Key findings
- Registered nurses (ISCO 22): AI exposure 4.0/10 - below average, physical care and clinical judgment protect the role
- 3.47 million registered nurses in the US at median $97,550/yr (BLS OOH, 2025 reference)
- Globally: 28 million nurses and midwives (WHO Global Health Workforce Statistics 2024)
- Highest-risk nursing sub-role: triage nurses and clinical documentation specialists (~6.0/10)
- Lowest-risk: ICU and surgical nurses - procedural, high-acuity, physical presence required (3.0/10)
Why nursing scores below the global average
The ISCO 22 group - health professionals, which includes nurses, midwives, and allied health practitioners - scores 5.0/10 on AI exposure in aggregate. But registered nursing sits meaningfully below that group average at 4.0/10. The gap reflects what nursing actually consists of in practice.
A nurse's shift involves physical assessment (palpation, auscultation, observation of patient appearance and movement), medication administration, wound care, patient repositioning, emotional support across acute distress situations, and rapid judgment under incomplete information - often with family members present asking questions. None of these tasks are text or data manipulation. They require a human body, physical dexterity, and the kind of contextual judgment that degrades when the situation departs from any expected pattern.
The tasks within nursing that are AI-susceptible - shift handover notes, medication reconciliation documentation, care plan updates, ordering from standard protocols - are being automated now by AI documentation tools like Abridge and Nuance DAX. But this automation removes administrative burden from nurses rather than nurses from wards. The clinical hours freed are reabsorbed by the care workload immediately, given the persistent global nursing shortage documented in the WHO Global Health Workforce Statistics 2024.
AI exposure by nursing sub-role
| Role | AI score | Robotics | US median wage |
|---|---|---|---|
| Triage nurses (ED, urgent care) | 6.0/10 | 2.0/10 | $90,400 |
| Clinical documentation specialists | 6.5/10 | 1.5/10 | $55,000 |
| Registered nurses (general) | 4.0/10 | 2.5/10 | $97,550 |
| Community and public health nurses | 4.5/10 | 1.5/10 | $78,200 |
| ICU and critical care nurses | 3.0/10 | 3.0/10 | $96,300 |
| Surgical / perioperative nurses | 3.0/10 | 4.0/10 | $89,400 |
| Nurse practitioners (advanced practice) | 3.5/10 | 1.5/10 | $134,920 |
| Nursing assistants / aides (ISCO 53) | 2.0/10 | 2.5/10 | $41,870 |
Source: BLS Occupational Outlook Handbook (OOH), 2025 reference (wage data). AI scores are WorldJobsData estimates per sub-role task bundle. ISCO-08 group 22 aggregate score is 5.0/10 across all health professionals; nursing sub-roles shown above reflect nursing-specific task analysis.
Why triage nurses score higher - and what that means
Triage - the initial assessment of patient acuity when they present at an emergency or urgent care facility - is the one nursing function where AI has documented, deployed capability. The Manchester Triage System and Emergency Severity Index are algorithmic frameworks that nurses apply; AI can match or exceed nurse performance on these structured scoring tasks in controlled settings (JAMA Internal Medicine, 2023).
That matters because triage is the gatekeeping function of an emergency department. If AI triage is accepted clinically and legally (the liability question is not yet resolved in most jurisdictions), the volume of nurses needed at intake changes. But we expect this to be a gradual transition rather than a 2026 event (our estimate, not a model forecast). Current AI triage tools work as decision-support aids alongside nurses, not replacements for them. The documentation role within triage - recording chief complaint, vitals, preliminary assessment in structured fields - is already being automated.
Clinical documentation specialists are a distinct sub-role created specifically to handle the administrative burden that registered nurses generated but found non-clinical. These roles, scoring 6.5/10, are genuinely at risk as AI scribes handle structured note-taking with increasing accuracy. BLS does not count these separately; they fall under medical records and health information technicians at a median of $55,000.
The global nursing shortage changes the displacement calculus
The WHO Global Health Workforce Statistics 2024 estimates a global shortfall of 5.9 million nurses, concentrated in sub-Saharan Africa and South and Southeast Asia. This shortage has a structural implication for AI displacement: even if AI increases nursing productivity by 20-30% by automating administrative tasks, the unmet clinical demand absorbs that capacity entirely. Displacement requires surplus supply. In nursing, there is no surplus.
In higher-income countries the picture is more nuanced. In the US, BLS projects registered nurse employment to grow 6% through 2032 - faster than average - driven by an aging population. In countries with ageing demographics and flat healthcare budgets (Japan, Germany, Italy), AI-assisted nursing is more likely to be framed as a way to extend existing nurse capacity than to reduce headcount.
Explore nursing workforce data by country
Nursing workforce density, health professional concentration, and AI exposure scores vary significantly across 206 countries in the WorldJobsData dataset.
Was this comparison useful?
Let us know what you think - your reaction helps us understand what to cover next.
Thanks for your reaction!
Get new country analyses, data updates, and AI labour market insights. No spam - one email when something worth reading drops.
Methodology
AI exposure scores are WorldJobsData estimates per ISCO-08 occupation group, derived from ILO ILOSTAT data (CC BY 4.0) and cross-referenced against Frey-Osborne (2013), OECD (2019), and IMF (2024) task-automation research. Sub-role scores for nursing are derived from task-level analysis within ISCO 22. Employment and wage figures for the US are from the BLS Occupational Outlook Handbook (OOH), 2025 reference. Global nursing workforce figures are from WHO Global Health Workforce Statistics 2024. Scores reflect AI exposure of the task bundle, not a prediction of job loss timing.
AI exposure, robotics, offshoring and work-from-home scores are model-estimated: an AI language model (Claude Opus 4.7) scored each ISCO-08 occupation group on 28 May 2026, guided by published research. They are informed judgements, not measurements, and not forecasts of job losses. "Risk velocity" and its time windows come from a digital-infrastructure index (broadband and secure servers), not from occupations or measured AI adoption. Read the full methodology and how the scores were checked.
Frequently asked questions
Are nurses safe from AI replacement in 2026?
How many nurses could be affected by AI in the US?
Which nursing roles face the most AI risk?
Where does the nursing AI risk data come from?
Data sources
- BLS Occupational Outlook Handbook (OOH), 2025 reference
- WHO Global Health Workforce Statistics 2024
- ILO ILOSTAT Labour Force Statistics 2024 (CC BY 4.0)
- JAMA Internal Medicine - AI triage performance study 2023
- Frey & Osborne (2013) "The Future of Employment"
- OECD Employment Outlook 2019
- IMF Staff Discussion Note SDN/2024/001
Spotted an error or know a better source? Report it and we will check it against the source data.