Nursing Assistants

31-1131.00
Median wage $42,260/yr1,448,910 employed (US)Rank #812 of 923 scored · top 88% by substitution

Provide or assist with basic care or support under the direction of onsite licensed nursing staff. Perform duties such as monitoring of health status, feeding, bathing, dressing, grooming, toileting, or ambulation of patients in a health or nursing facility. May include medication administration and other health-related tasks. Includes nursing care attendants, nursing aides, and nursing attendants.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution14
Exposure11
Augmentation31

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

33 rated tasks, binned by substitution score.

Position among all scored occupations

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

Tasks with substitution ≥ 70

0%

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

Why this score

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

Task automatabilityw 35%11

panel mean rating 1.4/5 → substitution pressure 11/100

Technical feasibility todayw 20%12

panel mean rating 1.5/5 → substitution pressure 12/100

Cost vs. human wagew 15%12

panel mean rating 1.5/5 → substitution pressure 12/100

Adoption barriersw 20%inverted — strong barriers lower the score26

panel mean rating 4.0/5 (barrier strength) → substitution pressure 26/100

Sector adoption velocityw 10%10

panel mean rating 1.4/5 → substitution pressure 10/100

Task breakdown (33 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.

Provide information, such as directions, visiting hours, or patient status information to visitors or callers.

56

CI 37–74 · exposure 58 · augmentation 50 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Healthcare has rapidly deployed phone systems, chatbots, and portal automation for routine inquiries. Major hospital systems now offer AI-driven visitor information and appointment lines; adoption is broad and accelerating in digitized health systems.
Sector adoption velocityclaude-sonnet-52/5Healthcare facilities lag in adopting AI for patient-facing communication tasks due to compliance concerns, though basic chatbot/kiosk adoption for directions is more common.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist a nursing assistant by pre-screening calls, looking up standard information, and drafting responses, freeing the human to handle complex or sensitive inquiries. However, the task is simple enough that human-only workflow is still common; augmentation is useful but not transformative.
Augmentation potentialclaude-sonnet-53/5AI-powered directory assistants and scheduling bots can offload simple informational requests, freeing nursing assistants to focus on higher-sensitivity communications.
Task automatabilityclaude-haiku-4-5-202510014/5This task is highly routine and information-retrieval based. Current AI systems can reliably answer FAQs about visiting hours, directions, and provide general patient status updates (when authorized) through chatbots or voice agents. The primary bottleneck is integration with hospital systems and handling edge cases, but the core task easily meets the 50% time-saving threshold.
Task automatabilityclaude-sonnet-53/5Simple informational queries (directions, visiting hours) could be handled by chatbots or automated phone systems, but conveying accurate patient status requires access to real-time records and judgment about privacy/appropriateness, limiting full automation.
Adoption barriersclaude-haiku-4-5-202510013/5HIPAA and privacy regulations require careful handling of patient information disclosure; hospitals may restrict automated release of certain patient status details and prefer human verification for sensitive inquiries. Organizational inertia and staff preferences for human contact add friction, but no hard legal requirement forbids automation.
Adoption barriersclaude-sonnet-54/5HIPAA and patient privacy regulations create strong barriers to disclosing status information via automated systems without verified authorization, requiring human oversight for sensitive parts of this task.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI inference and chatbot hosting cost pennies per interaction, while a nursing assistant's loaded wage is $25–35/hour. Even with oversight and integration costs, AI is orders of magnitude cheaper per handled inquiry.
Cost vs. human wageclaude-sonnet-53/5Automated directory/hours systems are cheap to run, but integrating with EHR for status updates with proper safeguards adds cost, making overall cost roughly comparable once compliance is factored in.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed chatbots and IVR systems (e.g., hospital appointment and information lines) already handle visitor inquiries and basic directional/procedural questions in production. However, error rates on complex multi-part queries and integration gaps with real-time patient data systems prevent a full 5 rating.
Technical feasibility todayclaude-sonnet-52/5Hospitals deploy basic wayfinding kiosks and general FAQ bots, but no widespread deployed product reliably provides patient-status updates to visitors/callers due to privacy and accuracy concerns.

Remind patients to take medications or nutritional supplements.

44

CI 39–50 · exposure 39 · augmentation 75 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Healthcare settings have adopted automated reminder systems in pilots and some production environments, but uptake remains uneven; many facilities still rely on staff-delivered reminders due to liability concerns and integration challenges with existing workflows.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially direct patient care settings employing nursing assistants, is a slower-adopting sector for AI-driven task automation due to physical care requirements and regulatory caution.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-driven reminder systems substantially assist nursing assistants by automating scheduling, tracking adherence, and flagging non-compliant patients, allowing staff to focus on patients who need intervention rather than blanket reminders.
Augmentation potentialclaude-sonnet-54/5Smart reminder systems, alerts, and scheduling tools meaningfully help nursing assistants track and prompt multiple patients' medication schedules, improving efficiency while the human remains responsible for delivery and verification.
Task automatabilityclaude-haiku-4-5-202510012/5Sending automated reminders is straightforward, but nursing assistants must often verify patient understanding, adjust timing for individual needs, and handle refusals or complications—tasks requiring human judgment and presence that current AI cannot reliably manage end-to-end.
Task automatabilityclaude-sonnet-52/5Physical presence, verification of ingestion, and interpersonal reassurance are core to this task; automated reminder systems can prompt but cannot perform the hands-on verification and assistance nursing assistants provide.
Adoption barriersclaude-haiku-4-5-202510013/5Regulatory oversight exists around medication administration accuracy and patient safety, and many healthcare facilities prefer human contact for medication management and patient education, creating moderate friction against full automation.
Adoption barriersclaude-sonnet-53/5No licensing barrier prevents automated reminders, but liability concerns around missed doses, adverse events, and patient safety create moderate organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510014/5Automated reminder systems cost pennies per patient per day, while nursing assistant labor costs $15–25/hour; the cost differential strongly favors automation for reminder delivery alone.
Cost vs. human wageclaude-sonnet-53/5Reminder software/devices are cheap per alert, but they don't cover the full task (confirming compliance, physical assistance), so cost comparison to a human performing the complete task is roughly comparable once integration and oversight are included.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed reminder systems (SMS, app-based, smart pill dispensers) work reliably in production for straightforward medication schedules, though integration with patient response verification and clinical workflows remains inconsistent across organizations.
Technical feasibility todayclaude-sonnet-53/5Automated medication reminder apps and devices (pill dispensers, smart alerts) are deployed in some care settings, but they operate alongside human staff rather than replacing the verification and physical assistance role.

Stock or issue medical supplies, such as dressing packs or treatment trays.

33

CI 30–35 · exposure 25 · augmentation 50 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare is moderately digitized, but adoption of autonomous supply-chain robotics remains limited outside large hospital systems; most healthcare settings continue manual or semi-automated stocking practices.
Sector adoption velocityclaude-sonnet-52/5Healthcare facility logistics are a laggard sector for AI/robotic adoption in physical inventory tasks, with automation more common in large hospital systems but not widespread among nursing assistants generally.
Augmentation potentialclaude-haiku-4-5-202510013/5Inventory tracking software and demand-forecasting tools meaningfully assist nursing assistants by highlighting low-stock items and optimizing supply placement, though the core physical task of stocking remains largely human-driven.
Augmentation potentialclaude-sonnet-53/5AI-driven inventory management systems can alert staff to low stock, predict usage patterns, and streamline reordering, meaningfully assisting but not replacing the physical task.
Task automatabilityclaude-haiku-4-5-202510012/5Stocking and issuing supplies involves inventory management and retrieval, which could be partially automated with robotic systems or inventory tracking software. However, the physical manipulation, human judgment about supply levels, and real-time responsiveness to clinical demands create significant automation hurdles that prevent end-to-end performance at 50% time savings today.
Task automatabilityclaude-sonnet-52/5This requires physical manipulation of supplies in a real clinical space, which current AI systems cannot perform end-to-end; only inventory-tracking software components could be automated.:
Adoption barriersclaude-haiku-4-5-202510013/5Healthcare supply chains face regulatory oversight and infection-control requirements that add friction to automation. Hospital operations also value human oversight and flexibility in supply distribution, though no strict legal barrier prevents mechanization of routine stocking.
Adoption barriersclaude-sonnet-52/5No licensure is strictly required for this specific stocking task, though facility protocols and inventory control policies create some procedural friction and accountability requirements.
Cost vs. human wageclaude-haiku-4-5-202510012/5Implementing automated inventory and robotic stocking systems requires substantial capital investment (hardware, software, integration), while nursing assistant labor is relatively low-cost. The all-in automation cost typically exceeds manual labor for routine supply stocking in most settings.
Cost vs. human wageclaude-sonnet-52/5Robotic/automated dispensing systems require significant capital investment and integration, and are not yet cheaper than having a nursing assistant handle routine stocking.
Technical feasibility todayclaude-haiku-4-5-202510012/5Inventory management software exists and is widely deployed, but fully autonomous supply stocking in clinical environments—requiring physical manipulation, location awareness, and integration with hospital workflows—remains largely manual or in early pilot stages, not production-grade automation.
Technical feasibility todayclaude-sonnet-52/5Automated inventory/RFID and smart cabinet systems exist for supply tracking, but the physical stocking and issuing of items to point-of-care still relies on staff or robotic carts in limited pilot deployments.

Record height or weight of patients.

31

CI 23–40 · exposure 30 · augmentation 38 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of AI for routine clinical tasks remains slow and cautious. Nursing assistant measurement recording is a low-priority automation target in production systems, and most facilities rely on direct human measurement and manual or simple electronic logging.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially long-term care and nursing assistant settings, is a historically slow adopter of automation technology relative to information-sector industries.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by auto-populating electronic health records from sensor data or images, reducing transcription time. However, the core task—accurate measurement—remains entirely human-dependent, limiting the augmentation potential to modest data-entry assistance.
Augmentation potentialclaude-sonnet-53/5Digital scales and EHR auto-population can speed up the recording step and reduce transcription errors, meaningfully aiding the human worker on part of the task.
Task automatabilityclaude-haiku-4-5-202510012/5Recording height or weight requires physical measurement and data entry, but the physical measurement itself cannot be automated without specialized equipment (scales, height measures) already in place. AI can process and log data once obtained, but cannot autonomously perform the in-person measurement and documentation required in most clinical settings.
Task automatabilityclaude-sonnet-52/5The physical act of measuring a patient's height/weight and positioning them cannot be done by AI; only the recording portion could be automated with smart scales or voice input, so full task automation is limited.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare settings have regulatory requirements for documented patient measurements performed by licensed personnel (nursing assistants or nurses); liability and patient safety concerns around measurement errors create strong friction. Chain-of-custody and verification requirements in medical records also limit pure automation.
Adoption barriersclaude-sonnet-52/5No licensing requirement specifically restricts recording vital measurements, though clinical protocols and liability for accurate patient records create some institutional friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5The labor cost of a nursing assistant performing this simple task is low (typically $15–25/hour loaded). AI inference and integration overhead, combined with the need for human oversight of measurement accuracy, does not achieve a significant cost advantage over direct human performance.
Cost vs. human wageclaude-sonnet-52/5Automated measurement devices require upfront hardware investment and integration, and a human aide is still needed for patient handling, so cost savings are modest rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI systems can extract and record numerical data from images or structured input, no deployed product reliably performs the full task of physically measuring patients and recording results end-to-end. Current systems lack the embodied capability to interact with patients and measurement devices in real clinical environments.
Technical feasibility todayclaude-sonnet-53/5Smart scales, stadiometers, and EHR-integrated devices that auto-record measurements exist and are used in some clinics, but many facilities still rely on manual measurement and manual data entry.

Review patients' dietary restrictions, food allergies, and preferences to ensure patient receives appropriate diet.

30

CI 25–35 · exposure 30 · augmentation 50 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While healthcare is digitizing, actual AI-driven automation of dietary-restriction verification remains in pilot phase at most institutions. Most hospitals still rely on nursing staff manually cross-checking charts, patient communication, and food-service tickets rather than autonomous AI systems.
Sector adoption velocityclaude-sonnet-52/5Healthcare support roles adopt digital record systems steadily but physical, hands-on care tasks in this sector show slow AI-driven displacement.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist nursing assistants by auto-populating dietary flags from EHR records, highlighting potential conflicts, and flagging newly documented allergies, reducing manual chart review time while the assistant retains responsibility for final verification and patient communication.
Augmentation potentialclaude-sonnet-53/5EHR alerts and dietary databases can meaningfully assist nursing assistants in cross-referencing restrictions, reducing error risk while the human remains responsible for verification and delivery.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can analyze dietary restrictions and allergies from medical records with high accuracy, the task requires judgment about individual patient needs, preferences, and medical context that varies case-by-case. Current systems cannot reliably handle the nuanced coordination with kitchen staff, physician orders, and real-time preference changes without substantial human oversight.
Task automatabilityclaude-sonnet-52/5AI can cross-check dietary restriction data against menus, but confirming patient identity, verbal preferences, and real-time bedside delivery still requires physical human presence and judgment.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare regulations, patient safety liability, and the legal requirement for a licensed provider to verify and sign off on dietary orders create substantial barriers. Dietary errors can cause serious harm (anaphylaxis, aspiration), so automation faces high compliance and insurance liability friction.
Adoption barriersclaude-sonnet-53/5No licensing requirement specifically for this, but patient safety liability for allergy errors and requirement for direct patient interaction create meaningful friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5The integration cost and oversight required to deploy AI dietary-check systems across healthcare settings approaches or exceeds the hourly wage of nursing assistants, especially when liability and error-checking overhead are factored in.
Cost vs. human wageclaude-sonnet-52/5Software flagging can be cheap, but the physical verification and delivery portion still requires paid human labor, keeping overall cost comparable to human-only workflows.
Technical feasibility todayclaude-haiku-4-5-202510013/5Clinical decision-support systems and EMR integrations can flag dietary restrictions and allergies in deployed hospital systems, but they often produce false positives or miss context-specific exceptions. No mature AI product fully replaces the nursing assistant's role of verifying restrictions and ensuring appropriate meals without human verification.
Technical feasibility todayclaude-sonnet-52/5Some EHR-integrated dietary tracking systems flag allergies automatically, but no deployed product independently reviews and ensures correct diet delivery to a specific patient without human execution.

Record vital signs, such as temperature, blood pressure, pulse, or respiration rate, as directed by medical or nursing staff.

30

CI 25–35 · exposure 30 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While healthcare is digitizing, vital sign measurement automation remains limited to device-to-EHR integration, not autonomous measurement. Nursing assistant roles remain largely manual and labor-intensive in most care settings, with slow meaningful displacement.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially direct patient care settings, has been slower than other sectors to adopt AI/automation for hands-on tasks, though remote monitoring tech is growing in specific contexts like ICUs or home health.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-assisted vital sign monitoring—automated device integration, real-time alerts for abnormal readings, and EHR auto-population—meaningfully augment nursing assistant productivity and decision support, allowing them to focus on patient care and exception handling rather than manual documentation.
Augmentation potentialclaude-sonnet-53/5AI-enabled devices and EHR systems can streamline recording and flag abnormal readings, helping assistants document faster and catch issues, but the core measurement still requires human action.
Task automatabilityclaude-haiku-4-5-202510012/5While vital sign *recording* (data entry) can be partially automated via integration with monitoring devices, the actual *measurement* of vital signs requires direct patient contact and skilled technique. Current AI cannot independently perform the full end-to-end task of approaching a patient, taking accurate measurements, and recording them without human supervision.
Task automatabilityclaude-sonnet-52/5Recording vitals requires physical presence to measure or read a patient's temperature, pulse, BP, or respiration; AI cannot perform the hands-on measurement, though it could log/transcribe data from connected devices. Most of the physical task cannot be automated end-to-end today.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare regulation, patient safety liability, and the requirement for a licensed clinician to oversee and validate vital sign measurements create substantial barriers. Direct patient contact and the legal responsibility of nursing staff to ensure accuracy impose hard friction against full automation.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement for this specific task, but patient-contact norms, liability for missed/inaccurate readings, and integration into clinical workflows create meaningful friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5Monitoring equipment and EHR integration add capital and integration costs. The labor savings are modest because human staff still position devices, verify readings, and document; the human wage for this task remains a significant cost driver.
Cost vs. human wageclaude-sonnet-52/5Connected monitoring hardware and software have upfront and maintenance costs that, for routine spot-checks, are not clearly cheaper than a nursing assistant's time, especially in lower-resource facilities.
Technical feasibility todayclaude-haiku-4-5-202510013/5Automated vital sign monitoring devices (pulse oximeters, blood pressure cuffs, thermometers) exist and integrate with EHRs, but these are narrowly scoped tools requiring human placement and oversight. No end-to-end AI system reliably measures and records all vital signs autonomously in production healthcare settings.
Technical feasibility todayclaude-sonnet-52/5Automated vital-sign devices and EHR integrations exist (e.g., smart BP cuffs, wearables) but full autonomous measurement and recording without a human present is not deployed at scale in most care settings.

Measure and record food and liquid intake or urinary and fecal output, reporting changes to medical or nursing staff.

28

CI 25–30 · exposure 25 · augmentation 50 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare has been slow to adopt AI-driven automation of bedside observation tasks; most intake/output tracking remains manual or semi-manual within existing EHR systems. Adoption is well behind finance or tech sectors, with pilots limited to data entry augmentation rather than autonomous measurement.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially long-term care and nursing assistant roles, is a slower-adopting sector for AI-driven physical monitoring, though EHR digitization is increasing gradually.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by auto-populating charts from manually recorded measurements, flagging outliers, and summarizing trends for clinical review. However, the assistance is modest—the human still does the core measurement and initial observation; AI primarily streamlines documentation and analysis.
Augmentation potentialclaude-sonnet-53/5AI-enabled charting, voice-to-text notes, and alert systems for abnormal trends can meaningfully help nursing assistants document and flag changes faster, even though physical measurement remains manual.
Task automatabilityclaude-haiku-4-5-202510012/5While intake/output recording involves structured data entry that AI could theoretically assist with, the critical measurement and observation components require direct physical presence and human judgment to detect abnormalities. Current AI systems cannot autonomously measure bodily outputs or reliably infer changes without structured data input from a human observer.
Task automatabilityclaude-sonnet-52/5Recording data is simple but requires physical presence to observe, measure fluids, and interact with patients; AI can log/aggregate data but cannot physically measure intake/output itself.
Adoption barriersclaude-haiku-4-5-202510014/5Clinical settings have strong regulatory and liability requirements; documentation of intake/output is part of the legal medical record and typically requires attestation by licensed staff. Hospitals have established workflows and nursing protocols that embed human accountability, creating organizational and legal friction against full automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement for this specific task, but patient safety concerns, liability for missed changes, and facility protocols create moderate friction against pure automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5The task involves low-wage manual work (nursing assistant salary ~$30–35k/year). AI overhead for vision systems, integration into EHRs, and compliance checking would likely exceed the cost of direct human labor for this straightforward recording function.
Cost vs. human wageclaude-sonnet-52/5Physical measurement still requires human labor or specialized sensor hardware, so AI-only solutions don't yet undercut the low-wage human cost by an order of magnitude when hardware and integration costs are included.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed AI system autonomously measures and records intake/output in clinical settings today. Charting assistants exist but still require human nurses to perform the actual measurement and observation; AI plays only a documentation role, not the core task.
Technical feasibility todayclaude-sonnet-52/5Some EHR and smart-monitoring devices exist to digitize intake/output tracking, but widespread reliable automated measurement without human observation is not standard in most care settings.

Document or otherwise report observations of patient behavior, complaints, or physical symptoms to nurses.

28

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare remains a laggard in clinical automation adoption despite digital infrastructure; pilot projects exist for documentation support but production-scale replacement of observational tasks is minimal. Regulatory burden and conservative risk culture slow deployment significantly.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially long-term and direct care settings, has historically been slow to adopt AI tools relative to information/professional service sectors.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-assisted documentation tools (voice transcription, symptom prompts, template suggestions) substantially improve nursing assistant efficiency and note quality when integrated into workflows. These systems enhance productivity while keeping the human observer and validator in the loop.
Augmentation potentialclaude-sonnet-53/5Speech-to-text and structured note-taking tools can meaningfully speed up documentation, though the underlying observation and clinical judgment remain human-driven.
Task automatabilityclaude-haiku-4-5-202510012/5While AI systems can process and categorize basic symptom descriptions from patient input, the task requires real-time observation of subtle behavioral and physical cues that current systems cannot reliably capture end-to-end. Documentation of complex, context-dependent patient states still requires human oversight to meet clinical standards.
Task automatabilityclaude-sonnet-52/5The physical observation and interpretation of patient behavior requires human presence and judgment; only the documentation/transcription portion could be AI-assisted, so full end-to-end automation is not achievable today.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory and legal barriers are substantial: clinical documentation is subject to HIPAA, accreditation standards, and liability requirements that typically mandate human accountability. Healthcare organizations also maintain strong preferences for direct human observation in care settings, and nursing staff are required to sign off on patient care records.
Adoption barriersclaude-sonnet-53/5No licensing requirement for documentation itself, but patient safety concerns, accuracy requirements for medical records, and reliance on direct human observation create meaningful friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5The all-in cost of AI observation systems (hardware, integration, continuous retraining, required human oversight) remains comparable to or exceeds the loaded wage of nursing assistants, especially when accounting for liability and error-correction overhead.
Cost vs. human wageclaude-sonnet-52/5AI documentation tools require devices, integration, and oversight; nursing assistants are relatively low-wage workers, so cost savings from AI transcription are modest rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510012/5Existing AI products can assist with documentation (e.g., speech-to-text, template-based note generation), but no production system reliably performs the full task of independently observing patients and generating clinically valid reports without human validation. Deployed systems remain in support roles, not autonomous operation.
Technical feasibility todayclaude-sonnet-52/5Voice-to-text and ambient documentation tools exist in some clinical settings but are not yet reliably deployed for CNA-level observational reporting at scale.

Explain medical instructions to patients or family members.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare systems have been slow to deploy AI for direct patient communication; most pilots remain experimental or supplementary rather than replacing human contact. Adoption is constrained by regulatory caution, liability concerns, and institutional preference for human accountability in clinical instruction.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially direct patient care roles like nursing assistants, has been slow to adopt AI for interpersonal communication tasks compared to administrative or diagnostic support functions.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist nursing assistants by drafting clear written explanations, translating medical jargon, or generating multilingual summaries that the human can then tailor and deliver to the patient. This supportive role improves clarity and consistency without removing the human judgment required for safe, empathetic communication.
Augmentation potentialclaude-sonnet-54/5AI can generate patient-friendly explanations, translate medical jargon, and provide multilingual materials that nursing assistants use to prepare and support their communication with patients and families.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can generate explanations of medical instructions, the task requires contextual adaptation to patient literacy, emotional state, and cultural background—factors that demand real-time judgment. Current AI systems lack reliable capacity to assess comprehension, handle follow-up questions, or build the trust needed for medical compliance, making full end-to-end automation with 50% time savings unlikely.
Task automatabilityclaude-sonnet-52/5Explaining instructions requires real-time reading of patient comprehension, emotional state, and follow-up questioning that current AI cannot reliably handle in-person, though AI can draft explanatory content in advance.tools.rated conservatively
Adoption barriersclaude-haiku-4-5-202510014/5Medical explanation carries legal and liability weight: failure to communicate correctly can result in patient harm, non-compliance, and litigation. Regulatory frameworks, institutional risk management policies, and patient expectations for human contact create strong organizational and legal barriers to full automation.
Adoption barriersclaude-sonnet-54/5Patient safety, liability for miscommunication of medical instructions, and regulatory/accreditation expectations for direct patient communication create strong barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI systems capable of medical explanation require specialized training data, compliance infrastructure, and continuous human oversight to prevent harm; integration and liability management costs are substantial. The loaded wage of a nursing assistant remains competitive with the full system cost, especially given legal and safety requirements.
Cost vs. human wageclaude-sonnet-52/5While generating text is cheap, the task requires human presence and interaction, so AI cannot fully substitute for the labor cost of an in-person explanation and check for understanding.
Technical feasibility todayclaude-haiku-4-5-202510012/5Deployed products can generate template explanations of medical procedures, but healthcare organizations rarely substitute AI for live explanation due to liability concerns, regulatory requirements for human verification, and the critical importance of immediate clarification of misunderstandings. Pilot systems exist but lack production-scale deployment in patient-facing settings.
Technical feasibility todayclaude-sonnet-52/5Chatbots and patient-education apps exist but are not deployed to replace in-person explanation by nursing staff at the bedside; adoption is limited to supplementary materials.

Gather information from caregivers, nurses, or physicians about patient condition, treatment plans, or appropriate activities.

23

CI 20–25 · exposure 20 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI for core clinical tasks remains slow and cautious. Nursing assistants work in regulated, risk-averse settings (hospitals, long-term care) where automation of patient information gathering is not yet a production trend. Pilot projects exist but far from widespread deployment.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially direct patient care roles like nursing assistants, has historically slow AI adoption due to regulatory, safety, and workflow constraints.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could assist a nursing assistant by pre-drafting summaries of existing notes, flagging potential follow-up questions, or organizing information templates, but the core interpersonal and clinical reasoning work requires human judgment. Moderate augmentation potential through decision support rather than autonomous execution.
Augmentation potentialclaude-sonnet-53/5AI-powered note-taking, EHR summarization, and voice-to-text tools can help capture and organize information gathered during these interactions, improving efficiency without replacing the human exchange.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could transcribe or parse existing notes, nursing assistants must actively elicit nuanced information from clinical staff about patient state, medication changes, and individualized care plans. The task requires real-time dialogue, contextual judgment about what to ask, and synthesis of oral communication that goes beyond template-based extraction. Current AI cannot reliably conduct these adaptive conversations with clinical accuracy.
Task automatabilityclaude-sonnet-52/5This requires real-time verbal communication, physical presence, and building rapport with care team members and patients, which current AI cannot fully replicate end-to-end in a clinical setting.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare is heavily regulated; patient safety, liability, and malpractice concerns create strong disincentives to automating information gathering from clinical staff. Institutional policies, physician and nurse oversight requirements, and accreditation standards expect human accountability in clinical communication chains.
Adoption barriersclaude-sonnet-54/5Patient care coordination involves compliance, safety, and accountability requirements where human staff must verify and act on information; liability concerns and human-contact norms create strong barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI systems capable of gathering clinical information through dialogue would require significant integration, oversight, and human verification to ensure safety and accuracy. The loaded cost of a nursing assistant remains low relative to specialized clinical AI infrastructure and the liability exposure of automated information gathering errors.
Cost vs. human wageclaude-sonnet-52/5While transcription or note-summarization tools are cheap, the actual information-gathering interaction with humans still requires a present, trained nursing assistant, so cost savings are minimal for the full task.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs this task independently. AI chatbots can simulate information retrieval but lack integration into actual clinical workflows, real-time communication with human staff, and the ability to validate clinical accuracy. Healthcare remains heavily reliant on human-to-human handoff communication.
Technical feasibility todayclaude-sonnet-51/5No deployed product independently gathers this kind of contextual, multi-source clinical information from human staff in live care settings; this remains a human interpersonal task.

Prepare or serve food trays.

19

CI 5–33 · exposure 13 · augmentation 13 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a laggard sector for physical automation due to regulatory burden, infection-control protocols, and the fragmented nature of hospital operations. Nursing-assistant roles show minimal AI/robotic displacement; adoption is measured in pilots, not production at scale.
Sector adoption velocityclaude-sonnet-51/5Healthcare support/physical care roles are among the slowest sectors to adopt AI or robotics for hands-on tasks, with automation limited to logistics/delivery robots rather than the human-contact serving step.
Augmentation potentialclaude-haiku-4-5-202510012/5AI offers minimal direct assistance to a nursing assistant preparing trays. Dietary-management software and inventory systems may help planning, but the core task—physically assembling and serving food—sees little productivity gain from current AI tools in the hands of the worker.
Augmentation potentialclaude-sonnet-51/5AI provides essentially no direct assistance to a human performing the physical act of preparing or serving a food tray.
Task automatabilityclaude-haiku-4-5-202510012/5Food tray preparation involves physical manipulation of trays, dishes, and food items in varied layouts and sequences. While robotic systems could theoretically handle standardized tray assembly in controlled settings, current general-purpose AI/robotics lack the dexterity and adaptability to handle the full task (responding to dietary restrictions, irregular tray configurations, temperature management) at 50% time savings without significant setup.
Task automatabilityclaude-sonnet-51/5Preparing and physically serving food trays to patients requires manual dexterity, mobility, and physical presence that current AI systems cannot perform; this is a physical-world manipulation and delivery task, not a cognitive/informational one.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare facilities face strong regulatory and liability requirements around food safety, sanitation, and traceability. Hospitals must verify proper handling, temperature control, and dietary compliance; human accountability and sign-off are often mandated, creating legal and operational barriers to full automation.
Adoption barriersclaude-sonnet-52/5No licensing requirement specifically for this task, but patient contact, safety checks (e.g., dietary restrictions, choking risk, allergies) and hospital workflow integration create moderate organizational friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Robotic food-prep and serving systems remain capital-intensive and require installation, maintenance, and supervision. The loaded wage for a nursing assistant ($30–40k/year) remains competitive to or cheaper than the per-task amortized cost of robotics plus human oversight in most healthcare settings.
Cost vs. human wageclaude-sonnet-51/5Robotic or automated tray delivery systems require expensive hardware, infrastructure, and maintenance that vastly exceed the low hourly cost of a nursing assistant performing this simple manual task.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some specialized food-service robotics exist in limited hospital deployments (e.g., tray delivery carts), but end-to-end preparation and serving remain largely manual. Deployed solutions handle only narrow subtasks (delivery, not preparation), with significant organizational friction and material gaps in real-world healthcare operations.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs tray preparation or bedside delivery to patients; robotic food service remains research/pilot stage in a few hospitals, not standard practice.

Clean and sanitize patient rooms, bathrooms, examination rooms, or other patient areas.

19

CI 5–33 · exposure 13 · augmentation 13 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare cleaning remains heavily manual and labor-intensive; adoption of robotic cleaning is minimal and mostly confined to pilot projects in large academic medical centers. The sector's fragmentation, regulatory conservatism, and infection-control liability have slowed meaningful displacement.
Sector adoption velocityclaude-sonnet-51/5Healthcare support and janitorial functions are among the least digitized, physically-dependent sectors with minimal AI/robotic adoption for this specific task type.
Augmentation potentialclaude-haiku-4-5-202510012/5AI tools offer minimal assistance for the core task; computerized scheduling and inventory management provide marginal support, but no AI system meaningfully augments the hands-on cleaning and sanitization work itself that defines this role.
Augmentation potentialclaude-sonnet-51/5Current AI offers essentially no meaningful assistance for the physical act of cleaning and sanitizing patient care areas.
Task automatabilityclaude-haiku-4-5-202510012/5Physical cleaning and sanitization requires dexterous manipulation of cleaning tools, maneuvering around varied furniture layouts, and detecting soiled areas—tasks that current robotics cannot reliably perform end-to-end in unstructured healthcare environments. While prototype cleaning robots exist, they cannot yet match the speed and thoroughness of human workers at scale.
Task automatabilityclaude-sonnet-51/5This is a physical cleaning and sanitizing task requiring manipulation of objects, surfaces, and equipment in varied environments, which current AI systems cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare environments impose strict infection-control standards, regulatory compliance requirements, and liability concerns around inadequate sanitization. Institutions face pressure to maintain human oversight of cleanliness for patient safety and regulatory certification, creating both legal and organizational friction against full automation.
Adoption barriersclaude-sonnet-52/5No licensing requirement specifically governs cleaning tasks, but infection-control protocols, liability for improper sanitization in clinical settings, and physical workspace constraints create moderate practical barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current cleaning robots and systems are capital-intensive ($50k–$500k per unit) with limited throughput; per-task cost remains higher than employing nursing assistants at typical wage rates when amortization, maintenance, and human supervision are included.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI system for this physical task, so any hypothetical robotic solution would require expensive specialized hardware far exceeding the cost of human labor for this task.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some autonomous floor-cleaning and UV disinfection robots are deployed in limited hospital settings, but they handle only floor/large-surface work and cannot perform comprehensive room cleaning, toilet cleaning, or adaptive response to different room types. Deployment remains sparse and requires significant human oversight.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs general patient-room cleaning and sanitizing; commercial cleaning robots exist for narrow floor-cleaning tasks but not comprehensive room sanitization including bathrooms and equipment.

Restock patient rooms with personal hygiene items, such as towels, washcloths, soap, or toilet paper.

16

CI 5–28 · exposure 8 · augmentation 13 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare, particularly nursing roles, is a laggard sector for physical task automation. Adoption remains minimal; most facilities continue with manual restocking and have shown little real-world deployment of robots for this function despite decades of robotics research.
Sector adoption velocityclaude-sonnet-51/5Healthcare support and physical labor tasks in nursing settings show very low AI/robotics adoption for routine physical logistics work like this.
Augmentation potentialclaude-haiku-4-5-202510012/5AI offers minimal augmentation for restocking—the task is largely mechanical and routine. Predictive inventory systems or route optimization could provide modest assistance, but no AI system meaningfully enhances a nursing assistant's productivity while performing this physical task.
Augmentation potentialclaude-sonnet-51/5AI offers essentially no meaningful assistance for this manual restocking task, as it involves no cognitive or informational component that generative AI tools could enhance.
Task automatabilityclaude-haiku-4-5-202510012/5This task involves physical manipulation and placement of items in patient rooms, requiring dexterous robotic systems and real-time navigation in varied environments. Current mobile manipulation robots struggle with the variability of patient room layouts and the need to handle diverse soft goods (towels, washcloths) reliably, making end-to-end automation with 50% time savings infeasible today.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring navigating patient rooms, handling objects, and restocking supplies, which current AI systems cannot perform end-to-end without robotic embodiment far beyond deployed capabilities.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare facilities have strong preferences for human staff presence in patient areas for safety, infection control oversight, and regulatory compliance. Hospitals also face liability concerns around autonomous systems in patient care environments, and union/labor agreements often protect human positions.
Adoption barriersclaude-sonnet-52/5No licensing or regulatory requirement mandates a human for restocking supplies, but practical barriers around hospital logistics, hygiene protocols, and physical environment complexity limit automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current robotic systems capable of this task (if they existed at production quality) would require significant capital investment, integration, and maintenance costs that far exceed the modest hourly wage of nursing assistants, making the economic case weak.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI-driven substitute for this physical task, so any automation would require expensive robotics infrastructure vastly exceeding the low wage cost of a human aide performing this simple task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed production systems reliably perform autonomous restocking of patient rooms with hygiene items. While robotic arms and mobile robots exist in research, real healthcare facilities do not rely on AI/robotic systems for this task at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously restocks patient rooms with hygiene items in production healthcare settings; this remains outside current physical AI/robotics deployment.

Transport specimens, laboratory items, or pharmacy items, ensuring proper documentation and delivery to authorized personnel.

15

CI 0–30 · exposure 13 · augmentation 25 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare is a traditionally slow-adopting sector for labor displacement; most hospitals still rely on human staff for specimen transport despite regulatory pressure to improve efficiency. Pilot programs exist but have not achieved meaningful production deployment across hospital networks.
Sector adoption velocityclaude-sonnet-52/5Healthcare facility logistics automation (robots, tube systems) is adopted unevenly and slowly, concentrated in a minority of large well-funded hospitals.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with routing optimization or documentation auto-filling, but the core task of physically transporting and securely handing off items to authorized personnel offers limited augmentation potential since a human must remain the responsible party.
Augmentation potentialclaude-sonnet-52/5AI can help with documentation, tracking, and alerts for specimen status, but offers limited assistance to the core physical transport activity.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical transportation of items in a healthcare environment with chain-of-custody integrity. Current AI systems lack mobile manipulation robots deployable at scale in hospitals, and the task involves real-world navigation, human handoffs, and compliance verification that remains beyond end-to-end automation today.
Task automatabilityclaude-sonnet-52/5Physical transport of specimens and pharmacy items requires manipulation and navigation in a healthcare facility, which current AI/robotics cannot reliably perform end-to-end without heavy infrastructure investment.4The documentation portion is automatable but is a minor part of the overall task.
Adoption barriersclaude-haiku-4-5-202510015/5Healthcare logistics are subject to strict chain-of-custody requirements, HIPAA compliance, regulatory oversight of specimen handling, and institutional liability frameworks that typically require a licensed human to authorize and sign off on delivery to authorized personnel.
Adoption barriersclaude-sonnet-53/5Chain-of-custody and proper handling of specimens/medications involves compliance and accountability requirements, though not strictly a licensed-professional-only task, creating moderate procedural friction.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current mobile robots capable of secure hospital transport (with authentication, security integration, and maintenance) cost tens of thousands of dollars per unit and require infrastructure investment, making them more expensive than a nursing assistant's hourly wages per delivery, all-in.
Cost vs. human wageclaude-sonnet-52/5Robotic transport systems have high capital costs and are only cost-effective in large hospitals with high volume; for most settings, a human aide is cheaper than automated logistics systems.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs specimen and pharmacy logistics with the required documentation, security, and regulatory compliance in production hospital workflows. Autonomous delivery robots exist in research and limited pilots but do not meet healthcare's authentication and liability standards at scale.
Technical feasibility todayclaude-sonnet-52/5Hospital delivery robots and pneumatic tube systems exist in some large facilities, but they are not general-purpose AI and require significant infrastructure; most nursing assistants still perform this manually.

Observe or examine patients to detect symptoms that may require medical attention, such as bruises, open wounds, or blood in urine.

13

CI 0–25 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of autonomous patient monitoring remains very limited; the sector is risk-averse, heavily regulated, and dependent on licensed human judgment. Current deployments are supplementary (monitoring devices with alarms) rather than substituting for direct nursing observation.
Sector adoption velocityclaude-sonnet-52/5Healthcare direct-care settings (nursing homes, hospitals) show slow AI adoption for hands-on physical tasks, though administrative and documentation aspects of nursing are seeing more tools introduced.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted wound imaging, automated vital sign alerts, and visual screening tools can augment a nursing assistant's efficiency in detecting certain symptoms, but the human remains in the loop for all clinical interpretation and escalation decisions.
Augmentation potentialclaude-sonnet-52/5AI can assist with documentation, flagging patterns from recorded vitals, or triage support, but offers minimal help with the actual physical observation and examination process itself.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can analyze visual data (bruises, wounds) from images with moderate accuracy, real-time bedside observation of patients requires continuous physical presence, contextual judgment about symptom severity, and immediate escalation decisions that current AI cannot reliably handle end-to-end. The task involves subtle clinical triage that AI cannot yet replicate at the ≥50% time-saving threshold.
Task automatabilityclaude-sonnet-51/5This requires hands-on physical examination, direct observation of the patient's body, and tactile/visual assessment in real time, which current AI cannot perform without embodiment.
Adoption barriersclaude-haiku-4-5-202510014/5Legal and regulatory frameworks require licensed clinical staff (RNs, nursing assistants) to perform patient assessment and symptom detection. Patient safety liability, the need for human judgment in escalation, and mandatory human oversight create substantial organizational and legal barriers to automation.
Adoption barriersclaude-sonnet-55/5Direct patient examination for medical symptoms is a core clinical duty requiring licensed/certified human personnel, with strict regulatory and liability requirements around hands-on care and reporting.
Cost vs. human wageclaude-haiku-4-5-202510012/5Integration of vision systems for bedside monitoring, combined with the high cost of false negatives (clinical liability), makes the all-in cost per task comparable to or potentially exceeding the wage of a nursing assistant who performs multiple overlapping duties.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical examination task, so cost comparison favors the human by default since AI cannot deliver the output at all.
Technical feasibility todayclaude-haiku-4-5-202510012/5Computer vision systems can detect certain visual abnormalities in controlled settings (dermatology, wound imaging), but deployed products lack the reliability and scope needed for continuous patient monitoring in dynamic hospital environments. No mature production system performs this core nursing function independently.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs physical patient examination for symptom detection; this remains a human physical-presence task with no robotic or sensor system in production use for this purpose.

Communicate with patients to ascertain feelings or need for assistance or social and emotional support.

9

CI 7–11 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI for direct patient communication remains cautious and limited; most deployments are supervisory or supplementary, not substitutive, and institutional risk-aversion around patient-facing emotional tasks slows adoption.
Sector adoption velocityclaude-sonnet-52/5Healthcare support occupations, especially direct patient care roles, show slow AI adoption due to the hands-on, relational nature of the work and regulatory caution.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist by flagging keywords or prompting a nursing assistant to ask follow-up questions, but current systems add little value to the core skill of listening, interpreting emotional cues, and responding authentically.
Augmentation potentialclaude-sonnet-52/5AI tools like sentiment-analysis apps or communication aids offer marginal support (e.g., translation, symptom-tracking prompts) but do not meaningfully transform the core interpersonal task.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires authentic human-to-human interaction to detect nuanced emotional states, build trust, and provide genuine social and emotional support—capabilities that current AI systems cannot reliably replicate in clinical settings. While AI can simulate conversation, it cannot provide the genuine empathy, presence, and adaptive emotional attunement that patients require.
Task automatabilityclaude-sonnet-51/5This requires physical presence, empathetic human interaction, and real-time responsiveness to bedridden or vulnerable patients that current AI cannot replicate end-to-end.dedicated
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: healthcare regulations require licensed personnel to assess patient mental/emotional status; patient safety liability is high if emotional distress is missed; and institutional liability for delegating sensitive patient communication to unaccountable systems creates legal and compliance friction.
Adoption barriersclaude-sonnet-54/5Direct patient care involving physical presence and emotional support is generally protected by facility staffing requirements, liability concerns, and patient preference for human contact, though not always formally licensed.
Cost vs. human wageclaude-haiku-4-5-202510012/5While AI inference is cheap, the human oversight, liability, and need for human backup when emotional crises emerge mean the all-in cost is comparable to or higher than a nursing assistant performing the task directly.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this in-person emotional/physical care task, so cost comparison favors the human by default.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs emotional assessment and genuine emotional support communication in clinical environments today; chatbots exist but are typically used for symptom screening or triage, not for the therapeutic communication this task requires.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously ascertains patient feelings or emotional needs and provides in-person support; this remains firmly in the domain of human caregivers.

Assist nurses or physicians in the operation of medical equipment or provision of patient care.

8

CI 0–16 · exposure 13 · augmentation 25 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare has historically lagged in automation of direct care roles due to regulatory, liability, and relational barriers. Adoption of AI/robotics for nursing assistant tasks remains experimental and marginal; no measurable production displacement has occurred.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially direct bedside nursing support, is a slow-adopting sector for AI-driven task replacement due to physical, regulatory, and safety constraints, though software-based clinical documentation tools are advancing elsewhere.
Augmentation potentialclaude-haiku-4-5-202510012/5AI can offer modest assistance (alerts on patient vital changes, scheduling reminders, documentation help), but patient care itself—the core of this task—offers little room for AI augmentation; human judgment and physical presence dominate the work.
Augmentation potentialclaude-sonnet-52/5AI may assist indirectly through smart monitoring alerts or equipment diagnostics, but it offers limited direct augmentation to the hands-on assistance component of this task.
Task automatabilityclaude-haiku-4-5-202510012/5Only narrow components of this task—monitoring vitals displays, alerting on thresholds—are automatable. The core responsibilities of hands-on patient care (repositioning, hygiene, comfort) and real-time equipment operation require physical presence, situational judgment, and human touch that current AI cannot provide end-to-end at 50% time savings.
Task automatabilityclaude-sonnet-51/5This task requires hands-on physical assistance with patients and equipment in real time, which current AI systems cannot physically perform; no end-to-end automation is feasible today.
Adoption barriersclaude-haiku-4-5-202510015/5Healthcare regulation, patient safety liability, physical contact requirements, and organizational/clinical governance create hard barriers. Patient care tasks generally require a credentialed human present; substitution faces legal, accreditation, and malpractice barriers that are not easily circumvented.
Adoption barriersclaude-sonnet-55/5Direct patient care involves licensing, liability, physical safety, and legal requirements for human presence and judgment, creating hard regulatory and safety barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The loaded cost of a nursing assistant (wages, benefits, training) is far lower than the capital, integration, maintenance, and liability costs of any robotics or AI system capable of safe patient handling and personal care.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical task, so the comparison defaults to the human being the only viable option, making AI effectively more costly (infinite) for full task substitution.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed AI system reliably performs nursing assistant duties in production. While monitoring and alert systems exist, actual patient care—bathing, mobility assistance, catheter care—remains entirely human-performed; no commercial product substitutes for these tasks at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical patient care assistance or hands-on equipment operation support; this remains firmly in the physical/robotics research stage.

Change bed linens or make beds.

7

CI 5–10 · exposure 0 · augmentation 13 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare facilities remain low-automation laggards for routine physical care tasks; adoption of robotics for bed-making is negligible despite decades of potential. Organizational inertia, capital constraints, and integration complexity keep this largely manual.
Sector adoption velocityclaude-sonnet-51/5Healthcare support and direct patient care work, especially physical tasks like this, show minimal AI/robotics adoption in production settings today.
Augmentation potentialclaude-haiku-4-5-202510012/5Current AI and robotics offer minimal assistance to nursing assistants performing bed-making; no mainstream tool meaningfully augments human productivity on this specific task. Any assistance is marginal.
Augmentation potentialclaude-sonnet-51/5Current AI tools offer no meaningful assistance for the physical act of changing bed linens or making beds.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of fabric, navigation of confined spaces around patients, and real-time adjustment to patient conditions—capabilities current AI robotics cannot reliably perform at scale. No deployed system can fully automate bed-making end-to-end with equal quality.
Task automatabilityclaude-sonnet-51/5Bed-making requires physical dexterity, manipulation of soft materials, and often assisting or moving a patient in bed, none of which current AI or robotics can perform reliably or affordably today.atab
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist due to infection control requirements, patient safety liability, regulatory oversight of in-bed patient contact, and the institutional and legal requirement for human staff to manage patient care environments. Healthcare licensing and duty-of-care standards create high friction.
Adoption barriersclaude-sonnet-53/5While no licensing specifically restricts bed-making, direct physical patient contact and safety considerations (e.g., moving frail patients) create practical and liability-related barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current robotic systems capable of any form of bed manipulation are prohibitively expensive (hundreds of thousands of dollars) compared to the loaded labor cost of a nursing assistant performing this routine task.
Cost vs. human wageclaude-sonnet-51/5Robotic systems capable of this fine physical task do not exist commercially at scale, so any hypothetical solution would vastly exceed the cost of a human aide performing the task.
Technical feasibility todayclaude-haiku-4-5-202510011/5While experimental robotic bed-making systems exist in research, no production system is deployed in healthcare organizations to perform this task reliably. The task remains almost entirely manual in clinical practice.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs bed-linen changing in healthcare settings; this remains a physical manipulation task well beyond current robotics deployment in clinical environments.

Turn or reposition bedridden patients.

5

CI 0–10 · exposure 5 · augmentation 25 · importance 4.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare, especially long-term care and hospitals, has been slow to adopt automation for patient-facing physical tasks. Uptake of repositioning robots remains negligible; the sector remains heavily dependent on human labor for these activities.
Sector adoption velocityclaude-sonnet-51/5Direct care work in nursing/long-term care is a low-digitization, physically intensive sector with minimal AI/robotic adoption for hands-on tasks.
Augmentation potentialclaude-haiku-4-5-202510012/5While exoskeletons or lifting aids can assist nursing staff with the physical strain of repositioning, current AI systems provide minimal cognitive or decision-support augmentation for the actual repositioning task itself. The assistance is primarily mechanical, not AI-driven.
Augmentation potentialclaude-sonnet-52/5Mechanical lift assist devices and some sensor-based aids help staff reposition patients more safely, but this is more mechanical assistance than AI-driven productivity enhancement.
Task automatabilityclaude-haiku-4-5-202510011/5Turning and repositioning bedridden patients requires physical manipulation of the human body in a care context, which no current AI or robotic system can reliably perform end-to-end in typical hospital or care settings. While specialized repositioning robots exist in narrow deployment, they cannot match the adaptability, force control, and responsiveness needed for general patient care.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring lifting and repositioning fragile human bodies safely, which current AI (software or robotics) cannot perform end-to-end without human hands-on involvement.
Adoption barriersclaude-haiku-4-5-202510015/5This task involves direct patient contact and physical care requiring immediate human judgment, responsiveness to patient pain or discomfort, and liability. Healthcare regulations, patient safety standards, and the legal requirement for human accountability create hard barriers to full automation.
Adoption barriersclaude-sonnet-54/5Patient safety regulations, fall/injury liability, and hands-on care standards in nursing facilities create strong barriers to full automation of physical patient handling.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized repositioning robots that exist today cost tens of thousands of dollars with high integration and maintenance overhead, far exceeding the cost of a nursing assistant's labor for the same task.
Cost vs. human wageclaude-sonnet-51/5Robotic lifting/repositioning systems, where they exist, are expensive capital equipment still requiring staff supervision, making them costlier per task-equivalent than an aide's time.
Technical feasibility todayclaude-haiku-4-5-202510012/5A handful of experimental repositioning robots (e.g., Panasonic RIBA, some Japanese prototypes) have been piloted in controlled settings, but none operate reliably in production at scale across typical healthcare environments. Material reliability and safety concerns remain unresolved.
Technical feasibility todayclaude-sonnet-51/5No deployed commercial product autonomously repositions bedridden patients in production care settings; patient-lifting robots remain experimental or require heavy human operation.

Provide physical support to assist patients to perform daily living activities, such as getting out of bed, bathing, dressing, using the toilet, standing, walking, or exercising.

5

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a laggard sector for labor automation; nursing assistant roles are still predominantly human-performed, with minimal displacement by AI or robotics in practice despite decades of robotic development.
Sector adoption velocityclaude-sonnet-51/5Long-term care and healthcare support sectors show minimal deployment of robotic physical assistance; adoption is essentially confined to pilot studies, not production use.
Augmentation potentialclaude-haiku-4-5-202510012/5Assistive exoskeletons or lifting devices can reduce caregiver injury risk, but current AI systems offer minimal cognitive or decision-support augmentation for the core physical assistance work performed by nursing assistants.
Augmentation potentialclaude-sonnet-52/5AI can support scheduling, monitoring, or fall-detection alerts around this task, but offers little direct assistance to the hands-on physical act of helping a patient move or bathe.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of human bodies in variable environments and involves safety-critical decisions about patient mobility, strength, and dignity that cannot be reliably automated by current robots or AI systems.
Task automatabilityclaude-sonnet-51/5This requires direct physical manipulation and strength to move, lift, and stabilize human bodies, which current AI systems (software-based) cannot perform; even physical robots are not deployed for this at scale.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory and liability frameworks require trained human care workers; patient safety, infection control, and dignity concerns create high organizational and legal barriers to substitution with automation.
Adoption barriersclaude-sonnet-54/5Direct physical care involves safety, dignity, liability, and often regulatory/licensing oversight (e.g., certified nursing assistant training), plus strong patient preference for human touch and trust.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current robotic systems capable of physical assistance (exoskeletons, lifting robots) are expensive to purchase and maintain, far exceeding the cost of employing nursing assistants in most healthcare settings.
Cost vs. human wageclaude-sonnet-51/5Physical assistive robotics for bathing/toileting/transfers are far more expensive to develop, deploy, and maintain safely than employing a human aide, with no cost advantage today.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed products reliably perform full physical assistance with activities of daily living at scale; robotic exoskeletons and mobility aids exist but do not replace human assistants for the nuanced, context-dependent support this task demands.
Technical feasibility todayclaude-sonnet-51/5No commercially deployed robotic product reliably assists patients with bathing, toileting, or transferring in real care settings; such systems remain research/prototype stage.

Undress, wash, and dress patients who are unable to do so for themselves.

5

CI 5–5 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of AI remains concentrated in back-office analytics and triage; hands-on patient care remains one of the slowest sectors for physical automation due to regulatory, safety, and human-preference constraints.
Sector adoption velocityclaude-sonnet-51/5Direct care work in nursing/long-term care is a low-digitization, physically intensive sector with minimal AI/robotics adoption for hands-on tasks.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with scheduling, patient history retrieval, or care reminders, but offers minimal real-time productivity enhancement during the physical act of washing and dressing a dependent patient.
Augmentation potentialclaude-sonnet-52/5AI offers little direct assistance during the physical act itself, though adjacent tools (scheduling, care plan documentation, lift equipment guidance) provide marginal support around the task.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of a vulnerable patient's body, close tactile contact, and adaptation to individual patient needs and comfort. Current AI robots lack the dexterity, gentleness, and real-time responsiveness to safely and respectfully perform personal care tasks at human equivalence.
Task automatabilityclaude-sonnet-51/5This is a physical hands-on caregiving task requiring dexterity, mobility assistance, and delicate physical manipulation of a person's body that current AI and robotics cannot perform.
Adoption barriersclaude-haiku-4-5-202510014/5Healthcare regulations, patient dignity and consent requirements, liability for errors in intimate care, and strong institutional/patient preference for human caregivers create significant barriers to substitution. Direct physical care is inherently contact-dependent and organizationally embedded.
Adoption barriersclaude-sonnet-54/5Direct physical care involves safety, dignity, infection control, and liability concerns, and most facilities require trained/certified staff to perform hands-on patient care, creating strong institutional and regulatory friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized care robots with sufficient dexterity and safety features would cost far more to acquire, maintain, and integrate than a nursing assistant's loaded wage, with no current mature market to drive economies of scale.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute at any cost; specialized care robots (where they exist in research) are far more expensive than human aide labor for this task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed commercial product reliably performs full personal hygiene and dressing tasks for patients in clinical settings. While robotic arms exist in research, they are not in production use for this intimate, safety-critical task at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed products perform full bathing/dressing of dependent patients; robotic bathing/dressing systems remain experimental research prototypes, not production tools.

Supply, collect, or empty bedpans.

5

CI 0–10 · exposure 0 · augmentation 0 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a laggard sector for physical automation due to high liability, regulatory conservatism, infection control demands, and deep organizational commitment to human-centered care. No measurable production adoption of bedpan automation exists.
Sector adoption velocityclaude-sonnet-51/5Direct patient care and physical hygiene tasks in healthcare facilities show minimal AI or robotic adoption due to the physical, sanitary, and interpersonal nature of the work.
Augmentation potentialclaude-haiku-4-5-202510011/5Current AI systems offer minimal assistance for bedpan supply, collection, or emptying—the task is almost entirely physical and sensorimotor with no meaningful digital augmentation pathway that would enhance a human worker's performance.
Augmentation potentialclaude-sonnet-51/5AI offers no meaningful assistance for this physical, hands-on caregiving task involving direct patient contact and sanitation.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of bedpans in close proximity to patients and involves significant variability in patient mobility, positioning, and hygiene conditions that current robotics cannot reliably handle at scale. No current AI system can perform the full end-to-end task of supplying, collecting, or emptying bedpans with the dexterity, judgment, and safety margins required.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring hands-on interaction with patients and their waste; no current AI or software system can perform this end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5This task is heavily protected by regulatory requirements (patient dignity, HIPAA, infection control), liability concerns, organizational standards of care, and the strong expectation that human contact and judgment are necessary. Healthcare regulators and institutions would likely require human involvement or at minimum close oversight.
Adoption barriersclaude-sonnet-53/5While not licensed work requiring certification beyond CNA training, patient contact, hygiene, and dignity concerns create practical and organizational barriers to non-human handling.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized robotic systems capable of handling this task would be extremely expensive to develop, deploy, and maintain, far exceeding the loaded wage of a nursing assistant. The infrastructure and safety requirements make automation economically infeasible compared to direct human labor.
Cost vs. human wageclaude-sonnet-51/5There is no AI-based substitute for this physical task, so any comparison would require expensive specialized robotics far costlier than a human aide performing it.
Technical feasibility todayclaude-haiku-4-5-202510011/5While some research exists in robotic toileting assistance, no deployed commercial product reliably performs bedpan management in production healthcare settings. The task requires real-time adaptation to patient comfort, dignity, and varied physical configurations that exceed current system capabilities.
Technical feasibility todayclaude-sonnet-51/5No deployed products handle physical bedpan supply, collection, or emptying; this remains purely a human/robotic-hardware task not addressed by AI systems.

Lift or assist others to lift patients to move them on or off beds, examination tables, surgical tables, or stretchers.

4

CI 0–7 · exposure 0 · augmentation 38 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare is digitizing slowly; while some facilities trial mechanical lift-assist devices, adoption of automated patient transfer remains minimal and concentrated in wealthy academic medical centers, not widespread across nursing homes or general hospitals.
Sector adoption velocityclaude-sonnet-51/5Direct care and physical patient handling in healthcare settings show minimal AI/robotic adoption; this remains a highly manual, low-digitization task.
Augmentation potentialclaude-haiku-4-5-202510013/5Passive and powered lift-assist devices (mechanical arms, ceiling lifts) meaningfully reduce strain and injury risk for nursing assistants, improving their productivity and safety, though these are tools rather than AI systems.
Augmentation potentialclaude-sonnet-52/5Mechanical lift assist devices and sensors can support safer transfers, but AI-specific augmentation of this physical task is currently minimal.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of human bodies in variable positions and conditions, demanding real-time tactile feedback, balance adjustment, and responsiveness to patient comfort and safety—capabilities current AI robotics cannot reliably execute at scale without extensive task-specific engineering.
Task automatabilityclaude-sonnet-51/5This is a physical manual-handling task requiring direct human strength, balance, and coordination with a patient's body; no off-the-shelf AI system can perform it end-to-end.atorio
Adoption barriersclaude-haiku-4-5-202510015/5Patient safety, liability, and regulatory requirements (OSHA, CMS, state nursing boards) effectively require human judgment and physical presence; the care relationship itself and duty-of-care standards create hard legal and ethical barriers to full automation.
Adoption barriersclaude-sonnet-54/5Direct physical contact with vulnerable patients, safety liability, and clinical protocols requiring trained staff create strong barriers to automation of physical transfers.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized patient-lift robotic systems, where they exist, cost tens of thousands of dollars plus installation and maintenance, far exceeding the annual loaded wage of a nursing assistant performing this task.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical task, so any comparison favors the human worker; robotic lifting solutions remain expensive, experimental, and rare.
Technical feasibility todayclaude-haiku-4-5-202510011/5While some prototype robotic lifts exist in niche research settings, no deployed commercial product reliably performs safe patient transfers end-to-end across the variability of patient types, bed configurations, and clinical environments at production scale.
Technical feasibility todayclaude-sonnet-51/5No deployed product lifts or transfers patients autonomously in production; patient lift devices exist but are human-operated mechanical aids, not AI systems performing the task.

Answer patient call signals, signal lights, bells, or intercom systems to determine patients' needs.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of AI for routine task automation remains slow due to regulatory caution, patient safety requirements, and resistance to removing direct human availability on units where patients need immediate assistance.
Sector adoption velocityclaude-sonnet-51/5Healthcare direct-care settings, especially nursing assistant roles, show very low AI adoption for physical response tasks due to the hands-on nature of the work.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could marginally assist by prioritizing or logging call signals, but current systems offer minimal meaningful augmentation since nursing assistants must respond immediately to signals regardless; the assistance would be marginal and not transform productivity.
Augmentation potentialclaude-sonnet-52/5Smart call-light systems or triage software might help prioritize or route requests, but they offer only marginal assistance to the core physical response task.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires detecting an intentional patient signal and physically responding to it in real time. Current AI systems cannot autonomously monitor signal systems and respond to patient requests in clinical settings; they lack embodied presence and cannot replace the human availability required.
Task automatabilityclaude-sonnet-51/5This requires physically going to a patient's bedside, assessing their condition, and providing hands-on assistance or care, which current AI cannot perform.
Adoption barriersclaude-haiku-4-5-202510015/5Healthcare facilities have strong regulatory requirements, liability concerns, and patient safety standards that mandate immediate human response to patient signals. Legal and accreditation requirements typically require nursing staff presence and accountability, creating hard barriers to full automation.
Adoption barriersclaude-sonnet-54/5Direct physical patient care and safety response typically requires a present, often certified, caregiver, with liability and regulatory expectations for immediate human response in care facilities.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of deploying AI infrastructure (signal monitoring, interpretation, coordination systems) would far exceed the cost of a single nursing assistant available on the unit to respond to calls.
Cost vs. human wageclaude-sonnet-51/5AI has no comparable offering for physical patient response, so there is no viable cost comparison; the human is the only option.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs this task end-to-end. AI systems do not monitor nursing call lights or patient signals in production healthcare settings; this remains entirely a human responsibility.
Technical feasibility todayclaude-sonnet-51/5No deployed product physically responds to patient call signals and addresses in-person needs; this remains entirely a human physical-presence task.

Feed patients or assist patients to eat or drink.

3

CI 0–5 · exposure 0 · augmentation 13 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare is a laggard sector for physical automation; budget constraints, regulatory caution, and the centrality of human contact to patient experience make robotic feeding adoption negligible in practice today.
Sector adoption velocityclaude-sonnet-51/5Direct patient care and long-term care/nursing settings show very low AI/robotic adoption for physical caregiving tasks, lagging far behind information-work sectors.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist marginally by reminding caregivers of dietary restrictions or monitoring intake, but the core task—hand-feeding and swallowing support—depends entirely on human presence, so augmentation potential is limited.
Augmentation potentialclaude-sonnet-51/5Current AI tools offer negligible assistance for the physical act of feeding a patient; there's no meaningful software layer that improves this specific hands-on task.
Task automatabilityclaude-haiku-4-5-202510011/5Feeding patients requires physical manipulation of food/utensils, safe swallowing assistance, and real-time interaction with the patient's mouth and body. Current AI cannot physically perform this task or reliably assess choking risk and patient readiness to swallow.
Task automatabilityclaude-sonnet-51/5This is a hands-on physical caregiving task requiring manual dexterity, real-time responsiveness to swallowing/choking risk, and physical presence; no AI system can perform the physical feeding act itself.
Adoption barriersclaude-haiku-4-5-202510015/5Healthcare settings have strict liability frameworks, infection control protocols, and regulatory requirements (e.g., CMS, state nursing board rules) that typically mandate human care workers for direct patient feeding due to aspiration risk and duty of care.
Adoption barriersclaude-sonnet-54/5Feeding assistance involves aspiration/choking risk, requires trained personnel monitoring patient safety, and is often subject to care standards and liability concerns, creating strong barriers to automation even though not always a licensed sign-off task.
Cost vs. human wageclaude-haiku-4-5-202510011/5Any robot capable of safe feeding would cost tens of thousands of dollars in capital plus maintenance, far exceeding the hourly cost of a nursing assistant, with integration and oversight overhead.
Cost vs. human wageclaude-sonnet-51/5No viable AI substitute exists, so any comparison defaults to the human being far cheaper and more effective than any experimental feeding robot, which would be costly and unreliable.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can physically feed a patient or assist them to eat/drink. Robotic arms exist in labs but lack the dexterity, safety assurance, and adaptive control needed for the delicate, person-specific variations this task demands.
Technical feasibility todayclaude-sonnet-51/5There are no deployed products that physically feed or assist patients with eating and drinking; this remains outside current robotic and AI product capability at scale.

Position or hold patients in position for surgical preparation.

3

CI 0–5 · exposure 0 · augmentation 13 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a conservative, highly regulated sector with slow automation adoption; patient positioning in surgery is deeply embedded in clinical workflow and human judgment about patient welfare.
Sector adoption velocityclaude-sonnet-51/5Healthcare direct patient-care tasks involving physical handling show minimal AI/robotic adoption due to safety, regulatory, and physical dexterity constraints.
Augmentation potentialclaude-haiku-4-5-202510012/5AI offers minimal assistance for this task; visualization or positioning guidance systems exist in niche robotic surgery contexts, but they do not meaningfully augment the core activity of manually positioning and holding patients.
Augmentation potentialclaude-sonnet-51/5Current AI tools offer no meaningful assistance for the physical act of positioning or holding a patient during surgical prep.
Task automatabilityclaude-haiku-4-5-202510011/5Positioning and holding patients requires physical manipulation of a human body in a sterile surgical environment, demanding real-time tactile feedback, anatomical knowledge, and responsiveness to patient comfort and safety—capabilities entirely beyond current AI systems.
Task automatabilityclaude-sonnet-51/5This is a physical manipulation task requiring hands-on handling of a patient's body, which current AI systems (software or robotics) cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Patient safety, liability, sterility requirements, and direct human contact with vulnerable patients create strong regulatory and organizational barriers; a licensed human must legally oversee or perform patient handling in surgery.
Adoption barriersclaude-sonnet-54/5Patient safety, sterile field protocols, and clinical supervision requirements mean only trained staff can perform this task, creating strong procedural and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Surgical robotic systems capable of patient positioning are extremely expensive to acquire, integrate, and maintain, making human nursing assistants far more cost-effective for this task.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI/robotic substitute performing this physical task, so any hypothetical automation would require expensive specialized robotics far costlier than human labor today.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI robotic system reliably performs patient positioning and holding in surgical settings; this remains a purely human task in clinical practice today.
Technical feasibility todayclaude-sonnet-51/5No deployed robotic or AI product positions or physically holds surgical patients in production settings; this remains firmly a manual human task.

Set up treating or testing equipment, such as oxygen tents, portable radiograph (x-ray) equipment, or overhead irrigation bottles, as directed by a physician or nurse.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a laggard sector for physical automation; equipment setup is performed by human nursing staff in virtually all facilities due to liability, regulatory requirements, and the need for real-time judgment about patient context and safety.
Sector adoption velocityclaude-sonnet-51/5Direct patient care and physical equipment handling in healthcare settings show very slow AI adoption due to physical, regulatory, and safety constraints.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could potentially assist with checklists or safety reminders for equipment setup procedures, but the task is inherently physical and hands-on, limiting meaningful AI augmentation to documentation or training support rather than task performance.
Augmentation potentialclaude-sonnet-52/5AI could potentially assist with checklists, equipment tracking, or reminders, but offers minimal direct assistance for the physical setup task itself.
Task automatabilityclaude-haiku-4-5-202510011/5Setting up medical equipment requires physical manipulation in clinical environments, spatial reasoning, and real-time adaptation to patient conditions and equipment variations. Current AI systems cannot reliably perform hands-on assembly and positioning tasks in physical healthcare settings.
Task automatabilityclaude-sonnet-51/5This is a physical, hands-on setup task requiring manipulation of medical equipment in a patient's room, which current AI systems cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Healthcare equipment setup is legally regulated; equipment must be assembled correctly for patient safety, and liability for improper setup falls on healthcare providers. A licensed nurse or qualified assistant must verify correct setup before clinical use, creating a hard barrier to substitution.
Adoption barriersclaude-sonnet-54/5Direct patient care equipment setup typically requires trained, often certified staff following clinical protocols under supervision, creating strong regulatory and safety-driven barriers to substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5Any robotic system capable of reliable physical equipment setup in healthcare would require substantial hardware, maintenance, and integration costs far exceeding the loaded wage of a nursing assistant performing this task.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute for this physical task, so the human labor cost remains the only real option, making AI comparatively unusable rather than cheaper.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs physical equipment setup in clinical environments today. While robotics research exists, production systems for nursing assistant equipment setup do not exist in healthcare organizations at meaningful scale.
Technical feasibility todayclaude-sonnet-51/5No deployed product exists that physically sets up oxygen tents, portable x-ray equipment, or irrigation bottles; this remains firmly in the physical/manual domain.

Wash, groom, shave, or drape patients to prepare them for surgery, treatment, or examination.

3

CI 0–5 · exposure 0 · augmentation 13 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Despite acute nursing shortages, adoption of autonomous systems for patient grooming and hygiene remains negligible in production settings; the physical, regulatory, and cultural barriers have prevented meaningful deployment even in technology-forward healthcare systems.
Sector adoption velocityclaude-sonnet-51/5Healthcare direct-care work, especially hands-on physical patient care, is among the slowest sectors for AI/robotic adoption due to physical dexterity demands, safety requirements, and lack of mature robotic solutions.
Augmentation potentialclaude-haiku-4-5-202510012/5Current AI offers minimal assistance for this task; while scheduling or documentation tools might support workflow, they do not enhance the core manual grooming and washing work that remains fundamentally human-dependent.
Augmentation potentialclaude-sonnet-51/5AI offers essentially no meaningful assistance for the physical acts of washing, shaving, or draping a patient; this is fundamentally manual labor requiring human touch and judgment in the moment.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical manipulation of patients in varied positions and anatomies, direct body contact, and real-time responsiveness to patient comfort and safety. Current AI cannot perform end-to-end physical care tasks of this nature; robotic systems for patient washing/grooming remain in research phases without proven equal-quality outcomes at scale.
Task automatabilityclaude-sonnet-51/5This requires direct physical manipulation of a patient's body—washing, shaving, draping—which current AI systems cannot perform; no robotic system does this reliably in clinical settings today.
Adoption barriersclaude-haiku-4-5-202510015/5Regulations require direct human care contact for patient hygiene and dignity; liability exposure for robotic or automated systems performing intimate care is extremely high, and patient preference for human touch in pre-procedure care creates strong legal and organizational barriers to substitution.
Adoption barriersclaude-sonnet-54/5Physical patient contact for hygiene and surgical prep carries safety, infection control, dignity, and liability concerns that require trained human staff; while not always requiring a specific license beyond CNA certification, the physical care nature creates strong practical barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized robotics capable of safe patient handling would require substantial capital investment, maintenance, and oversight, far exceeding the cost of a nursing assistant's labor for routine hygiene tasks.
Cost vs. human wageclaude-sonnet-51/5There is no AI or robotic alternative to compare costs against; a human nursing assistant is the only viable option, making AI infinitely more 'expensive' by virtue of non-existence for this task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed commercial product reliably performs full-body patient grooming, washing, or draping independently. Robotic research prototypes exist but lack the dexterity, safety certification, and real-world validation needed for production healthcare settings.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs patient bathing, grooming, shaving, or draping; this remains entirely manual, hands-on caregiving work with no robotic substitutes in production.

Exercise patients who are comatose, paralyzed, or have restricted mobility.

3

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a laggard sector for physical automation of patient care due to regulatory burden, liability concerns, and the entrenched role of human caregivers. Current adoption of patient-handling robots is minimal and largely experimental.
Sector adoption velocityclaude-sonnet-51/5Healthcare direct patient care, especially physical assistance tasks, remains a low-digitization, high-touch sector with minimal robotic adoption for hands-on procedures like this.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could potentially assist by tracking exercise protocols or alerting staff to timing, but the physical manipulation itself requires human or advanced robotic execution, limiting meaningful augmentation of the nursing assistant's core task.
Augmentation potentialclaude-sonnet-52/5AI could assist with scheduling, tracking exercise regimens, or monitoring vital signs during sessions, but offers minimal augmentation to the actual physical act of exercising a patient.
Task automatabilityclaude-haiku-4-5-202510011/5Exercising immobilized patients requires physical manipulation of limbs, repositioning, and real-time assessment of patient comfort and response. Current AI systems lack embodied robotics with the dexterity, sensitivity, and adaptability to safely perform passive range-of-motion exercises and detect patient distress signals in real time.
Task automatabilityclaude-sonnet-51/5Physically manipulating and exercising immobile or comatose patients requires direct manual manipulation, tactile feedback, and physical strength that current AI systems cannot perform without embodied robotics, which are not deployed for this purpose.
Adoption barriersclaude-haiku-4-5-202510015/5Patient safety liability, regulatory oversight (FDA approval for medical devices), clinical judgment requirements, and direct physical contact with vulnerable patients create hard legal and organizational barriers. A licensed healthcare worker must remain responsible for patient welfare during therapeutic interventions.
Adoption barriersclaude-sonnet-54/5Physical patient handling carries significant liability and safety risk, often requires trained personnel and sometimes physician-directed protocols, creating strong barriers to non-human execution even if technology existed.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized medical robotics capable of safe patient handling remain extremely expensive to acquire, maintain, and integrate into healthcare facilities, far exceeding the loaded wage of a nursing assistant per patient interaction.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI/robotic substitute performing this physical task, so any hypothetical automation would require expensive specialized robotics far exceeding the cost of a nursing assistant's labor.
Technical feasibility todayclaude-haiku-4-5-202510011/5While some research robots exist for patient handling, no deployed commercial products reliably perform therapeutic exercise on comatose or paralyzed patients in production healthcare settings. The task requires safe physical contact and responsiveness that current robotic systems do not demonstrate at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs manual range-of-motion exercises or physical therapy movements on paralyzed or comatose patients; this remains outside current robotics capabilities in clinical settings.

Apply clean dressings, slings, stockings, or support bandages, under direction of nurse or physician.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare organizations, especially those employing nursing assistants, remain largely non-automated in direct patient care tasks due to regulatory, safety, and liability constraints. Adoption of robotics for this specific task is negligible in production settings.
Sector adoption velocityclaude-sonnet-51/5Direct patient care in nursing/long-term care settings is a low-digitization, physical-labor sector with minimal AI-driven displacement of hands-on tasks.
Augmentation potentialclaude-haiku-4-5-202510012/5AI offers minimal augmentation for applying physical dressings and bandages; perhaps minor assistance in scheduling or documenting dressing changes, but the core hands-on task cannot be meaningfully assisted by current AI without removing the human from direct execution.
Augmentation potentialclaude-sonnet-52/5AI could support documentation of wound status or provide training guidance, but it offers little direct assistance during the physical act of applying dressings or bandages.
Task automatabilityclaude-haiku-4-5-202510011/5Applying physical dressings, slings, stockings, and support bandages requires dexterous manipulation of fabric, precise positioning on varied patient anatomy, and real-time tactile feedback. Current robotic systems and AI cannot reliably perform this hands-on task end-to-end, and the requirement to work under clinical direction with real patients creates insurmountable practical barriers.
Task automatabilityclaude-sonnet-51/5This is a hands-on physical task requiring manual dexterity, tactile assessment of skin/wounds, and physical manipulation of a patient's body, none of which current AI systems can perform without robotic embodiment far beyond today's capabilities.
Adoption barriersclaude-haiku-4-5-202510015/5This task carries hard regulatory and liability barriers: a licensed or supervised healthcare worker must legally perform wound care and dressing application under clinical governance, and error costs (infection, patient harm) are high and asymmetric, preventing full automation.
Adoption barriersclaude-sonnet-54/5Direct physical patient contact under clinical supervision, infection control protocols, and liability concerns around wound care create strong practical and safety barriers to any automated substitute.
Cost vs. human wageclaude-haiku-4-5-202510011/5Robotic systems capable of performing this task (if they existed at production quality) would require significant capital investment, maintenance, and infrastructure costs far exceeding the loaded wage of a nursing assistant, making any substitution economically infeasible.
Cost vs. human wageclaude-sonnet-51/5There is no AI system that can substitute for the physical labor involved, so any comparison favors the human aide who can actually perform the task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI systems or robots can independently apply dressings and support devices to patients in clinical settings at scale. While robotic research exists, production systems do not exist that can handle the variability of patient bodies, wound conditions, and the need for safety and comfort during application.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs physical dressing or bandage application on patients; this remains entirely a human manual care task in clinical practice.

Transport patients to treatment units, testing units, operating rooms, or other areas, using wheelchairs, stretchers, or moveable beds.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare is a cautious, heavily regulated sector with slow AI adoption for patient-facing tasks. No measurable displacement of nursing assistants by autonomous transport systems has occurred in production healthcare settings.
Sector adoption velocityclaude-sonnet-51/5Healthcare support/physical care roles are among the slowest sectors for AI/robotic adoption due to safety-critical physical interaction requirements.
Augmentation potentialclaude-haiku-4-5-202510012/5Limited assistance could come from routing optimization or call-system improvements, but current AI cannot meaningfully augment the core physical act of safely moving patients. Wearable sensors or scheduling tools offer marginal gains.
Augmentation potentialclaude-sonnet-52/5AI could assist with scheduling/routing transport logistics, but offers minimal help with the core physical act of moving patients.
Task automatabilityclaude-haiku-4-5-202510011/5Physical transportation of patients requires mobile robots capable of safe patient handling, obstacle navigation in hospitals, and real-time interaction with patients—capabilities that do not yet exist in deployed form. Current AI systems cannot reliably perform this embodied, safety-critical task end-to-end.
Task automatabilityclaude-sonnet-51/5This requires physical manipulation of patients, wheelchairs, and stretchers through complex, dynamic hospital environments—far beyond current AI/robotics capability for reliable deployment.
Adoption barriersclaude-haiku-4-5-202510015/5Patient safety, liability concerns, and regulatory requirements (HIPAA, medical device regulation) create hard barriers. Hospitals have legal and ethical responsibility for patient welfare during transport, and no autonomous system is approved to assume this duty without human supervision.
Adoption barriersclaude-sonnet-54/5Patient safety, liability for falls/injury during transport, and requirements for trained staff to handle medical equipment and patient conditions create strong barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Autonomous patient transport robots (where they exist at all) cost hundreds of thousands of dollars and require infrastructure modification, IT integration, and continuous oversight—far exceeding the loaded wage of a nursing assistant.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical task, so any hypothetical robotic system would be far more expensive than a human aide given current robotics costs and integration needs.
Technical feasibility todayclaude-haiku-4-5-202510011/5No commercially deployed product reliably transports patients autonomously in hospital settings today. Experimental robotics exist but require extensive supervision and cannot match human performance in dynamic hospital environments with stairs, crowded corridors, and variable patient needs.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs whole-patient physical transport in hospitals today; robotic patient transport remains research/pilot stage at best.

Collect specimens, such as urine, feces, or sputum.

3

CI 0–5 · exposure 0 · augmentation 0 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of automation for direct patient contact tasks remains very low; patient care delivery remains highly human-dependent, and this specific task is not being displaced by automation in production settings.
Sector adoption velocityclaude-sonnet-51/5Direct care and physical patient-handling tasks in healthcare facilities show minimal AI/robotic adoption; this remains a low-digitization, hands-on task.
Augmentation potentialclaude-haiku-4-5-202510011/5AI offers no meaningful assistance in the core task of physically collecting biological specimens; the task is inherently manual and interpersonal, leaving no augmentation opportunity for current AI.
Augmentation potentialclaude-sonnet-51/5AI offers negligible assistance for the physical act of collecting specimens, though it may help with subsequent labeling or documentation, which is outside this specific task.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires physical collection of biological specimens directly from patients, which demands in-person contact, manual dexterity, and interaction with living individuals. Current AI systems lack embodied presence and cannot perform the hands-on collection.
Task automatabilityclaude-sonnet-51/5This is a physical hands-on task requiring direct patient contact, positioning, and handling of biological materials, which current AI systems (software or general robotics) cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Significant legal and regulatory barriers exist: patient contact and dignity, infection control protocols, and clinical liability all require a trained human professional to perform or directly supervise specimen collection.
Adoption barriersclaude-sonnet-54/5Direct physical patient contact, infection control protocols, and certification/training requirements for handling bodily specimens create strong practical and regulatory barriers to non-human performance.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI systems cannot perform this task at all, making any cost comparison theoretical. The actual cost of deploying autonomous collection systems would far exceed the loaded wage of a nursing assistant.
Cost vs. human wageclaude-sonnet-51/5There is no AI system offering this physical service, so any comparison to human labor cost is moot; AI cannot substitute at any price point today.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product can physically collect biological specimens from patients. This is fundamentally a physical, embodied task that requires a human or robot with sophisticated manipulation capabilities, neither of which are in clinical production for this purpose.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs specimen collection from patients; this remains purely a human manual-care task with no robotic substitute in production healthcare settings.

Administer medications or treatments, such as catheterizations, suppositories, irrigations, enemas, massages, or douches, as directed by a physician or nurse.

0

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare remains a laggard sector for robotic or autonomous task substitution in hands-on nursing care; physical patient-contact work is heavily regulated, unionized in many settings, and organizations have strong incentives to retain human oversight for liability and quality-of-care reasons.
Sector adoption velocityclaude-sonnet-51/5Healthcare direct-care work, especially hands-on nursing assistant tasks, shows minimal automation adoption due to physical and regulatory constraints.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with task scheduling, documentation, or protocol reminders, but current systems offer minimal augmentation for the core physical and interpersonal acts of administering medications or treatments at the bedside.
Augmentation potentialclaude-sonnet-52/5AI can support documentation, reminders, or protocol checklists around these tasks, but offers little assistance in the physical execution of treatments themselves.
Task automatabilityclaude-haiku-4-5-202510011/5This task involves direct patient contact, physical manipulation of medical devices and bodily administration of treatments, and requires real-time assessment of patient response. Current AI systems cannot physically perform these interventions or safely handle the variability in patient anatomy, comfort, and medical status that demands human judgment in real time.
Task automatabilityclaude-sonnet-51/5This requires direct physical manipulation of patients' bodies (catheterization, enemas, massages) which current AI systems cannot perform; no robotic system is deployed for this purpose in care settings.
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and regulatory barriers exist: nursing assistants work under direct supervision of licensed nurses or physicians, liability for medication errors and patient harm is severe, and most jurisdictions legally require a credentialed human to perform or directly oversee administration of medical treatments and intimate care tasks.
Adoption barriersclaude-sonnet-55/5Direct patient care involving invasive procedures requires licensed/certified personnel, physician or nurse direction, and carries significant liability and safety risk, making substitution legally and practically barred.
Cost vs. human wageclaude-haiku-4-5-202510011/5The capital and integration costs of hypothetical robotic systems capable of intimate patient care far exceed the loaded wage of a nursing assistant, and no such system is commercially available to compare.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical task, so cost comparison favors the human by default since no automated alternative exists at any price.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI systems can autonomously perform catheterizations, suppositories, irrigations, enemas, massages, or douches. These require physical robots with dexterous manipulation in unstructured human settings, which do not exist in clinical production environments.
Technical feasibility todayclaude-sonnet-51/5No commercial product performs hands-on administration of these treatments; this remains firmly in the physical/manual domain outside AI product scope.

Related occupations — Healthcare Support

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.