and real-time adaptation to perform this reliably at scale.
Task automatability	claude-sonnet-5	1/5	This is a physical manipulation task requiring mobility, dexterity, and navigation of real-world clinical environments that current AI systems cannot perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Healthcare facilities face licensing and liability concerns around patient safety and infection control; regulatory bodies oversee cleanliness standards, and many organizations prefer human staff for quality assurance and patient interaction during room maintenance.
Adoption barriers	claude-sonnet-5	2/5	No licensing or legal requirement mandates a nurse specifically perform this task, but practical/organizational barriers (infection control protocols, physical environment complexity) limit automation via non-robotic AI.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Specialized cleaning and bed-making robots remain extremely expensive to purchase, maintain, and deploy compared to the hourly wage of LPNs/LVNs, especially when accounting for integration, downtime, and human oversight.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI (software) solution for this physical task, so cost comparison favors human labor entirely; any robotic solution would be far more expensive than a human worker for this task.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed commercial product reliably cleans hospital rooms or makes beds autonomously in production healthcare settings. Prototype robotics exist but are not yet reliable enough or cost-effective for real-world healthcare environments.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product cleans patient rooms or makes hospital beds; this remains firmly in the domain of human labor or basic robotics still in limited pilot use.

Prepare patients for examinations, tests, or treatments and explain procedures.

CI 9–11 · exposure 16 · augmentation 50 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	This task is embedded in heavily regulated healthcare workflows with mandatory licensure; adoption of automation is near-zero because legal and patient-safety requirements mandate human nurses.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare, especially bedside nursing tasks, has been slow to adopt AI for hands-on physical care due to regulatory, safety, and infrastructure constraints.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI could assist by drafting procedure explanations or generating patient education materials that a nurse then personalizes and delivers, or by prompting checklists for preparation—useful supporting tools but the nurse remains the essential interface.
Augmentation potential	claude-sonnet-5	3/5	AI can help generate patient-facing explanations, translate instructions, or provide checklists that support the nurse's communication with patients, improving efficiency around the task even if it can't perform the physical component.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI could potentially draft explanations of procedures, the task fundamentally requires physical preparation of patients (positioning, equipment setup, vitals) and real-time, empathetic communication that adapts to individual patient anxiety and comprehension levels—elements that current AI cannot reliably perform end-to-end in a clinical setting.
Task automatability	claude-sonnet-5	2/5	Explaining procedures verbally could be partially scripted or delivered via AI-generated materials, but physical preparation of patients (positioning, vitals, gowning, IV setup) requires hands-on human action that current AI cannot perform.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Licensing laws require an RN or LPN to perform patient preparation and assessment; liability and patient-care standards mandate direct human oversight and accountability for explanations and physical handling.
Adoption barriers	claude-sonnet-5	5/5	Direct patient care, physical contact, and clinical judgment during preparation are legally restricted to licensed nursing staff, creating strong regulatory and liability barriers.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI has no cost advantage here because the task cannot be meaningfully automated—a licensed nurse's time is the baseline cost, and AI tools capable of assisting would add overhead rather than replace the work.
Cost vs. human wage	claude-sonnet-5	1/5	Physical patient preparation still requires an on-site human nurse, so AI cannot substitute for the labor cost of this task.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system can independently prepare a patient physically or deliver procedure explanations with the clinical judgment and responsiveness required; this remains a task requiring direct human-patient interaction.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product physically prepares patients for exams or treatments; at most, chatbots or patient education apps supplement verbal explanations but do not replace the nurse's role.

Provide medical treatment or personal care to patients in private home settings, such as cooking, keeping rooms orderly, seeing that patients are comfortable and in good spirits, or instructing family members in simple nursing tasks.

CI 0–14 · exposure 8 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Home health care remains a low-digitization, labor-intensive sector with fragmented small providers, older patient populations, and strong human-contact requirements. Adoption of AI agents for autonomous patient care in homes is negligible; the sector lags in automation broadly.
Sector adoption velocity	claude-sonnet-5	1/5	Home health and personal care sectors have very low AI/robotics adoption for hands-on physical tasks; this is among the most laggard segments for automation given physical and regulatory constraints.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can meaningfully assist nurses with administrative tasks (scheduling, documentation, patient reminders) and provide quick reference information, but the core caregiving, comfort assessment, and family teaching still require deep human judgment and presence. Assistance is real but limited in scope.
Augmentation potential	claude-sonnet-5	2/5	AI could help with scheduling, care instructions, or reminders for family members, but offers minimal assistance for the core physical caregiving, cooking, and housekeeping components of this task.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI could assist with scheduling care, reminding patients about medications, or providing basic health information, the core tasks—physical patient care (cooking, comfort positioning), emotional support (keeping patients 'in good spirits'), and real-time responsiveness to patient needs in a home environment—require embodied presence and human judgment. Current AI systems cannot reliably perform end-to-end home care at 50% time savings.
Task automatability	claude-sonnet-5	1/5	This task requires hands-on physical care, cooking, cleaning, and direct human presence in a private home—none of which current AI systems can perform end-to-end; robotics for this level of unstructured physical caregiving is not deployable.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Licensed practical and vocational nurses must hold state licensure to practice; many jurisdictions have scope-of-practice laws that explicitly require a licensed human to assess patient needs, provide hands-on care, and instruct family members. Liability, patient safety, and legal requirements create hard barriers to autonomous AI substitution.
Adoption barriers	claude-sonnet-5	4/5	Licensing requirements for LPN/LVN scope of practice, liability concerns for medical tasks performed in unsupervised home settings, and the inherently physical/relational nature of care create strong barriers to automation.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI assistive tools (chatbots, monitoring devices) cost money to deploy and require human oversight, while the bulk of the work—physical caregiving and presence—still requires licensed nurses or aides. The total cost of AI-plus-human is likely higher than human-alone for this task.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI substitute performing physical in-home care tasks, so AI cost is not comparable—human labor remains the only viable option, making AI effectively far more 'expensive' (i.e., non-functional) for this task.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product reliably performs integrated home nursing care autonomously. Narrow subcomponents (medication reminders, telehealth triage) exist, but cooking, physical comfort care, and family instruction in actual home settings remain fully human-dependent with no production-grade AI alternative.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product provides physical in-home nursing care, housekeeping, or hands-on patient comfort; this remains entirely a human physical and interpersonal task.

Assemble and use equipment, such as catheters, tracheotomy tubes, or oxygen suppliers.

CI 0–9 · exposure 8 · augmentation 25 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare adoption of autonomous robotic nursing assistants for device assembly and insertion remains minimal; most clinical settings still rely on human nurses, and regulatory/liability concerns slow pilot-to-production transitions significantly.
Sector adoption velocity	claude-sonnet-5	1/5	Physical, hands-on clinical care tasks in nursing show minimal AI/robotic adoption; healthcare bedside procedures remain a low-digitization, high-touch domain.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI offers limited augmentation for this task; while AR guidance or real-time coaching might assist a nurse, the core activity is manual and physical, and current AI tools provide minimal productivity gain for an already-practiced clinical skill.
Augmentation potential	claude-sonnet-5	2/5	AI can assist with training simulations, checklists, or documentation related to equipment use, but offers little direct augmentation of the physical assembly and application task itself.
Task automatability	claude-haiku-4-5-20251001	2/5	Physical assembly and insertion of medical equipment (catheters, tracheotomy tubes, oxygen suppliers) requires manual dexterity, real-time tactile feedback, and precise positioning that current AI and robotics cannot reliably perform in uncontrolled clinical settings. While some robotic prototypes exist, they are not deployed at scale for routine nursing tasks.
Task automatability	claude-sonnet-5	1/5	This is a hands-on physical task requiring manual dexterity, patient contact, and real-time clinical judgment that current AI cannot perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task is a direct hands-on patient intervention that is legally and professionally required to be performed or directly supervised by a licensed healthcare professional; liability, patient safety, and regulatory frameworks (state nursing boards, medical device regulations, patient consent) create hard barriers to autonomous substitution.
Adoption barriers	claude-sonnet-5	5/5	This is an invasive/medical procedure requiring licensed personnel, direct physical patient contact, and legal/regulatory accountability, making substitution essentially barred.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Current robotic systems capable of medical device assembly and insertion are extremely expensive (hundreds of thousands to millions), far exceeding the loaded cost of a licensed practical/vocational nurse performing these tasks.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this physical procedure, so cost comparison favors the human by default; any robotic solution would be far more expensive than a nurse's wage.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product reliably performs physical assembly and insertion of catheters or tracheotomy tubes in clinical production today. Research prototypes exist but lack the reliability, safety clearance, and integration into workflows needed for real-world deployment.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product or robotic system autonomously assembles and applies catheters, tracheostomy tubes, or oxygen equipment on patients in production settings.

Collect samples, such as blood, urine, or sputum from patients, and perform routine laboratory tests on samples.

CI 0–5 · exposure 5 · augmentation 25 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare adoption of AI for this specific task is minimal; sample collection and routine laboratory testing remain firmly within human clinical workflows. The heavily regulated and hands-on nature of this work limits automation velocity.
Sector adoption velocity	claude-sonnet-5	1/5	Healthcare's physical, hands-on clinical procedures show minimal AI adoption; this sub-task within nursing remains untouched by automation trends seen in administrative or diagnostic support areas.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could provide minimal assistance such as suggesting appropriate tests based on patient history or flagging unusual results for review, but the core task of physical collection and hands-on test execution offers limited augmentation opportunities.
Augmentation potential	claude-sonnet-5	2/5	AI can assist with lab result interpretation, tracking, or documentation after sample collection, but offers negligible support for the physical act of collecting samples itself.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires physical collection of biological samples from patients, which demands hands-on interaction with patients and sterile technique that current AI systems cannot perform. Even laboratory analysis of samples requires specialized lab equipment operation, which current AI cannot autonomously execute.
Task automatability	claude-sonnet-5	1/5	Physical sample collection requires manual dexterity, patient contact, and physical manipulation that current AI systems cannot perform; this is a hands-on clinical task with no software substitute.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Significant legal and regulatory barriers exist: licensed practical/vocational nurses are required by law to perform or directly supervise sample collection, and clinical laboratory tests must be performed by licensed personnel under CLIA regulations. Patient contact requirements and liability for specimen handling further strengthen barriers to automation.
Adoption barriers	claude-sonnet-5	5/5	Sample collection is a licensed clinical procedure requiring hands-on human contact, infection control protocols, and legal scope-of-practice authorization, making substitution essentially impossible under current regulation and physical constraints.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI systems cannot yet perform the physical sample collection or operate the specialized laboratory equipment needed for routine tests, making a meaningful cost comparison infeasible. The task requires licensed human nurses and laboratory technicians.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI system capable of performing this physical task, so cost comparison is moot; the human labor cost is the only viable option currently.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	While AI can assist with interpreting some laboratory test results (e.g., image analysis), no deployed product can autonomously collect patient samples or perform routine laboratory tests end-to-end. Limited applications exist for specific result interpretation, but the core task remains human-dependent.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product collects blood, urine, or sputum samples from patients; this remains entirely a human physical task performed by trained clinical staff.

Help patients with bathing, dressing, maintaining personal hygiene, moving in bed, or standing and walking.

CI 0–5 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare remains a relatively low-adoption sector for AI-driven automation of clinical tasks; physical care automation is in pilot or research phase, with minimal production deployment in nursing.
Sector adoption velocity	claude-sonnet-5	1/5	Direct patient physical care in healthcare/long-term care settings shows minimal AI or robotic adoption due to physical, regulatory, and safety constraints.
Augmentation potential	claude-haiku-4-5-20251001	2/5	Limited assistive devices (e.g., mechanical lifts, mobility aids) exist, but AI augmentation of the decision or assessment portions of this task is minimal; the core task remains fundamentally hands-on and human-delivered.
Augmentation potential	claude-sonnet-5	2/5	AI can support scheduling, monitoring, or documentation around these tasks, but offers little direct assistance to the hands-on physical act of bathing, dressing, or moving patients.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires direct physical contact, assistance with intimate personal care, and real-time adaptation to patient mobility and comfort needs. Current AI systems cannot perform physical manipulation, and no robotic system is deployed at scale for these activities in clinical settings.
Task automatability	claude-sonnet-5	1/5	This is direct physical care requiring manual dexterity, mobility support, and hands-on assistance with fragile or vulnerable patients—no current AI system can perform physical bathing, dressing, or ambulation support.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Regulatory and licensing frameworks legally require a licensed nurse to provide direct patient care; liability and duty-of-care standards create hard barriers, and patient dignity and safety requirements necessitate human contact and judgment.
Adoption barriers	claude-sonnet-5	4/5	Physical care of vulnerable patients carries high liability and safety risk, often requires trained/certified personnel, and patients strongly expect human touch and judgment during hygiene and mobility assistance.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The capital and integration costs of robotic systems capable of safe physical patient care far exceed the loaded wage of a licensed practical nurse, with no mature cost-effective alternative available today.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute for this physical labor, so any comparison favors the human caregiver entirely; specialized robotics would be far more costly than nursing aide labor today.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	While robotic assistance exists in research, no deployed AI or robotic products reliably perform bathing, dressing, or physical mobility assistance at the scale and quality required for clinical care. The task demands tactile feedback, safety judgment, and human presence that remain infeasible to automate.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product performs hands-on physical patient care like bathing, dressing, or transfer assistance; robotic patient-lifting aids exist only in limited research/pilot contexts.

Apply compresses, ice bags, or hot water bottles.

CI 0–5 · exposure 0 · augmentation 0 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare remains a laggard sector for physical automation; comfort-care tasks are low-priority for automation investment and face strong institutional resistance to replacing direct patient contact.
Sector adoption velocity	claude-sonnet-5	1/5	Direct bedside physical care tasks in healthcare show minimal AI adoption; this remains a manual, low-digitization task performed by humans.
Augmentation potential	claude-haiku-4-5-20251001	1/5	AI offers minimal assistance for a straightforward manual and observational task; the nurse's direct presence, tactile judgment, and patient reassurance cannot be meaningfully augmented by current AI.
Augmentation potential	claude-sonnet-5	1/5	AI offers essentially no assistance for the physical act of applying compresses or hot/cold packs; it's a manual procedural task.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires physical manipulation of medical equipment and direct patient contact to assess comfort and skin integrity—capabilities beyond current AI systems. Robotic arms exist in research but are not deployed for routine clinical comfort care.
Task automatability	claude-sonnet-5	1/5	This is a physical hands-on task requiring direct patient contact and manual dexterity; current AI systems cannot apply physical items to a patient's body.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Patient care tasks involving physical contact and clinical judgment are legally and organizationally restricted to licensed nursing staff; regulatory frameworks require human licensure for direct patient care delivery.
Adoption barriers	claude-sonnet-5	4/5	Direct patient care involving physical contact and clinical judgment about skin integrity/temperature tolerance typically requires a licensed nurse, creating strong practical and regulatory barriers to automation.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Purpose-built robotics for compress/compress application would be far more expensive than the few minutes of nurse labor required, with significant integration and maintenance overhead.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI-based substitute performing this physical task, so no meaningful cost comparison exists; human labor is the only option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI or robotic system reliably performs this task in clinical settings today. The task demands real-time tactile feedback, temperature adjustment, and patient interaction.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product performs physical application of compresses or thermal therapy devices to patients; this remains purely a manual nursing task.

Wash and dress bodies of deceased persons.

CI 0–5 · exposure 0 · augmentation 0 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Mortuary and nursing sectors show minimal AI adoption for clinical or end-of-life tasks, and cultural and regulatory factors make substitution unlikely even in digitized healthcare settings.
Sector adoption velocity	claude-sonnet-5	1/5	Healthcare direct-care and mortuary-adjacent physical tasks show minimal AI adoption; this is a low-digitization, physical-labor task with no observed automation trend.
Augmentation potential	claude-haiku-4-5-20251001	1/5	AI offers no meaningful assistance for the core physical task of washing and dressing a body; the work is primarily manual and interpersonal, with no clear role for computational support.
Augmentation potential	claude-sonnet-5	1/5	AI provides no meaningful assistance to the physical, respectful handling and preparation of a deceased person's body.
Task automatability	claude-haiku-4-5-20251001	1/5	This task involves physical contact with human remains and requires manual dexterity, tactile sensitivity, and contextual judgment that current robotics cannot reliably perform. No AI or robotic system today can meaningfully automate the end-to-end washing, dressing, and positioning of a deceased body.
Task automatability	claude-sonnet-5	1/5	This is a physical, manual task requiring hands-on manipulation of a body with dignity and care; no current AI/robotic system can perform this end-to-end.'
Adoption barriers	claude-haiku-4-5-20251001	5/5	This task is legally and culturally bound to human practitioners; many jurisdictions require licensed personnel to prepare bodies, and families typically expect dignified human care. Strong regulatory and human-contact barriers protect this work.
Adoption barriers	claude-sonnet-5	4/5	Handling deceased bodies involves strong cultural, ethical, and often religious/legal expectations for human care and dignity, along with facility protocols, though not always a strict licensure requirement specific to this act.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The cost of specialized robotics hardware, software, integration, and maintenance would far exceed the hourly wage of a licensed practical nurse performing this task, making automation economically infeasible today.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI-based alternative to compare cost against; the task must be performed by human staff, making AI substitution infeasible at any cost.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed commercial products perform this task. Specialized mortuary equipment exists for other purposes, but no robotic or AI system in production can autonomously wash and dress human remains.
Technical feasibility today	claude-sonnet-5	1/5	No deployed products perform postmortem care; this remains entirely outside the scope of current AI or robotics offerings in healthcare settings.

Administer prescribed medications or start intravenous fluids, noting times and amounts on patients' charts.

CI 0–3 · exposure 0 · augmentation 38 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare remains heavily regulated and safety-critical; no adoption of AI autonomously administering medications is occurring or foreseeable. This task is among the most protected in clinical care.
Sector adoption velocity	claude-sonnet-5	2/5	Healthcare bedside care remains a laggard sector for AI-driven task automation, though EHR/documentation tools are increasingly used to support charting.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist with documentation (charting times/amounts post-administration) or reminder systems, but offers minimal augmentation to the core clinical skills of medication/IV administration itself.
Augmentation potential	claude-sonnet-5	3/5	AI-enabled voice dictation, smart pumps, and EHR alerts can streamline documentation and reduce dosage errors, offering moderate assistance to the charting and monitoring portion of the task.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires direct physical administration of medications/IV fluids and clinical judgment about patient response. Current AI cannot perform the hands-on nursing actions, physical assessment, or real-time adjustments needed at the bedside.
Task automatability	claude-sonnet-5	1/5	Physical administration of medications and starting IVs requires hands-on manipulation of a patient's body, which current AI cannot perform; only the charting/documentation sliver is automatable.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Strict licensing requirements mandate that only licensed nurses perform medication administration and IV placement. Legal liability, regulatory requirements (state nursing boards), and patient safety standards create hard barriers to automation.
Adoption barriers	claude-sonnet-5	5/5	Medication administration and IV starts are licensed clinical acts governed by nursing scope-of-practice laws, requiring a credentialed human with legal accountability for patient safety.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI cannot substitute for the licensed nurse performing this task. There is no cost comparison because automation is infeasible; a human must physically perform the administration.
Cost vs. human wage	claude-sonnet-5	1/5	AI cannot replace the physical labor and clinical judgment involved, so the human cost is unavoidable; any AI role is a minor add-on to, not a substitute for, the nurse's wage.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product can autonomously administer medications or start IVs. This remains entirely human-dependent in clinical practice, with AI having no production role in the core clinical action.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product administers medications or starts IV lines; this remains purely a human physical-clinical task, with AI at most assisting documentation via voice-to-text or EHR integration.

Provide basic patient care or treatments, such as taking temperatures or blood pressures, dressing wounds, treating bedsores, giving enemas or douches, rubbing with alcohol, massaging, or performing catheterizations.

CI 0–0 · exposure 0 · augmentation 25 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Healthcare remains heavily regulated and risk-averse; robotics deployment in direct patient care is still nascent and confined to pilot programs in well-resourced academic medical centers, far from industry-wide production adoption.
Sector adoption velocity	claude-sonnet-5	1/5	Direct physical patient care in nursing remains one of the least digitized, most hands-on healthcare functions with minimal AI/robotic penetration in production settings.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI might assist with vital-signs monitoring dashboards or wound documentation, but most of the physical, tactile, and assessment components of this task offer limited scope for meaningful AI augmentation while the nurse remains primary.
Augmentation potential	claude-sonnet-5	2/5	AI can assist with documentation, vital sign trend analysis, or wound photo assessment, but offers minimal direct assistance during the physical performance of these hands-on procedures.
Task automatability	claude-haiku-4-5-20251001	1/5	These tasks require physical manipulation, tactile assessment, and direct patient contact that current robots cannot reliably perform outside controlled laboratory settings. Temperature and blood pressure reading are the only measurable components, but they represent a small fraction of this diverse, hands-on clinical work.
Task automatability	claude-sonnet-5	1/5	This task requires direct physical manipulation of patient bodies (wound dressing, catheterization, massage) that current AI systems, lacking robotic embodiment in clinical settings, cannot perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Nursing care is legally regulated; licensed nurses must provide or directly oversee most of these tasks. Patient safety liability, clinical judgment requirements, and state licensure laws create hard barriers to full automation.
Adoption barriers	claude-sonnet-5	5/5	Direct clinical procedures like catheterization and wound care require licensed personnel under nursing scope-of-practice laws and carry significant liability, making legal and safety barriers very high.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The capital, maintenance, and integration costs of medical robotics vastly exceed the loaded wage of an LPN/LVN, particularly when factoring in the specialized liability insurance, regulatory compliance, and human oversight required for patient safety.
Cost vs. human wage	claude-sonnet-5	1/5	There is no AI system capable of substituting for the physical labor involved, so cost comparison is moot—the human is the only viable option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI or robotic systems reliably perform wound dressing, catheterization, enema administration, or bedside massage in production healthcare environments. Isolated sensor-based vital signs collection exists, but the full scope of physical clinical care remains research-stage.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product performs hands-on physical nursing procedures like catheterization or wound dressing; this remains entirely in the domain of human clinical staff.

Related occupations — Healthcare Practitioners & Technical

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.