Registered Nurses

29-1141.00
Median wage $97,550/yr3,379,720 employed (US)Rank #719 of 923 scored · top 78% by substitution

Assess patient health problems and needs, develop and implement nursing care plans, and maintain medical records. Administer nursing care to ill, injured, convalescent, or disabled patients. May advise patients on health maintenance and disease prevention or provide case management. Licensing or registration required.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution18
Exposure19
Augmentation58

Substitution — the headline: capability discounted by cost, barriers and adoption.

Exposure — technical capability alone, regardless of whether anyone deploys it.

Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.

Tasks on the substitution scale

27 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%19

panel mean rating 1.8/5 → substitution pressure 19/100

Technical feasibility todayw 20%20

panel mean rating 1.8/5 → substitution pressure 20/100

Cost vs. human wagew 15%18

panel mean rating 1.7/5 → substitution pressure 18/100

Adoption barriersw 20%inverted — strong barriers lower the score14

panel mean rating 4.4/5 (barrier strength) → substitution pressure 14/100

Sector adoption velocityw 10%22

panel mean rating 1.9/5 → substitution pressure 22/100

Task breakdown (27 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.

Record patients' medical information and vital signs.

56

CI 39–74 · exposure 62 · augmentation 75 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Healthcare is a digitized sector with rapid EHR and automation adoption; vital-sign monitoring systems and automated data capture are already deployed at scale in most hospitals and many outpatient settings.
Sector adoption velocityclaude-sonnet-52/5Healthcare overall lags in AI adoption due to regulatory, interoperability, and liability constraints, though pilots of ambient documentation tools are growing.br
Augmentation potentialclaude-haiku-4-5-202510014/5AI-assisted charting and vital-sign alerts significantly assist nurses by reducing manual entry time and flagging abnormal values, allowing them to focus on patient care and clinical decision-making while remaining responsible for validation.
Augmentation potentialclaude-sonnet-54/5AI-powered vital sign monitors and ambient scribes substantially reduce charting time and cognitive load, letting nurses focus more on patient care while remaining responsible for accuracy.
Task automatabilityclaude-haiku-4-5-202510014/5Recording vital signs and medical information from monitors and structured sources is largely automatable with current EHR integration and AI-assisted data capture systems, achieving >50% time savings. However, the task retains some human judgment for unusual values or patient context that may require manual verification.
Task automatabilityclaude-sonnet-53/5Vital signs from connected devices can auto-populate EHRs, and ambient documentation tools can transcribe and structure clinical notes, but nurses still must verify accuracy, contextualize observations, and handle exceptions manually.br
Adoption barriersclaude-haiku-4-5-202510013/5Healthcare regulations (HIPAA, documentation standards) and liability concerns require human review and sign-off; many institutions retain mandatory nurse verification of automated records for safety and compliance, creating meaningful organizational friction.
Adoption barriersclaude-sonnet-54/5Clinical documentation is subject to regulatory, legal, and accreditation requirements (e.g., accurate charting, licensure accountability) that require a qualified nurse to verify and be responsible for the record.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated vital-sign capture and EHR integration cost a fraction of a nurse's hourly wage, especially at scale; the all-in cost (hardware, software, oversight) is an order of magnitude lower than manual recording and entry.
Cost vs. human wageclaude-sonnet-52/5Devices and software require significant integration, licensing, and IT support costs, and human oversight is still needed, so savings versus nurse time are moderate rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510014/5Mature EHR systems with automated vital-sign logging from devices and structured data entry are widely deployed in hospitals and clinics; AI-assisted documentation (e.g., voice-to-text for charting) is common. Some integration challenges and verification steps remain standard practice.
Technical feasibility todayclaude-sonnet-53/5Ambient AI scribes and integrated monitoring-to-EHR systems are deployed in some hospitals today, but adoption is uneven and human verification/correction remains standard practice.

Maintain accurate, detailed reports and records.

54

CI 39–70 · exposure 62 · augmentation 88 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Healthcare organizations, especially larger hospital systems, have been rapidly adopting AI documentation and EHR enhancements since ~2020. Ambient intelligence and voice-to-text clinical documentation are now commonplace in many hospital workflows, reflecting fast adoption within the healthcare information sector.
Sector adoption velocityclaude-sonnet-52/5Healthcare is a historically slow-adopting sector for AI tools, though documentation-focused AI is one of the faster-growing niches within it.
Augmentation potentialclaude-haiku-4-5-202510015/5AI documentation assistance meaningfully augments nurse productivity by drafting notes, auto-populating templates, and organizing data, allowing nurses to focus on patient care rather than clerical work. This is among the most successful human–AI collaborative workflows in healthcare today.
Augmentation potentialclaude-sonnet-54/5AI-assisted documentation tools can significantly speed up note-taking and reduce administrative burden while the nurse remains responsible for final accuracy.
Task automatabilityclaude-haiku-4-5-202510014/5Registered nurses maintain electronic health records (EHRs) documenting patient vitals, medications, observations, and care plans. Current AI systems can auto-populate fields from voice/image input, extract and summarize clinical notes, and flag missing documentation—reducing documentation time by 40–60% at comparable accuracy. Full end-to-end automation is limited by the need for human clinical judgment on what to record and verification.
Task automatabilityclaude-sonnet-53/5AI can draft and structure clinical documentation from dictation or EHR data, but nurses must verify accuracy and add clinical judgment, so only partial time savings occur today.
Adoption barriersclaude-haiku-4-5-202510013/5Regulatory bodies (CMS, state boards) require that nurses verify and sign off on medical records, and liability for errors remains with the healthcare provider. While AI can assist, nurses must retain final authority and accountability, creating a permanent human-in-the-loop requirement.
Adoption barriersclaude-sonnet-54/5Medical record accuracy is legally and clinically critical, with licensure, liability, and regulatory (HIPAA, compliance) requirements mandating human accountability for entries.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI documentation tools cost $5–15k annually per user, versus a registered nurse's fully loaded wage of ~$80–120k/year. Automation of 40–50% of documentation time yields substantial cost savings relative to hiring additional nursing staff, making the cost ratio highly favorable.
Cost vs. human wageclaude-sonnet-52/5AI scribe/documentation tools add licensing and integration costs plus required human review, so savings versus nurse time are moderate, not order-of-magnitude cheaper.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed EHR systems, voice-to-text AI documentation tools (e.g., Nuance Dragon, ambient intelligence platforms), and note-generation systems are actively used in hospitals and clinics. These handle substantial portions of routine documentation reliably, though they still require human review and correction for accuracy and completeness.
Technical feasibility todayclaude-sonnet-53/5Ambient documentation and clinical note-generation products (e.g., AI scribes) are deployed in some health systems, but adoption is uneven and error-checking remains necessary.

Provide or arrange for training or instruction of auxiliary personnel or students.

30

CI 25–35 · exposure 25 · augmentation 75 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Healthcare organizations are experimenting with AI-assisted training and simulation tools, but adoption remains mixed; financial constraints and regulatory caution slow deep deployment, though pilot adoption is becoming more common in larger systems.
Sector adoption velocityclaude-sonnet-52/5Healthcare training adoption of AI is emerging (e.g., simulation, e-learning modules) but is generally slower than in software/finance sectors due to compliance and hands-on requirements.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can significantly assist nurses in training by generating customized learning materials, tracking trainee progress, flagging knowledge gaps, and enabling asynchronous review—substantially raising instructor productivity while the nurse remains actively engaged in assessment and feedback.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by creating training materials, quizzes, simulations, and personalized learning paths, augmenting the nurse's instructional efforts.
Task automatabilityclaude-haiku-4-5-202510012/5Training and instruction require responsiveness to learner needs, assessment of understanding, and adaptive feedback—capabilities where current AI falls short at scale. While AI can generate training materials or suggest instructional approaches, the core mentoring, assessment, and interpersonal adjustment that defines effective instruction remains difficult to fully automate.
Task automatabilityclaude-sonnet-52/5Designing and delivering hands-on clinical training and mentorship involves demonstration, supervision, and adaptive feedback that current AI cannot fully replicate end-to-end, though it can help produce materials.
Adoption barriersclaude-haiku-4-5-202510014/5Nursing education and competency verification often require licensed nurses to directly oversee, evaluate, and sign off on learner performance; many hospitals and educational accreditation standards mandate human nurse involvement in clinical skills assessment and sign-off.
Adoption barriersclaude-sonnet-54/5Clinical training often requires licensed nurse oversight for competency verification and regulatory/accreditation compliance, creating strong barriers to full substitution.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-generated instructional content and automated tutoring modules are substantially cheaper than dedicating a registered nurse's time to training, especially for basic onboarding. Even with oversight, AI support for training materials and initial instruction is orders of magnitude less costly than human-led training.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply generate training materials, but the core supervisory/hands-on instruction still requires paid nurse time, so overall cost savings are limited.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably conducts end-to-end nursing training or instruction at production scale. Learning management systems exist, but they lack the real-time adaptation, clinical judgment application, and one-on-one corrective feedback that nursing instruction demands.
Technical feasibility todayclaude-sonnet-52/5Some e-learning and AI-generated training content tools exist, but no deployed product reliably runs full clinical instruction or supervision of auxiliary staff/students in practice settings.

Engage in research activities related to nursing.

29

CI 25–34 · exposure 25 · augmentation 63 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare and academic research sectors show moderate digitization but slow AI adoption for core research design and execution. Most nursing research remains traditional and human-led, with AI tools adopted only for peripheral tasks like literature review or manuscript drafting.
Sector adoption velocityclaude-sonnet-52/5Healthcare research settings are adopting AI tools for literature review and data analysis, but broader clinical/nursing environments show slower, more cautious AI adoption compared to fully digital sectors.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist nurses conducting research through literature synthesis, data management, statistical guidance, and manuscript drafting, materially raising productivity on those components while the researcher retains full intellectual and ethical responsibility.
Augmentation potentialclaude-sonnet-54/5AI tools substantially assist nurses in literature review, data analysis, drafting manuscripts, and identifying research trends, meaningfully boosting productivity while humans retain oversight of study design and ethics.
Task automatabilityclaude-haiku-4-5-202510012/5Research activities in nursing often require designing studies, synthesizing complex medical literature, conducting analyses, and interpreting nuanced clinical findings. While AI can assist with literature searches, data organization, and preliminary analysis, the core design, judgment, and hypothesis-generation elements remain human-dependent and cannot achieve 50% time savings at equal quality end-to-end.
Task automatabilityclaude-sonnet-52/5Nursing research involves literature review, study design, data collection, and analysis, some of which AI can accelerate, but hypothesis generation, ethical oversight, and clinical judgment remain human-driven, so full end-to-end automation is not yet achievable at 50% time savings with equal quality.
Adoption barriersclaude-haiku-4-5-202510014/5Nursing research is often conducted under institutional review boards (IRBs), grant requirements, and regulatory oversight; publications require researcher accountability and institutional affiliation; liability and ethical responsibility for study design and participant safety rest with human researchers.
Adoption barriersclaude-sonnet-53/5Research involving human subjects requires IRB approval, ethical oversight, and often credentialed investigators, creating moderate regulatory and organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for research assistance (cloud databases, language models) carry meaningful ongoing costs (subscriptions, integration), while a nurse researcher's full wage includes domain expertise that AI cannot yet fully replace, making all-in costs roughly comparable or favoring human researchers.
Cost vs. human wageclaude-sonnet-53/5AI can cheaply handle literature synthesis and data analysis portions, but the overall research process still requires substantial human labor for design, IRB compliance, and interpretation, keeping costs roughly comparable when the full task is considered.
Technical feasibility todayclaude-haiku-4-5-202510012/5No mature deployed products perform nursing research end-to-end reliably. Existing tools (literature databases, statistical software, AI writing assistants) handle narrow subtasks, but research planning, clinical context integration, and peer-review standards require human researchers in production settings.
Technical feasibility todayclaude-sonnet-52/5AI literature search and summarization tools are used in research support, but no deployed product independently conducts nursing research studies reliably; current use is assistive and narrow.

Perform administrative or managerial functions, such as taking responsibility for a unit's staff, budget, planning, or long-range goals.

27

CI 25–29 · exposure 25 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare remains slow to adopt AI-driven autonomous management due to liability concerns, regulatory conservatism, and entrenched hierarchies. While analytics tools proliferate, actual replacement of nurse manager decision authority is minimal and proceeding slowly.
Sector adoption velocityclaude-sonnet-52/5Healthcare administration adopts digital tools unevenly and cautiously, with slower AI integration into managerial decision-making compared to sectors like finance or tech.
Augmentation potentialclaude-haiku-4-5-202510013/5AI augments managerial work well via predictive analytics, budget modeling, and staff scheduling recommendations; nurse managers routinely use dashboards and data tools that raise their efficiency. However, augmentation is limited to decision support rather than transformative replacement of core judgment.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist nurse managers via predictive staffing models, budget dashboards, and reporting automation, improving efficiency while the manager retains decision authority.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with budget analysis, scheduling optimization, and data reporting, core managerial functions—accountability for unit performance, strategic decision-making, and staff oversight—require sustained human judgment and cannot be fully automated today. AI can automate perhaps 20–30% of routine administrative work but cannot independently manage unit goals or staff relationships at the required quality threshold.
Task automatabilityclaude-sonnet-52/5This is a broad managerial task involving human judgment, staff leadership, budget accountability, and strategic planning that requires accountability and interpersonal authority AI cannot assume; only sub-components like data compilation or scheduling drafts can be automated.
Adoption barriersclaude-haiku-4-5-202510014/5Strong legal and organizational barriers exist: nursing leadership carries direct liability for unit outcomes, regulatory compliance, and staff performance. Healthcare accreditation bodies and state regulations typically require licensed practitioners (RNs) to hold managerial accountability, not automated systems.
Adoption barriersclaude-sonnet-54/5Unit leadership requires a licensed, accountable individual with legal and organizational responsibility for staff and budget outcomes, creating strong institutional and regulatory barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510013/5AI-powered analytics and administrative automation are cost-competitive with junior administrative staff on specific subtasks, but cannot replace mid-to-senior nurse managers. The integrated cost of oversight, training, and integration roughly matches the mid-level wage it would displace.
Cost vs. human wageclaude-sonnet-52/5While analytics tools reduce some administrative overhead cheaply, the accountable managerial role still requires a paid nurse manager, so overall cost savings are modest rather than transformative.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs managerial accountability end-to-end. AI tools exist for budget dashboards and scheduling suggestions, but real management systems in hospitals still require human oversight and sign-off; they do not operate autonomously.
Technical feasibility todayclaude-sonnet-52/5AI tools exist for scheduling optimization, budget analytics, and staffing forecasts, but no deployed product performs unit management, accountability, or long-range planning autonomously in healthcare settings today.

Instruct individuals, families, or other groups on topics such as health education, disease prevention, or childbirth and develop health improvement programs.

27

CI 25–29 · exposure 25 · augmentation 75 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare remains a relatively slow-adopting sector for autonomous AI; regulatory caution, patient safety mandates, and institutional risk-aversion limit deployment of AI-led health instruction to pilots and supplementary roles, not mainstream replacement.
Sector adoption velocityclaude-sonnet-52/5Healthcare overall is a slower-adopting sector for patient-facing AI due to regulatory, safety, and trust concerns, with pilots more common than widespread deployment for this specific task.
Augmentation potentialclaude-haiku-4-5-202510014/5AI tools demonstrably assist nurses by generating evidence-based educational summaries, drafting program materials, and suggesting tailored content; nurses retain authority and personalization, substantially raising their output while maintaining the human-nurse relationship as essential.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully assist nurses by drafting educational materials, translating health information, and helping design health improvement program content, boosting efficiency while the nurse retains the interpersonal teaching role.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can generate educational content and draft health materials, delivering effective health instruction requires assessing individual readiness, cultural sensitivity, and behavioral adaptation—capabilities current AI systems lack reliably. The task involves real-time pedagogical judgment and relationship-building that does not meet the 50% time-saving threshold for end-to-end automation.
Task automatabilityclaude-sonnet-52/5AI can generate educational content and personalized materials, but delivering instruction requires interpersonal trust, adaptive communication, and physical/emotional presence that current systems cannot fully replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5State nursing boards legally restrict health education and program development to licensed nurses or physicians; liability and malpractice concerns create strong incentives to keep human professionals accountable for instruction accuracy. Patient trust and informed-consent requirements further cement the human role.
Adoption barriersclaude-sonnet-54/5Nursing scope-of-practice regulations, licensure requirements, and liability for health education/counseling create strong barriers to full AI substitution, especially for clinical judgment-based instruction.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI content generation and delivery systems have non-negligible costs (platform subscriptions, integration, quality oversight), and still require nurses for validation and customization, making all-in automation costs comparable to or higher than nurse-led instruction.
Cost vs. human wageclaude-sonnet-53/5AI-generated content is cheap to produce, but a nurse's time for tailored counseling and program development still requires human oversight, making blended costs roughly comparable rather than dramatically cheaper.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI chatbots and educational platforms exist but are not deployed at scale by healthcare organizations as primary instructors; they function as supplementary tools rather than replacements. Production systems typically require human nurses for authoritative health counseling and program design in clinical settings.
Technical feasibility todayclaude-sonnet-52/5Chatbots and patient education apps exist and are used for supplementary health information, but no deployed product independently conducts patient/family health teaching or childbirth education at scale in clinical settings.

Prepare rooms, sterile instruments, equipment, or supplies and ensure that stock of supplies is maintained.

26

CI 21–30 · exposure 25 · augmentation 50 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Hospitals have adopted inventory management software, but adoption of physical automation for room prep and sterilization remains slow and limited to large academic centers. Most facilities still rely on nursing staff and technicians for these tasks, indicating laggard sector adoption of full automation.
Sector adoption velocityclaude-sonnet-52/5Healthcare facilities adopt supply-chain automation slowly due to safety-critical processes, legacy systems, and capital constraints, though RFID/inventory systems are gradually spreading.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted inventory forecasting, automated reordering, and mobile apps for supply tracking meaningfully improve nurse productivity in the planning and monitoring aspects of the task. However, the core physical work of preparation and sterilization validation remains human-dependent, limiting augmentation upside.
Augmentation potentialclaude-sonnet-53/5AI-enabled inventory management and automated restocking alerts can meaningfully assist nurses in maintaining supply levels, though the physical setup remains manual.
Task automatabilityclaude-haiku-4-5-202510012/5While some sub-tasks like inventory tracking and ordering can be partially automated, the physical preparation of rooms, handling of sterile instruments, and hands-on equipment setup require human presence and dexterity that current robotics cannot reliably perform end-to-end in hospital environments. The sterility requirements and need for human judgment about equipment readiness prevent ≥50% time savings at equal quality.
Task automatabilityclaude-sonnet-52/5This is largely a physical, hands-on preparation and inventory task involving handling sterile equipment and rooms, which current AI cannot physically perform; only inventory tracking/ordering aspects could be automated.
Adoption barriersclaude-haiku-4-5-202510014/5Strong regulatory requirements govern sterile field maintenance and instrument handling; healthcare licensing and Joint Commission standards typically require trained personnel to validate sterility and equipment readiness. Legal liability for contamination creates high error-cost asymmetry that restricts delegation to automated systems.
Adoption barriersclaude-sonnet-53/5Sterile technique and infection control protocols require trained personnel, though software-assisted supply tracking has fewer regulatory barriers than the physical task itself.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current solutions (inventory software, basic robotics) require significant capital investment and ongoing integration costs that exceed the loaded wage of nursing staff performing these preparatory duties. Full end-to-end automation remains economically unviable today.
Cost vs. human wageclaude-sonnet-52/5Physical preparation still requires human labor and specialized robotics are costly and rare; only the inventory-tracking software component offers cost savings.
Technical feasibility todayclaude-haiku-4-5-202510012/5Limited deployed products exist for full task automation. Inventory management systems are mature, but room preparation and sterile instrument handling remain largely manual in practice. Robotic systems for material transport exist in some hospitals, but they perform only narrow sub-tasks and do not reliably handle the complete workflow.
Technical feasibility todayclaude-sonnet-52/5Automated inventory management systems and smart cabinets exist in some hospitals for supply tracking, but no deployed product performs sterile room/instrument preparation itself.

Assess the needs of individuals, families, or communities, including assessment of individuals' home or work environments, to identify potential health or safety problems.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI for independent patient assessment is slow and limited to pilots; nursing assessment remains a core human function. Healthcare sectors show cautious adoption due to regulatory oversight and risk-aversion in clinical settings.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially home/community nursing, has been slower to adopt AI agents for hands-on assessment tasks compared to administrative or diagnostic support functions.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist nurses by flagging structured risk factors, organizing assessment data, or surfacing relevant patient history, which speeds documentation and decision-support. However, the core assessment remains nurse-driven.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by organizing intake data, flagging risk indicators, generating checklists, and supporting documentation, improving nurse efficiency during assessments.
Task automatabilityclaude-haiku-4-5-202510012/5Assessing health/safety needs requires contextual judgment, observation of non-verbal cues, and integration of complex social factors that current AI cannot reliably perform end-to-end. While AI can assist with preliminary data analysis or documentation review, the core assessment task demands human clinical expertise and environmental evaluation.
Task automatabilityclaude-sonnet-52/5Assessment relies heavily on in-person observation, physical examination, and contextual judgment about environments that AI cannot directly perceive or evaluate end-to-end today.
Adoption barriersclaude-haiku-4-5-202510014/5Strong regulatory and liability barriers exist: registered nurses are legally required to perform patient assessments, and liability for missed health/safety problems creates asymmetric error costs that deter substitution. Clinical judgment and professional accountability remain non-delegable.
Adoption barriersclaude-sonnet-54/5Licensed nursing judgment is typically required for health/safety assessments, with liability and regulatory scope-of-practice rules restricting delegation to non-human systems.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI solutions for health assessment (if deployed) require significant human oversight, integration, and clinical validation, making per-task costs comparable to or higher than direct nurse assessment time in most contexts.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply process intake data or flag risk factors, but the core in-person assessment still requires a nurse, so overall cost savings are limited.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed products reliably perform comprehensive needs assessments for individuals, families, or communities in production nursing settings. AI systems exist for narrow screening tasks (e.g., risk flagging from structured data), but independent, clinically validated assessment remains research-stage.
Technical feasibility todayclaude-sonnet-52/5Some clinical decision-support and risk-screening tools exist, but no deployed product independently conducts home/work environment safety assessments in production.

Refer students or patients to specialized health resources or community agencies furnishing assistance.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI is moderate and cautious; while some EHR-integrated resource tools exist, systematic AI-driven autonomous referral is rare and adoption remains slow due to clinical governance and liability concerns.
Sector adoption velocityclaude-sonnet-52/5Healthcare is a moderate-to-slow adopting sector for AI-driven decision tasks due to regulatory, liability, and workflow integration challenges, with most AI use still confined to documentation and administrative support.
Augmentation potentialclaude-haiku-4-5-202510013/5AI tools that surface relevant community resources, filter options by patient criteria, or flag gaps in local services can meaningfully assist nurses in making faster, more comprehensive referrals. The human nurse retains decision authority while AI improves information access.
Augmentation potentialclaude-sonnet-54/5AI tools can efficiently search and compile relevant community resources, insurance-covered specialists, or support agencies, meaningfully speeding up the nurse's research and reducing administrative burden.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can retrieve and match available resources, the task requires understanding patient context, barriers to access, and personalized judgment about appropriateness—dimensions that remain difficult for current systems without human oversight. The referral decision itself contains clinical and social reasoning that AI cannot reliably perform end-to-end at the 50% time-saving threshold.
Task automatabilityclaude-sonnet-52/5While AI can suggest referral options based on patient data, the actual judgment of appropriateness, timing, and personal follow-through with patients/students requires clinical judgment and relationship-based communication that current systems cannot fully replace.
Adoption barriersclaude-haiku-4-5-202510014/5Nurses are liable for inappropriate referrals and must ensure patient safety and consent; institutional policy typically requires human clinical judgment in referral decisions. The licensed professional carries accountability, creating a strong barrier to full automation.
Adoption barriersclaude-sonnet-54/5Referrals often require licensed clinical judgment, liability considerations, and adherence to care standards, meaning a nurse must typically authorize or perform this task in most healthcare settings.
Cost vs. human wageclaude-haiku-4-5-202510012/5Implementing AI-assisted referral systems, maintaining resource databases, and managing the oversight required to ensure appropriate recommendations costs significantly relative to a nurse's marginal time spent on a single referral. The integration overhead is substantial.
Cost vs. human wageclaude-sonnet-52/5AI could cheaply generate resource lists, but the human oversight, verification, and patient interaction needed still requires a nurse's time, keeping costs comparable rather than dramatically lower.
Technical feasibility todayclaude-haiku-4-5-202510012/5Some products exist for resource databases and matching, but deployed systems lack the clinical judgment integration and real-world reliability needed for actual patient referrals. Current tools remain aids rather than autonomous performers of the full task in production settings.
Technical feasibility todayclaude-sonnet-52/5Some clinical decision support tools and care coordination platforms suggest referral resources, but no deployed product independently manages the full referral process including patient counseling and follow-up at scale.

Monitor, record, and report symptoms or changes in patients' conditions.

25

CI 20–30 · exposure 30 · augmentation 75 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI-driven monitoring is slow and highly fragmented; most deployments remain pilot-stage or confined to specific units (ICUs). Organizational resistance, regulatory caution, liability concerns, and integration complexity with legacy EHR systems limit production-scale adoption.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially bedside nursing, is a slow-adopting sector for AI due to regulatory, safety, and workflow integration challenges; pilots exist but production-scale replacement is rare.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-powered dashboards, alert prioritization, and predictive analytics significantly assist nurses by surfacing changes, reducing manual chart review, and enabling proactive intervention. Nurses retain clinical judgment while AI amplifies their capacity to monitor and detect patterns across multiple patients.
Augmentation potentialclaude-sonnet-54/5AI-based monitoring systems, predictive alerts (e.g., sepsis or deterioration detection), and automated documentation tools meaningfully assist nurses in tracking and reporting patient changes.
Task automatabilityclaude-haiku-4-5-202510012/5Monitoring vital signs and recording structured data can be partially automated via sensors and EHR integration, but detecting subtle clinical changes, interpreting context, and deciding what constitutes a reportable change require significant human judgment. Current AI cannot reliably replace the full end-to-end task of determining clinical significance and making care decisions.
Task automatabilityclaude-sonnet-52/5AI can help aggregate and flag vitals data from monitors and EHRs, but the core task requires physical presence, hands-on assessment, and clinical judgment that current systems cannot perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510014/5Strong regulatory and liability barriers exist: nurses are legally and professionally accountable for accurate assessment and reporting; clinical decisions based on missed changes carry high error costs and malpractice exposure. Healthcare licensing and standard-of-care requirements make it difficult to remove the human from sign-off.
Adoption barriersclaude-sonnet-55/5Nursing licensure, scope-of-practice laws, and patient safety liability require a licensed nurse to assess and document patient conditions, making this a hard-barrier task.
Cost vs. human wageclaude-haiku-4-5-202510012/5Automated monitoring hardware and EHR integration have upfront costs and ongoing maintenance; when accounting for integration, interoperability, and required human oversight, the all-in cost remains comparable to or higher than the nursing labor for small-to-medium deployments.
Cost vs. human wageclaude-sonnet-52/5AI monitoring tools add cost on top of nursing labor since a nurse must still physically observe and validate patient status; software licensing and integration costs are non-trivial relative to marginal task savings.
Technical feasibility todayclaude-haiku-4-5-202510013/5Deployed systems exist for automated vital-sign capture and basic EHR documentation, but they operate in narrow contexts (ICU monitors, wearables) with limitations. Reliably detecting all clinically meaningful changes and integrating them into real clinical workflows remains challenging; most systems require substantial human oversight.
Technical feasibility todayclaude-sonnet-52/5Deployed clinical decision support and monitoring alert systems exist (e.g., early warning scores), but they are narrow, error-prone, and require nurse verification rather than replacing observation and reporting.

Order, interpret, and evaluate diagnostic tests to identify and assess patient's condition.

23

CI 20–25 · exposure 25 · augmentation 75 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Hospitals and health systems are adopting AI diagnostic aids (e.g., imaging software) in pilots and narrow deployments, but adoption is cautious due to liability, regulatory scrutiny, and the need for human oversight. Deep, production-scale replacement of this core nursing task remains limited.
Sector adoption velocityclaude-sonnet-52/5Healthcare overall adopts AI diagnostic aids cautiously due to regulatory approval, liability, and interoperability challenges, with pilots more common than widespread production deployment for this specific task.
Augmentation potentialclaude-haiku-4-5-202510014/5AI significantly augments nurses by rapidly flagging abnormal results, providing interpretive summaries, and surfacing relevant clinical patterns—enabling faster assessment and earlier intervention. These tools enhance nurse productivity and decision-making while the nurse retains full responsibility.
Augmentation potentialclaude-sonnet-54/5AI-based decision support, lab flagging, and imaging analysis tools meaningfully help nurses prioritize and interpret diagnostic data faster while the nurse retains final clinical judgment.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in interpreting some diagnostic test results (e.g., image analysis), nurses must exercise clinical judgment integrating patient history, symptoms, and context—tasks requiring human expertise. Ordering tests also involves nuanced decision-making about appropriateness and patient-specific factors that current AI cannot autonomously handle reliably.
Task automatabilityclaude-sonnet-52/5AI can assist in interpreting some diagnostic tests (e.g., imaging, lab values) but ordering appropriate tests and integrating results into a full clinical assessment requires contextual judgment, patient interaction, and accountability that current systems cannot fully replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Nursing practice is regulated; nurses are legally responsible for ordering and evaluating diagnostic tests as part of their licensure scope. Liability and patient safety requirements create strong barriers to full automation, and institutional protocols mandate nurse interpretation and judgment sign-off.
Adoption barriersclaude-sonnet-55/5Ordering and interpreting diagnostic tests to determine patient condition is a licensed clinical function with strict regulatory, liability, and scope-of-practice requirements mandating a credentialed professional.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI diagnostic support tools require substantial licensing, integration, and oversight infrastructure. When accounting for compliance, validation, and the nurse's continued involvement, the cost savings do not yet approach an order of magnitude compared to the nurse's labor.
Cost vs. human wageclaude-sonnet-52/5AI tools add cost as decision-support layered on top of nurse labor rather than replacing it, since human interpretation and legal responsibility remain necessary, so all-in cost savings are modest.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools exist for specific interpretations (radiology reports, lab result flagging), but deployed systems operate in narrow scopes with human oversight requirements. No production system autonomously orders, interprets, and evaluates tests across the full scope of nursing practice without significant human validation.
Technical feasibility todayclaude-sonnet-52/5Some deployed clinical decision-support tools flag abnormal labs or suggest tests, but no product autonomously orders, interprets, and evaluates diagnostics as a substitute for nurse judgment in production settings.

Modify patient treatment plans as indicated by patients' responses and conditions.

20

CI 20–20 · exposure 25 · augmentation 75 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While healthcare is digitizing, adoption of autonomous clinical decision-making is laggard due to regulatory caution, liability concerns, and physician resistance. Pilots of AI-assisted triage and monitoring exist, but autonomous modification of treatment plans remains rare in production healthcare systems.
Sector adoption velocityclaude-sonnet-52/5Healthcare overall is a slower-adopting, highly regulated sector; clinical decision-support tools are used as aids but full delegation of care plan changes to AI remains rare and cautious.
Augmentation potentialclaude-haiku-4-5-202510014/5AI systems that surface relevant clinical data, flag potential medication interactions, suggest evidence-based adjustments, and highlight concerning trends can significantly augment a nurse's ability to monitor and recommend changes. Current EHR-integrated tools and clinical decision-support systems already assist nurses in identifying when escalation or plan modification is warranted.
Augmentation potentialclaude-sonnet-54/5AI-based alerts, trend analysis, and decision-support systems meaningfully help nurses notice changes in patient status and consider treatment adjustments faster, while the nurse retains final judgment.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist in analyzing patient data and suggesting treatment adjustments, modifying treatment plans requires real-time clinical judgment, integration of subtle contextual patient information, and accountability for medical decisions. Current AI cannot reliably do this end-to-end with 50% time savings at equal quality because it lacks the ability to safely operate without physician oversight and cannot handle the full complexity of individual patient responses.
Task automatabilityclaude-sonnet-52/5This requires clinical judgment integrating physical assessment, patient history, and real-time observation, which current AI cannot reliably perform end-to-end; at most it can suggest options for a nurse to consider.ed
Adoption barriersclaude-haiku-4-5-202510015/5Registered nurses are licensed professionals; treatment plan modifications must be documented, validated, and signed off by licensed providers (usually physicians or nurse practitioners). Regulatory frameworks (FDA, state boards of nursing, hospital credentialing) and liability law require human accountability, creating hard barriers to autonomous AI substitution.
Adoption barriersclaude-sonnet-55/5Modifying treatment plans is a licensed clinical act with direct liability implications; nursing scope-of-practice laws and hospital protocols require a licensed professional to make and authorize such changes.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-based clinical decision support systems require expensive implementation, validation, integration with EHRs, and ongoing human oversight. When accounting for liability and the nurse salary cost, the all-in cost of an AI system is comparable to or exceeds the cost of a nurse performing this task.
Cost vs. human wageclaude-sonnet-52/5AI decision-support software has licensing and integration costs plus mandatory human oversight, so total cost is not dramatically cheaper than the marginal cost of a nurse making this judgment call.
Technical feasibility todayclaude-haiku-4-5-202510012/5Decision-support systems exist to flag treatment changes, but no deployed product reliably performs autonomous modification of treatment plans in production settings. Existing clinical AI tools are narrow in scope and require human clinician validation before implementation, meaning they support rather than perform the core task.
Technical feasibility todayclaude-sonnet-52/5Clinical decision support tools exist and flag abnormal values or suggest protocol adjustments, but no deployed product autonomously modifies treatment plans without nurse/physician oversight.

Consult and coordinate with healthcare team members to assess, plan, implement, or evaluate patient care plans.

18

CI 11–25 · exposure 20 · 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 adoption of AI for clinical decision support remains cautious and heavily regulated; while EHR automation and clinical documentation tools are spreading, actual AI-led care coordination in production at scale remains limited, with most efforts still in pilot or research stages.
Sector adoption velocityclaude-sonnet-52/5Healthcare is a highly regulated, relationship-driven sector with slow, cautious AI adoption for clinical coordination tasks despite growth in documentation tools.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can meaningfully assist nurses by aggregating patient data, flagging alerts, drafting documentation, and suggesting evidence-based care options, thereby raising nurse productivity in the coordination and information-synthesis parts of the task while the nurse retains clinical and team-leadership responsibility.
Augmentation potentialclaude-sonnet-53/5AI can assist by summarizing patient records, flagging changes, and drafting care plan updates, improving efficiency, but the core interpersonal coordination remains human-led.
Task automatabilityclaude-haiku-4-5-202510012/5Nurses use judgment, interpersonal communication, and contextual reasoning to assess and coordinate patient care—domains where AI today struggles with real-time nuance and accountability. While AI can help draft or organize information, end-to-end care plan coordination requires human-led clinical judgment and team synchronization that current systems cannot reliably replace.
Task automatabilityclaude-sonnet-52/5This requires real-time interpersonal coordination, clinical judgment, and physical presence with patients and staff, which current AI cannot perform end-to-end despite being able to draft summaries or flag data points.
Adoption barriersclaude-haiku-4-5-202510014/5Nursing care plan coordination is legally and professionally accountable to the registered nurse; liability, licensure requirements, patient safety regulations, and institutional protocols require a licensed nurse to own and sign off on care decisions, creating a hard legal barrier to substitution.
Adoption barriersclaude-sonnet-55/5Nursing licensure, scope-of-practice laws, and liability for patient care decisions legally require a qualified human to consult and coordinate care.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI integration for care coordination still requires nursing oversight, validation, and re-work; the cumulative cost of inference, system integration, and mandatory human review often exceeds the marginal cost savings, especially given liability and regulatory requirements that keep humans in the loop.
Cost vs. human wageclaude-sonnet-51/5Since AI cannot substitute for the coordination and consultation itself, the human cost remains necessary, making AI an add-on cost rather than a cheaper substitute.
Technical feasibility todayclaude-haiku-4-5-202510012/5Deployed AI products can generate documentation summaries and flag clinical data, but no production system reliably executes the full loop of assessment, plan formulation, team coordination, and care adjustment without human oversight. Real-world clinical coordinators remain rare and typically operate as narrow assistants rather than autonomous coordinators.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously consults with healthcare teams and coordinates patient care plans; AI is used only as a documentation or decision-support aid alongside human coordination.

Conduct specified laboratory tests.

17

CI 9–25 · exposure 20 · augmentation 50 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of AI for test conduction is minimal; laboratory automation exists but is not AI-driven displacement of nurses, and regulatory and clinical governance heavily limit any substitution pathway.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially bedside nursing tasks, remains a laggard in AI/robotic adoption for physical task execution compared to information-based professions.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-powered laboratory result interpretation, flagging of abnormal values, and decision support can assist nurses in understanding and acting on test outcomes, providing meaningful productivity gains in the analysis and communication phases of the task.
Augmentation potentialclaude-sonnet-53/5AI can assist with result interpretation, flagging abnormal values, and streamlining documentation, improving efficiency around the test process even though it doesn't replace the physical act.
Task automatabilityclaude-haiku-4-5-202510012/5Laboratory tests require precise specimen handling, procedural adherence, and hands-on clinical skills (phlebotomy, sample preparation). While AI can interpret some results post-analysis, the physical conduct of tests remains beyond current AI capabilities, and no meaningful subset achieves 50% time savings at equal quality.
Task automatabilityclaude-sonnet-52/5Physical specimen handling and many point-of-care tests require manual dexterity and hands-on manipulation that current AI cannot perform; automation exists only for lab analysis instruments, not the nurse's hands-on execution.
Adoption barriersclaude-haiku-4-5-202510015/5Registered nurse licensure and scope of practice legally mandate that clinical laboratory tests be ordered and often personally conducted or directly supervised by licensed healthcare professionals; regulatory frameworks (CLIA, state nursing boards) enforce this requirement.
Adoption barriersclaude-sonnet-54/5Clinical testing often requires licensed personnel for specimen collection, chain-of-custody, and interpretation, with regulatory and liability requirements around who may perform certain tests.
Cost vs. human wageclaude-haiku-4-5-202510011/5Dedicated laboratory analyzers exist but are costly capital equipment with ongoing maintenance and calibration. Their per-test cost plus infrastructure does not undercut the nurse labor cost for routine test administration at the task level described.
Cost vs. human wageclaude-sonnet-52/5Robotic or automated testing equipment can be costly to install and maintain versus a nurse performing quick point-of-care tests, so cost advantage is limited outside of high-volume centralized labs.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system reliably conducts physical laboratory tests end-to-end. Existing laboratory automation and POCT devices are narrow-scope instruments, not AI agents performing the full task of test conduction as a registered nurse would.
Technical feasibility todayclaude-sonnet-52/5Automated lab analyzers exist and are widely deployed for sample processing, but the nurse's role of physically conducting bedside/point-of-care tests is not replaced by any deployed AI product.

Work with individuals, groups, or families to plan or implement programs designed to improve the overall health of communities.

16

CI 7–25 · exposure 13 · augmentation 63 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI is growing but remains cautious; community health programs are typically delivered by larger health systems with slower IT integration, and regulatory/liability concerns slow deployment of autonomous or minimally supervised systems.
Sector adoption velocityclaude-sonnet-52/5Public health and community nursing settings are typically under-resourced and slower to adopt AI tools compared to fast-digitizing sectors like finance or tech.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can meaningfully assist nurses by analyzing community health data, identifying disparities, synthesizing evidence-based interventions, and drafting program proposals—enabling nurses to focus more on engagement, stakeholder coordination, and implementation while staying fully in control.
Augmentation potentialclaude-sonnet-53/5AI can help nurses analyze community health data, draft educational materials, identify at-risk populations, or generate program plans, meaningfully aiding parts of the task.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can support data analysis, literature review, and draft program designs, the task requires sustained engagement with communities, stakeholder negotiation, and adaptive implementation based on human relationships and trust—capabilities that current AI systems cannot perform end-to-end without substantial human oversight and involvement.
Task automatabilityclaude-sonnet-51/5This is a relational, community-organizing task requiring trust-building, local knowledge, and in-person engagement that current AI cannot substitute for end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Registered nurses are licensed professionals, and community health program implementation carries liability, regulatory requirements around program approval, and a fundamental requirement for human professional judgment and accountability in public health contexts.
Adoption barriersclaude-sonnet-54/5Nursing licensure, scope-of-practice rules, and community trust/accountability requirements mean a credentialed human nurse must lead this work, though some administrative parts face fewer barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for data support and program drafting are relatively inexpensive, but the core work—community engagement, stakeholder management, adaptive implementation—still requires highly trained nurses whose labor cost far exceeds the marginal AI cost of assistive tools.
Cost vs. human wageclaude-sonnet-51/5AI cannot perform the core relational and implementation work, so there is no meaningful AI cost basis to compare against the human wage for full task completion.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs community health program planning and implementation independently; AI tools exist for supporting components (data analysis, resource mapping) but production systems do not operate autonomously in this complex, relationship-dependent domain.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product plans or implements community health programs with populations; AI at best supports data analysis or documentation adjacent to this work.

Monitor all aspects of patient care, including diet and physical activity.

14

CI 3–25 · exposure 13 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI for clinical monitoring remains slow and pilot-heavy; hospitals deploy narrow tools (alarms, charting aides) but have not replaced nurse monitoring with autonomous systems due to liability, regulatory caution, and clinical culture. Adoption remains institutional and incremental.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially bedside nursing, has historically slow AI adoption for direct patient care tasks despite faster uptake in administrative and documentation areas.
Augmentation potentialclaude-haiku-4-5-202510014/5AI dashboards, predictive alerts, and automated data aggregation substantially augment nurse productivity by surfacing trends, freeing time for direct patient contact and enabling faster response to deterioration. Many hospitals now use these tools to enhance rather than replace bedside monitoring.
Augmentation potentialclaude-sonnet-53/5AI-enabled monitoring devices, EHR alerts, and wearable sensors can flag anomalies in vitals, diet logs, or activity levels, helping nurses prioritize attention, though human oversight remains central.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with data monitoring (vital signs, activity logs, dietary intake tracking) but cannot independently perform holistic patient care monitoring that requires clinical judgment, patient interaction, and contextual decision-making in response to observed changes. The task demands real-time responsiveness and relationship-based care assessment that current systems cannot reliably deliver without constant human oversight.
Task automatabilityclaude-sonnet-51/5This requires continuous physical presence, hands-on assessment, and real-time clinical judgment across a wide range of unpredictable patient conditions that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: nursing practice acts legally require a licensed nurse to assess and monitor patients, patient safety liability falls on the nurse, and healthcare regulation mandates human professional judgment in care decisions. Patient and institutional trust in human oversight further protects this role.
Adoption barriersclaude-sonnet-55/5Nursing licensure, scope-of-practice laws, and liability for patient safety require a licensed human to perform and be accountable for this monitoring.
Cost vs. human wageclaude-haiku-4-5-202510012/5Monitoring AI and wearables reduce some documentation burden, but a nurse's loaded compensation includes expertise and liability that AI integration does not replace at lower cost; the overhead of oversight, integration, and validation often exceeds the time saved for non-routine monitoring.
Cost vs. human wageclaude-sonnet-51/5Since AI cannot substitute for the core task, there is no viable cost comparison—human nursing labor remains necessary and AI adds cost as a supplementary tool rather than a replacement.
Technical feasibility todayclaude-haiku-4-5-202510012/5Monitoring tools and EMR dashboards exist and flag certain metrics, but no deployed product reliably monitors all aspects of patient care (diet, activity, clinical status, psychosocial factors) with the judgment required for safe independent operation. Existing systems are narrow, generate false positives, and require nurse validation on critical decisions.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously monitors comprehensive patient care including diet and activity; existing tools only support narrow data logging or alerts, not holistic monitoring.

Direct or coordinate infection control programs, advising or consulting with specified personnel about necessary precautions.

14

CI 3–25 · exposure 13 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare AI adoption is moderate and heavily regulated; while some hospitals use analytics for surveillance, active direction and coordination of infection control remains a human-led function with slow automation uptake in practice.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially clinical operations and infection control, is a slower-adopting sector for autonomous AI decision-making due to regulatory and safety constraints, though AI-assisted surveillance tools are emerging.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can meaningfully assist nurses by flagging outbreak signals, suggesting protocol updates based on data, and automating documentation—raising coordinators' productivity while they retain oversight and final decision-making authority.
Augmentation potentialclaude-sonnet-53/5AI can assist by analyzing infection surveillance data, flagging outbreak patterns, and summarizing guidelines, which helps inform the nurse's decisions without replacing the coordination role.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with infection control protocols and flagging data anomalies, directing and coordinating programs requires real-time judgment about personnel, policy adaptation, and stakeholder consultation that demand human authority and contextual judgment. Current AI cannot reliably replace the consultative and coordinative components end-to-end.
Task automatabilityclaude-sonnet-51/5This task requires clinical judgment, situational authority, and real-time coordination across departments that current AI cannot perform end-to-end.the human role of directing and consulting is inherently interactive and context-dependent.
Adoption barriersclaude-haiku-4-5-202510014/5Infection control programs carry significant regulatory requirements (OSHA, CDC, Joint Commission standards) and liability for outbreaks; human credentialing, professional judgment, and legal accountability for policy decisions create substantial adoption barriers.
Adoption barriersclaude-sonnet-55/5Infection control leadership typically requires a licensed, accountable clinical professional (e.g., RN or infection preventionist) with legal and regulatory responsibility, making this a hard-barrier task.
Cost vs. human wageclaude-haiku-4-5-202510012/5The oversight and coordination costs of AI-assisted infection control—requiring continued expert review, integration with hospital systems, and human decision-making—remain comparable to or exceed the cost of a nurse coordinator handling these tasks directly.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this coordination role, so cost comparison favors the human entirely; any AI use would be supplementary, not substitutive.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed products reliably direct infection control programs autonomously; AI tools exist for surveillance and protocol drafting but production systems require human epidemiologists and infection preventionists to make final decisions and coordinate across departments.
Technical feasibility todayclaude-sonnet-51/5No deployed product directs or coordinates infection control programs autonomously; AI exists only as decision-support or documentation tools, not as the coordinating authority.

Prescribe or recommend drugs, medical devices, or other forms of treatment, such as physical therapy, inhalation therapy, or related therapeutic procedures.

13

CI 5–20 · exposure 17 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Regulatory barriers and liability concerns mean healthcare organizations show minimal adoption of autonomous prescription or treatment recommendation systems. Adoption remains confined to narrow clinical decision support roles with human oversight retained.
Sector adoption velocityclaude-sonnet-52/5Healthcare overall has been a slower adopter of AI for high-stakes clinical decisions like prescribing due to regulatory, safety, and liability concerns, despite faster uptake in administrative healthcare tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist nurses by surfacing drug interactions, side effect profiles, alternative treatment literature, and patient-specific contraindications, helping them make better informed recommendations within their scope of practice. However, the augmentation is bounded by the nurse's legal authority rather than transformative.
Augmentation potentialclaude-sonnet-54/5AI clinical decision support and diagnostic aids meaningfully help nurses and prescribers by surfacing drug interactions, guidelines, and treatment options, improving speed and accuracy while the clinician retains final authority.
Task automatabilityclaude-haiku-4-5-202510011/5Prescribing and recommending drugs or medical devices requires clinical judgment, patient history synthesis, contraindication assessment, and legal authorization that current AI systems cannot perform end-to-end. While AI can assist in drug interaction checking or literature retrieval, only licensed humans can legally issue prescriptions in most jurisdictions.
Task automatabilityclaude-sonnet-52/5Prescribing requires clinical judgment, patient examination, and legal accountability that current AI cannot fully replicate end-to-end; AI can suggest options but cannot independently complete the task at equal quality with major time savings.
Adoption barriersclaude-haiku-4-5-202510015/5Prescribing is legally restricted to licensed practitioners in virtually all jurisdictions; regulatory bodies (FDA, state medical boards) explicitly govern who can authorize medications and treatments. Liability asymmetry is extreme: incorrect recommendations cause patient harm for which organizations and practitioners are legally accountable.
Adoption barriersclaude-sonnet-55/5Prescribing is legally restricted to licensed practitioners (nurses with prescriptive authority, physicians, etc.) with strict scope-of-practice and liability regulations, making this a hard-barrier task.
Cost vs. human wageclaude-haiku-4-5-202510011/5The specialized clinical judgment, legal liability, and regulatory requirements mean any AI system capable of this task would require substantial validation, governance, and human oversight infrastructure—far exceeding the cost savings of automating the recommendation itself.
Cost vs. human wageclaude-sonnet-52/5AI decision-support tools are relatively cheap to run, but the requirement for licensed oversight, integration into EHRs, and liability management keeps all-in costs comparable to or only modestly below human-only workflows.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product independently prescribes or recommends therapeutic interventions for nurses. Clinical decision support tools exist (e.g., drug interaction checkers), but they inform rather than replace human prescription authority and require expert oversight.
Technical feasibility todayclaude-sonnet-52/5Clinical decision support tools exist and are used to suggest treatments, but no deployed product autonomously prescribes or recommends treatment without licensed provider review and sign-off.

Provide health care, first aid, immunizations, or assistance in convalescence or rehabilitation in locations such as schools, hospitals, or industry.

9

CI 3–16 · exposure 13 · augmentation 63 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare sectors have piloted AI documentation and monitoring tools, but automation of core nursing tasks (first aid, immunizations, physical care, rehabilitation assistance) has seen minimal real-world displacement; hospitals and clinics remain heavily dependent on human nursing staff.
Sector adoption velocityclaude-sonnet-52/5Healthcare delivery of hands-on physical care remains a laggard sector for AI substitution, though administrative/documentation aspects of nursing are seeing pilots.
Augmentation potentialclaude-haiku-4-5-202510014/5AI significantly augments nursing productivity through real-time vital-sign monitoring, clinical decision support, automated documentation, patient education platforms, and rehabilitation exercise guidance—all while the nurse remains responsible for patient care and safety.
Augmentation potentialclaude-sonnet-53/5AI can assist with care documentation, triage decision support, and monitoring alerts, but the physical care delivery itself receives little direct augmentation.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with triaging, immunization scheduling, and rehabilitation exercise monitoring, the task requires hands-on physical care (injections, wound care, patient handling), clinical judgment in real-time patient assessment, and emotional support—none of which can be fully automated today. Most work remains human-dependent.
Task automatabilityclaude-sonnet-51/5This task requires hands-on physical care, administering injections, wound care, and physical assessment that current AI cannot perform without embodiment.
Adoption barriersclaude-haiku-4-5-202510015/5Nursing tasks are heavily regulated; many require a licensed Registered Nurse to legally perform (medication administration, injections, assessment). Scope-of-practice laws, liability for patient harm, and the physical human-contact requirement create hard legal and regulatory barriers to substitution.
Adoption barriersclaude-sonnet-55/5Licensure requirements, legal liability for medical care, and mandatory human presence for administering treatments like immunizations create hard regulatory and legal barriers.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI inference costs for supporting tasks (triage support, documentation) are negligible compared to nurse wages, but the automation does not reduce the need for nurses in patient-facing roles, making the per-task cost ratio unfavorable—human nurses remain necessary and expensive.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical labor involved, so there is no comparable AI cost basis; a human nurse remains necessary for all physical care components.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed AI system can independently deliver first aid, administer injections, monitor patients during convalescence, or provide the physical assistance nurses deliver. AI tools exist for documentation and scheduling, but the core clinical and physical tasks remain human-performed in all production healthcare settings.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product provides physical first aid, immunizations, or hands-on rehabilitation assistance; this remains entirely research-stage or nonexistent for physical care delivery.

Prepare patients for and assist with examinations or treatments.

4

CI 0–9 · exposure 8 · 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 remains heavily regulated and conservative in automation of direct patient care tasks. Adoption of AI in this specific hands-on nursing function is minimal; human nursing presence is a core operational requirement.
Sector adoption velocityclaude-sonnet-51/5Healthcare's hands-on clinical tasks are among the slowest to adopt AI due to physical, regulatory, and safety constraints, despite AI adoption in administrative/documentation aspects of healthcare.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by providing preparation checklists, managing patient workflows, flagging contraindications, or offering real-time procedure guidance, but the nurse remains essential for physical assistance and clinical judgment.
Augmentation potentialclaude-sonnet-52/5AI can support scheduling, checklists, or information retrieval around the task, but offers minimal direct assistance to the physical act of preparing and assisting patients during exams.
Task automatabilityclaude-haiku-4-5-202510012/5While AI systems could potentially automate communication about preparation steps and coordinate scheduling, the core task involves physical assistance, patient positioning, and real-time clinical judgment that require human presence. Current AI cannot physically assist patients or adapt to unpredictable patient responses in real-time.
Task automatabilityclaude-sonnet-51/5This task requires physical handling of patients, positioning, applying equipment, and hands-on assistance during procedures—none of which current AI systems (software/language models) can perform without embodiment.
Adoption barriersclaude-haiku-4-5-202510015/5Regulatory frameworks, licensure requirements, and patient safety standards mandate that registered nurses or other licensed professionals perform or directly oversee patient preparation and assistance. Legal and liability concerns create hard barriers to automation.
Adoption barriersclaude-sonnet-55/5Direct patient contact, safety, and clinical judgment during preparation and treatment require licensed nursing oversight, with strong regulatory, liability, and clinical safety barriers preventing automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task fundamentally requires human physical presence and clinical decision-making, so AI cannot reduce the cost of the primary care delivery—any AI support (scheduling, reminders) addresses only peripheral components.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical task, so any comparison favors the human nurse by default; robotic alternatives would be far more costly than nursing wages.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product reliably performs physical patient assistance or real-time clinical triage during examinations. While chatbots can provide preparation instructions, they cannot deliver the embodied care this task requires.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product physically prepares or assists patients during exams; this remains entirely research-stage in robotics with no production nursing deployments.

Administer medications to patients and monitor patients for reactions or side effects.

4

CI 0–7 · exposure 5 · augmentation 50 · importance 4.7/5 · 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 clinical-task automation due to safety requirements, regulatory constraints, and the direct human-contact mandate. Production deployment of AI for medication administration is virtually nonexistent.
Sector adoption velocityclaude-sonnet-52/5Healthcare bedside care remains a slow-adopting, highly regulated, physically-dependent sector where AI use is largely confined to documentation and decision support, not task execution.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist nurses by flagging drug interactions, monitoring vital signs for adverse reactions, and alerting to potential side effects via EHR integration, improving safety and reducing cognitive load on parts of the monitoring workflow.
Augmentation potentialclaude-sonnet-53/5AI-enabled systems (smart infusion pumps, EHR alerts, decision-support tools for drug interactions and dosing) meaningfully assist nurses in safely administering and monitoring medications, though the core task remains human-performed.
Task automatabilityclaude-haiku-4-5-202510011/5Administering medications requires physical handling of drugs, establishing IV lines, or other hands-on clinical procedures that current AI cannot perform. While monitoring could be partially automated, the core task—actually delivering medication—is inherently manual and requires human presence.
Task automatabilityclaude-sonnet-51/5Physically administering medications (injections, IV lines, oral meds with verification) and hands-on monitoring for reactions requires physical presence, dexterity, and real-time clinical judgment that no current AI system can perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Medication administration is legally restricted to licensed healthcare professionals (RNs, LPNs, physicians, or other licensed providers) in virtually all jurisdictions. Liability, regulatory oversight, and licensure create hard barriers to automation or delegation to unlicensed systems.
Adoption barriersclaude-sonnet-55/5Medication administration is legally restricted to licensed nurses/clinicians, with strict liability, scope-of-practice laws, and safety regulations requiring human accountability.
Cost vs. human wageclaude-haiku-4-5-202510011/5The physical and direct-care components of medication administration cannot be cost-effectively replaced by AI systems; human nurses remain necessary. Any AI monitoring tools add cost rather than replace the primary expense of nursing labor.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical task, so cost comparison favors the human nurse entirely; robotics for medication administration remain experimental and costly.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI systems can assist with medication monitoring through EHR data analysis and alert systems, but no deployed product can independently administer medications or reliably perform the full task end-to-end. Monitoring alone exists but remains supervised by human nurses.
Technical feasibility todayclaude-sonnet-51/5No deployed product administers medication autonomously or performs physical patient monitoring for reactions; existing tech (smart pumps, barcode scanning) only assists a human nurse, not automates the task.

Perform physical examinations, make tentative diagnoses, and treat patients en route to hospitals or at disaster site triage centers.

4

CI 0–7 · exposure 5 · augmentation 38 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Healthcare adoption of AI is selective and cautious; triage and emergency care remain under tight clinical governance. While digital triage tools are piloted, autonomous AI diagnosis and treatment at disaster sites sees minimal real-world deployment.
Sector adoption velocityclaude-sonnet-51/5Emergency and disaster response medicine is a highly physical, low-digitization environment with minimal AI agent deployment for hands-on care.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-driven diagnostic decision support, vital-sign monitoring alerts, and protocol reminders can assist nurses in faster differential diagnosis and triage decisions, but the core task of physical examination and patient interaction remains human-led.
Augmentation potentialclaude-sonnet-52/5AI can support triage protocols, decision support, or documentation but offers limited real-time assistance amid the chaotic, physical demands of disaster-site care.
Task automatabilityclaude-haiku-4-5-202510011/5Physical examinations require hands-on assessment (palpation, auscultation, vital-sign measurement) that current AI cannot perform without specialized robotic hardware. Tentative diagnosis and triage treatment decisions demand real-time clinical judgment under uncertain, rapidly changing conditions that AI cannot reliably execute end-to-end today.
Task automatabilityclaude-sonnet-51/5This requires hands-on physical examination, real-time judgment under chaotic conditions, and direct patient treatment (e.g., wound care, stabilization) that current AI cannot physically perform.
Adoption barriersclaude-haiku-4-5-202510015/5Nursing practice is strictly regulated; diagnosis and treatment en route or at triage sites are legally scoped to licensed healthcare professionals. Liability, patient safety, and statutory medical practice laws create hard barriers to substitution with AI.
Adoption barriersclaude-sonnet-55/5Licensure, scope-of-practice laws, and liability for medical diagnosis and treatment strictly require a qualified human provider, especially in emergency/disaster contexts.
Cost vs. human wageclaude-haiku-4-5-202510011/5Performing physical exams and administering treatment at emergency sites requires embodied presence and real-time intervention; remote AI systems or diagnostic software provide only partial support and cannot substitute for the cost of human deployment.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical labor and clinical presence required, so there is no meaningful cost comparison—human presence is mandatory.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI can assist with symptom-checkers and diagnostic decision support in controlled settings, but no deployed system reliably performs full physical examinations, makes autonomous diagnoses, or delivers treatments in dynamic emergency or disaster contexts where a nurse must act independently.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs hands-on triage, physical exams, or emergency treatment in mobile/disaster settings; this remains firmly in the domain of trained human responders.

Consult with institutions or associations regarding issues or concerns relevant to the practice and profession of nursing.

4

CI 0–7 · exposure 0 · augmentation 38 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Institutional and professional association reliance on nursing expertise for consultations remains fundamentally centered on licensed human professionals; there is no evidence of AI adoption for this trust-dependent, legally-accountable role.
Sector adoption velocityclaude-sonnet-52/5Healthcare institutions adopt AI slowly for administrative and clinical support, but this specific advocacy/consultation task shows little to no AI deployment.
Augmentation potentialclaude-haiku-4-5-202510012/5AI tools might assist a nurse consultant by drafting research summaries or organizing background data, but the core task of professional consultation requires human judgment and professional standing, limiting augmentation value.
Augmentation potentialclaude-sonnet-53/5AI can help nurses research policy issues, draft position statements, summarize regulatory changes, or prepare talking points, providing moderate support for the consultation process.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires human judgment, professional standing, and relational negotiation with institutional stakeholders. AI cannot credibly represent professional nursing interests or serve as a trusted consultant to institutions in matters of professional practice.
Task automatabilityclaude-sonnet-51/5This task requires professional judgment, relationship-building, and representing nursing interests in institutional dialogue—an interpersonal, advocacy-based activity that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Nursing licensure and professional ethics requirements mandate that a licensed, accountable human nurse represent the profession in institutional consultations; regulatory and liability frameworks prevent AI substitution in this advisory role.
Adoption barriersclaude-sonnet-54/5Professional representation and consultation on nursing practice typically requires credentialed expertise and institutional trust, creating strong organizational and professional barriers to AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task requires a registered nurse with professional credibility and authority; any AI-generated analysis would still need human nursing leadership to validate and present it, making the all-in cost roughly equivalent to hiring the human consultant.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this consultative role, so no meaningful cost comparison favors AI over the human nurse for this function.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can independently conduct institutional consultations on nursing practice issues; this requires licensed professional authority, accountability, and the ability to negotiate complex stakeholder relationships that exceed current AI capability.
Technical feasibility todayclaude-sonnet-51/5No deployed product substitutes for a nurse consulting with institutions or associations on professional practice issues; this remains a human relational and advisory function.

Observe nurses and visit patients to ensure proper nursing care.

1

CI 0–3 · exposure 0 · augmentation 38 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare adoption of AI remains cautious and regulated; direct patient care oversight is among the most protected clinical tasks and shows minimal displacement by automation.
Sector adoption velocityclaude-sonnet-52/5Healthcare adoption of AI is growing but remains slow for hands-on clinical oversight roles due to regulatory, safety, and workflow constraints.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with documentation or alert systems based on vital signs, but cannot replace the core act of bedside observation and clinical judgment that a nurse performs during patient visits.
Augmentation potentialclaude-sonnet-53/5AI can assist with documentation review, flagging anomalies in patient records, or scheduling rounds, but the core observational/supervisory judgment remains human-driven.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires direct human presence, physical assessment of patients, and real-time clinical judgment. Current AI systems cannot physically visit patients or perform the nuanced observations that warrant clinical intervention decisions.
Task automatabilityclaude-sonnet-51/5Direct observation of nurses' practice and in-person patient visits require physical presence, clinical judgment, and interpersonal assessment that current AI cannot replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Healthcare regulations, duty of care, patient safety laws, and licensure requirements mandate that a qualified human (RN) must directly observe and assess patients; automation is legally and ethically precluded.
Adoption barriersclaude-sonnet-55/5Nursing supervision and quality-of-care assurance require licensed personnel accountable under clinical governance and regulatory/legal frameworks; a machine cannot legally sign off on care quality.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI systems cannot perform the core function of physical patient visitation and real-time clinical assessment, making cost comparison moot—there is no feasible AI alternative to the human nursing task.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this supervisory/visitation task, so cost comparison favors the human entirely.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product can autonomously visit and observe patients to ensure proper nursing care; this fundamentally requires human presence and continuous bedside judgment in clinical settings.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs supervisory clinical observation or patient rounding autonomously; this remains a human oversight function in every clinical setting.

Inform physician of patient's condition during anesthesia.

1

CI 0–3 · exposure 0 · augmentation 38 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare, particularly operating rooms and anesthesia settings, remains highly resistant to autonomous decision-making in critical care. Adoption of AI in these roles is minimal due to safety, liability, and regulatory requirements that mandate licensed human judgment.
Sector adoption velocityclaude-sonnet-52/5Healthcare, especially perioperative/anesthesia care, adopts AI cautiously due to safety-critical nature, regulatory oversight, and the need for human presence at bedside.
Augmentation potentialclaude-haiku-4-5-202510012/5While monitoring devices and dashboards can assist nurses in tracking vital signs and alerting to changes, AI augmentation is limited because the core task—communicating clinical judgment to a physician—inherently requires the nurse's professional assessment and responsibility.
Augmentation potentialclaude-sonnet-53/5AI-enabled monitoring systems and alert algorithms can help flag anomalies in vital signs, aiding the nurse's situational awareness, though the core reporting and judgment remain human-driven.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time clinical judgment, human-to-human communication in a high-stakes operating room environment, and the ability to convey nuanced patient physiological changes to a physician. Current AI systems cannot reliably perform or replace this direct communication and assessment responsibility.
Task automatabilityclaude-sonnet-51/5This requires real-time physical monitoring, clinical judgment, and verbal communication during a live surgical/anesthesia event, which no current AI system can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and regulatory barriers exist: nursing licensure is required, physicians must maintain direct clinical responsibility for patient care during anesthesia, and liability for adverse outcomes rests on licensed providers who must personally assess patient status.
Adoption barriersclaude-sonnet-55/5This is a licensed clinical function with direct patient safety and legal liability implications, requiring a credentialed nurse or physician to perform and communicate in real time.
Cost vs. human wageclaude-haiku-4-5-202510011/5An AI system capable of monitoring and communicating patient status during surgery would require extensive custom infrastructure, continuous oversight, and liability coverage that would far exceed the cost of a nurse's labor for this task.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this task, so cost comparison favors the human by default; any AI monitoring aids still require full nursing staffing and cost.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs the core function of a nurse informing a physician of patient condition during anesthesia. This requires continuous bedside monitoring, clinical interpretation, and real-time communication that remains firmly in human territory in clinical practice.
Technical feasibility todayclaude-sonnet-51/5No deployed product independently monitors and reports patient status to a physician during anesthesia; this remains a core nursing responsibility with no automation in production.

Direct or supervise less-skilled nursing or healthcare personnel or supervise a particular unit.

0

CI 0–0 · 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 organizations have shown no meaningful adoption of AI for core supervisory functions; leadership roles remain deeply human-centric and resistant to automation due to legal, liability, and relational requirements.
Sector adoption velocityclaude-sonnet-51/5Healthcare management and clinical supervision show minimal AI adoption for direct personnel oversight, reflecting the sector's cautious, high-liability environment.
Augmentation potentialclaude-haiku-4-5-202510012/5AI tools could assist with scheduling optimization or performance data aggregation, but supervision fundamentally involves judgment calls, personnel feedback, and accountability that require sustained human presence and authority.
Augmentation potentialclaude-sonnet-52/5AI can assist with scheduling, documentation, or performance tracking data, but offers limited support for the interpersonal and judgment-based aspects of supervision.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time interpersonal judgment, conflict resolution, performance feedback, scheduling decisions, and accountability for staff outcomes—functions that demand human discretion, emotional intelligence, and organizational authority that current AI cannot exercise.
Task automatabilityclaude-sonnet-51/5Supervising staff involves real-time interpersonal leadership, delegation, performance evaluation, and accountability that require physical presence and clinical judgment AI cannot replicate.
Adoption barriersclaude-haiku-4-5-202510015/5Nursing supervision is a licensed responsibility; regulatory frameworks and institutional hierarchy require a qualified human RN to hold supervisory authority and be legally accountable for unit operations and personnel management.
Adoption barriersclaude-sonnet-55/5Nursing supervision requires licensure, legal accountability for patient safety, and regulatory/organizational structures mandating a qualified human supervisor.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI solutions for supervisory support (if they existed at scale) would still require human supervisors in the loop for decision-making, making the combined cost higher than a human supervisor working alone.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this supervisory role, so cost comparison favors the human by default.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can legally or reliably direct or supervise healthcare personnel in production environments; this requires licensed accountability and decision-making authority that remains exclusively with human supervisors.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product manages or supervises healthcare personnel; this remains an entirely human management function in practice.

Administer local, inhalation, intravenous, or other anesthetics.

0

CI 0–0 · 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 sectors remain heavily reliant on human clinical judgment for anesthesia; automation adoption in this domain is negligible. Regulatory requirements, patient safety liability, and the need for continuous clinical decision-making make rapid AI substitution infeasible.
Sector adoption velocityclaude-sonnet-51/5Healthcare's physical, high-stakes procedural tasks show minimal AI adoption; anesthesia administration itself has essentially zero automation deployment in practice.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with pre-operative planning, monitoring alerts, or documentation, but current systems cannot augment the core task of safely administering and titrating anesthetics in real time. The human clinician must remain in direct control throughout.
Augmentation potentialclaude-sonnet-52/5AI can support anesthesia-related monitoring, dosing calculators, or decision support tools, but it offers limited direct assistance to the physical act of administering anesthesia itself.
Task automatabilityclaude-haiku-4-5-202510011/5Administering anesthetics requires direct patient contact, real-time monitoring of physiological responses, and immediate clinical judgment to adjust dosages and manage complications. Current AI systems cannot physically deliver medications or make split-second clinical decisions that depend on continuous patient assessment.
Task automatabilityclaude-sonnet-51/5This is a hands-on, high-risk clinical procedure requiring physical manipulation of patients, real-time physiological monitoring, and split-second judgment; no AI system today can perform this end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Anesthesia administration is legally restricted to licensed professionals (nurses, physicians, or anesthesiologists) and subject to strict regulatory oversight (DEA, state boards, hospital credentialing). Direct patient contact and medication handling create hard legal and liability barriers to substitution.
Adoption barriersclaude-sonnet-55/5Administering anesthesia is tightly regulated and requires licensure (RN/CRNA/anesthesiologist), with severe liability and safety consequences, making this among the most legally protected clinical tasks.
Cost vs. human wageclaude-haiku-4-5-202510011/5The infrastructure, integration, and regulatory validation required to deploy autonomous anesthesia systems would exceed the cost of employing trained anesthesia nurses or anesthesiologists by orders of magnitude, with unacceptable liability and error-cost asymmetry.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical task, so cost comparison favors the human by default since AI cannot deliver the output at all.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product can independently administer anesthetics; this task is exclusively performed by licensed anesthesia professionals in real-world settings. Research into robotic drug delivery exists but remains limited to controlled laboratory or simulation environments.
Technical feasibility todayclaude-sonnet-51/5No deployed product administers anesthesia autonomously; this remains a physical, licensed clinical act performed by humans with no commercial automation.

Related occupations — Healthcare Practitioners & Technical

How to read this

A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.

What would change this score

New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.