Personal Care Aides
31-1122.00Rank #723 of 923 scored · top 78% by substitution
Provide personalized assistance to individuals with disabilities or illness who require help with personal care and activities of daily living support (e.g., feeding, bathing, dressing, grooming, toileting, and ambulation). May also provide help with tasks such as preparing meals, doing light housekeeping, and doing laundry. Work is performed in various settings depending on the needs of the care recipient and may include locations such as their home, place of work, out in the community, or at a daytime nonresidential facility.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution18
Exposure15
Augmentation42
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
11 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%16
panel mean rating 1.6/5 → substitution pressure 16/100
Technical feasibility todayw 20%14
panel mean rating 1.6/5 → substitution pressure 14/100
Cost vs. human wagew 15%18
panel mean rating 1.7/5 → substitution pressure 18/100
Adoption barriersw 20%inverted — strong barriers lower the score30
panel mean rating 3.8/5 (barrier strength) → substitution pressure 30/100
Sector adoption velocityw 10%12
panel mean rating 1.5/5 → substitution pressure 12/100
Task breakdown (11 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Provide clients with communication assistance, typing their correspondence or obtaining information for them.
41CI 30–52 · exposure 42 · augmentation 75 · importance 4.2/5 · click for rater detail
Provide clients with communication assistance, typing their correspondence or obtaining information for them.
41| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Personal care aide work is concentrated in small care facilities, home-based settings, and low-digitization sectors with limited AI adoption infrastructure; regulatory and human-contact requirements slow experimentation and deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Home care and personal care services are a low-digitization, high-touch sector with slow AI adoption despite consumer-grade tools like voice assistants being available. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI tools for drafting, autocomplete, and information retrieval can meaningfully enhance a care aide's ability to help clients with correspondence and research, allowing faster service delivery while the aide maintains the critical role of understanding client needs and ensuring accuracy. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Speech-to-text, translation, and AI drafting tools can meaningfully speed up correspondence tasks and information retrieval for aides assisting clients with communication needs. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft emails and retrieve information reliably, personal care aides must understand individual client preferences, accessibility needs, and often provide real-time interaction. The task requires human judgment about client intent and context that current AI systems cannot fully replicate without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can draft or transcribe correspondence and retrieve information effectively, but the task also involves physical presence, interpreting client intent in real time, and personalized interaction that current systems cannot fully replace. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: care settings often have strict data protection requirements (HIPAA, state regulations), clients may require in-person presence for communication assistance, and liability concerns around misrepresenting client intent in written correspondence create legal friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks AI-assisted writing or information lookup, though clients with disabilities may prefer trusted human assistance for sensitive communications. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | While AI inference is cheap, integrating reliable communication systems into care workflows, maintaining client privacy/security, and providing necessary human oversight still requires significant cost relative to a personal care aide's hourly wage. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI tools for transcription/drafting are cheap, but the aide still must be physically present for most of the job, so cost savings apply only to a sub-task, not the full role. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Existing AI tools (chatbots, voice assistants, automated information retrieval) can perform partial aspects like drafting and information lookup, but no deployed product reliably handles the full scope of personalized communication assistance requiring understanding of individual client needs and constraints. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Dictation, transcription, and AI writing assistants are mature and widely deployed, but integrating them into a personal care aide's in-person workflow with a specific client's needs is not a standard deployed product today. |
Prepare and maintain records of client progress and services performed, reporting changes in client condition to manager or supervisor.
33CI 25–41 · exposure 30 · augmentation 63 · importance 4.7/5 · click for rater detail
Prepare and maintain records of client progress and services performed, reporting changes in client condition to manager or supervisor.
33| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Healthcare settings are digitizing records, but adoption of autonomous condition-monitoring AI remains low; most facilities use aides or nurses with software assistance rather than full automation. Regulatory conservatism and liability concerns slow velocity. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Home care and personal care sectors have low digitization and slow AI adoption overall, though some agencies are beginning to use basic digital record-keeping tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI voice-to-text, form auto-population, and flagging of key clinical terms can substantially boost aide productivity in documentation and alert supervisors to potential condition changes, keeping the aide in the loop while reducing clerical burden. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Voice-to-text and note-taking apps can speed up documentation and help structure reports, providing meaningful but partial assistance to the aide. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Record-keeping and data entry are partially automatable via structured forms and templates, but assessing client condition changes requires human judgment and direct observation. AI cannot reliably detect subtle health or behavioral changes from raw notes without human oversight, limiting time savings to well under 50%. |
| Task automatability | claude-sonnet-5 | 2/5 | Documentation can be partially drafted or structured via voice-to-text and templates, but accurate observation of client condition and judgment about what to report still requires an in-person human aide.; the core task depends on physical presence and situational judgment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Healthcare regulations (HIPAA, state licensing laws, and clinical documentation standards) often require a licensed or credentialed human to certify that observations and condition assessments are accurate and complete. Liability and malpractice risk create strong pressure for human sign-off. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically for documentation, but liability concerns around accurate health/condition reporting and employer requirements for human accountability create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI-assisted documentation tools (voice-to-text, auto-complete, structured entry) are comparable in cost to the time saved relative to aide wages, especially when including oversight for accuracy of change detection. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI transcription/documentation tools are cheap, but they only handle a small slice of this task; the observation and reporting judgment component still requires the human aide's time, limiting overall cost savings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Clinical documentation systems and EHRs exist and perform routine data entry and templating, but they require human nurses or aides to interpret observations and determine what constitutes a reportable change. Production systems handle standardized forms but lack reliable autonomous judgment on condition changes. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some home-care software includes voice dictation and note templates, but no deployed product autonomously observes clients and generates accurate progress reports without human input. |
Perform healthcare-related tasks, such as monitoring vital signs and medication, under the direction of registered nurses or physiotherapists.
29CI 5–54 · exposure 33 · augmentation 63 · importance 4.5/5 · click for rater detail
Perform healthcare-related tasks, such as monitoring vital signs and medication, under the direction of registered nurses or physiotherapists.
29| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Healthcare organizations are piloting remote monitoring and smart dispensing systems, but widespread production deployment in care homes and clinical settings remains patchy; adoption is faster in larger institutions and slower in small private-care settings, reflecting moderate overall velocity. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Personal care and home health sectors are low-digitization, physically-oriented, and show minimal AI agent adoption in production settings. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-driven monitoring systems and alerts meaningfully assist aides and nurses by reducing time spent on rote vital-sign checks and medication schedule tracking, allowing staff to focus on hands-on care and patient interaction; automated flagging of concerning trends augments human judgment. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-enabled monitoring devices (e.g., smart vital sign sensors, medication reminder apps) can assist aides by flagging anomalies or ensuring medication schedules, improving oversight without replacing the hands-on task. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Monitoring vital signs (temperature, pulse, blood pressure, oxygen) and medication schedules can be substantially automated via wearable sensors, connected medical devices, and automated alert systems; this covers the time-consuming surveillance component. However, the requirement to work "under the direction of" licensed professionals and handle exceptions means full end-to-end automation without human oversight is limited, preventing a 5. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, hands-on monitoring, and physical administration of medication to a person; current AI systems cannot perform physical caregiving tasks. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant regulatory and liability barriers exist: vital-sign monitoring and medication administration are often legally scoped to licensed nurses in many jurisdictions, and any lapse in automated systems carries clinical risk that organizations are hesitant to bear without professional accountability; legal frameworks require documented human sign-off. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Healthcare tasks under supervision of licensed professionals carry liability, safety, and regulatory requirements, and physical/direct human contact is essential, creating strong barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | Deployed monitoring systems (sensors, cloud backends, alerts) have meaningful upfront and per-patient costs, and human oversight for interpretation and exception-handling remains necessary; all-in cost is roughly comparable to or slightly lower than paying aides for routine monitoring time, but not decisively cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so no meaningful cost comparison exists; a human aide remains the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Mature products exist for remote vital-sign monitoring and medication reminders (commercial wearables, EHR systems, pill dispensers), but deployment in care settings often requires integration work and human review of alerts; error rates and narrow scope (e.g., limited sensor accuracy in certain populations) mean reliability is not yet production-grade at scale across all vital-sign types. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical vital-sign monitoring or medication administration for personal care aides; wearable sensors assist but don't replace the hands-on task. |
Instruct or advise clients on issues, such as household cleanliness, utilities, hygiene, nutrition, or infant care.
28CI 18–37 · exposure 20 · augmentation 50 · importance 4.4/5 · click for rater detail
Instruct or advise clients on issues, such as household cleanliness, utilities, hygiene, nutrition, or infant care.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Personal care services remain heavily analog and human-dependent, concentrated in in-home, small-firm, or non-profit settings with low digital maturity and strong cultural preference for human judgment in intimate care contexts. Adoption of AI advice tools is minimal in production. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Personal care work is a low-digitization, high-touch, low-wage sector with minimal AI agent deployment in production; adoption of AI tools by personal care aides remains negligible. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist a human aide by generating draft talking points on hygiene or nutrition, or by helping organize information for a client with memory concerns, but the aide must still assess, adapt, and take responsibility for the guidance—offering real but limited augmentation. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help aides by supplying reference material, checklists, or answers to nutrition/hygiene questions to relay to clients, offering moderate support without transforming the core interpersonal task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate generic advice on household cleanliness, hygiene, and nutrition, the task requires understanding individual client circumstances, cultural preferences, mobility limitations, and adaptive communication tailored to each person—factors that current AI systems handle poorly in unstructured home environments. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate generic advice on hygiene, nutrition, or household topics, but delivering personalized, context-sensitive instruction to a specific client requires physical presence, observation, and relationship-building that current AI cannot replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Personal care aides operate under licensing requirements in many jurisdictions, and liability for harmful advice (e.g., unsafe nutrition or infant care guidance) falls on the care provider and employer. Regulatory oversight of healthcare/elder-care advice, duty of care, and the legal requirement for a human to be accountable for guidance create substantial adoption barriers. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement for this specific advisory task, but strong human-contact expectations, trust-building with vulnerable clients (elderly, infants), and liability concerns around health/safety advice create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-generated advice (via chatbot or agent) costs near zero per interaction; human personal care aides command substantial hourly wages. The cost ratio heavily favors AI, though integration and oversight of AI advice in care contexts adds modest overhead. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While generic AI content generation is cheap, the actual task requires a human presence for assessment and rapport, so AI alone cannot substitute for the labor cost of the aide performing this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Chatbots can deliver pre-scripted advice, but no deployed product reliably assesses a specific client's household situation, understands their constraints, or delivers instruction in a way that substitutes for human judgment about what advice is safe and appropriate for that individual. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product provides in-person, tailored advisory care to individual clients in this domestic caregiving context; existing chatbots offer generic information only, not situated instruction. |
Plan, shop for, or prepare nutritious meals or assist families in planning, shopping for, or preparing nutritious meals.
24CI 18–30 · exposure 20 · augmentation 50 · importance 4.0/5 · click for rater detail
Plan, shop for, or prepare nutritious meals or assist families in planning, shopping for, or preparing nutritious meals.
24| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of AI meal-planning tools is slow in home care sectors, which are fragmented, low-digitization, and heavily dependent on direct human relationships and trust; no broad deployment of autonomous meal-planning or preparation in production care settings. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Personal care aide work is a low-digitization, high-physical-labor, small-employer-dominated sector with minimal AI agent deployment in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist by suggesting nutritious meal options, generating shopping lists, and offering recipe guidance with dietary constraints, materially reducing planning time for the aide and family without replacing hands-on food preparation and quality judgment. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI apps can help generate recipes, nutrition-appropriate meal plans, and shopping lists, meaningfully assisting aides in the planning portion of this task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with meal planning and recipe generation, the task requires shopping (physical store navigation, price comparison, product selection) and meal preparation (cooking, food handling, taste adjustment) that involve sensorimotor skills and real-time adaptation current systems cannot perform end-to-end at 50% time savings. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate meal plans and shopping lists from dietary constraints, but physical shopping and food preparation cannot be done end-to-end by current AI systems, and in-home client interaction is required. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Personal care aides work in regulated home-care settings where direct human presence is often mandated or strongly preferred by families and licensing; liability for nutrition and safety outcomes typically requires human accountability and sign-off. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars AI from meal planning, but the task requires physical presence, judgment about client preferences/health needs, and trust built through personal caregiving relationships. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI meal planning tools are inexpensive, but integrating them with shopping and cooking oversight, plus the human assistance still required for physical tasks, does not yet achieve a cost advantage over a personal care aide's hourly wage for the bundled service. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply generate meal plans, but the human labor cost of shopping and physical food prep remains unchanged, so overall cost savings for the full task are minimal. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Meal-planning chatbots and recipe apps exist, but no deployed product reliably handles the full workflow of nutritional assessment, shopping execution, and meal preparation. Most products are narrow recipe databases or planning assistants without integration into actual shopping or cooking execution. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs the full physical task of shopping and preparing meals for a client; AI is limited to text-based planning assistance, not production-scale automation of the task itself. |
Participate in case reviews, consulting with the team caring for the client, to evaluate the client's needs and plan for continuing services.
13CI 0–25 · exposure 8 · augmentation 38 · importance 4.5/5 · click for rater detail
Participate in case reviews, consulting with the team caring for the client, to evaluate the client's needs and plan for continuing services.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task occurs in highly regulated, relationship-dependent care settings (nursing homes, in-home care agencies, social services) with slow digital adoption and strong compliance requirements; displacement is minimal and strongly resisted. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Home healthcare and personal care sectors are low-digitization, labor-intensive fields with slow AI adoption for direct care coordination tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Limited augmentation potential exists—AI might summarize prior notes or flag trends in client data before the meeting—but case review is fundamentally a human interpersonal and decision-making task where AI assistance is marginal and secondary to the human aide's participation. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help prepare summaries, track client history, and suggest care plan updates ahead of meetings, moderately aiding the human's preparation and follow-through. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time judgment about client needs, interpersonal coordination with a care team, and integration of nuanced observations that current AI cannot perform end-to-end. While AI might summarize existing records, the core activity—participating in collaborative case reviews with human judgment and accountability—cannot be automated. |
| Task automatability | claude-sonnet-5 | 2/5 | AI could help summarize case notes or draft talking points, but active participation in a live multidisciplinary case review requires real-time judgment, relationship context, and verbal interaction that current systems cannot fully replace. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong barriers exist: care plans typically require licensed professionals' sign-off, liability and fiduciary duty rest on identifiable humans, and regulations (Medicaid, state licensing boards) mandate human review and planning accountability for vulnerable populations. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no formal licensing mandates a human attend case reviews, care coordination often involves accountability, trust, and regulatory documentation expectations that favor human presence. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The oversight and liability costs of any AI involvement in case planning would exceed the cost of a human aide attending the meeting; the task generates legal and ethical accountability that AI cannot bear, making automation uneconomical. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Since AI cannot perform the core interactive task, any cost comparison is limited to augmentation tools which add cost without replacing the human's role, making all-in AI substitution costlier or infeasible. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed products perform genuine case review participation today; AI tools exist for documentation support, but no system can reliably substitute for a human team member in a collaborative clinical/care review meeting with enforceable accountability. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for a personal care aide's participation in case review meetings; at most AI assists with note-taking or scheduling behind the scenes. |
Train family members to provide bedside care.
12CI 5–19 · exposure 5 · augmentation 50 · importance 4.6/5 · click for rater detail
Train family members to provide bedside care.
12| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Personal care aides work in highly fragmented, low-digitization settings (homes, small facilities) with limited tech infrastructure. Adoption of any AI-based training tool is laggard, and the sector has minimal organizational incentive to automate this interpersonal training task. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Personal care work is a low-digitization, high physical-contact sector with minimal AI adoption for hands-on training tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist by generating instructional videos, checklists, and reference materials that a care aide uses while teaching, and by providing post-session quizzes or documentation aids. This would improve efficiency and consistency without replacing the human trainer's core role. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can supplement training with instructional videos, checklists, or chatbot-based Q&A support, but cannot replace the hands-on demonstration and personalized coaching central to the task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Training family members requires interpersonal communication, assessment of learning needs, adaptive teaching, and real-time feedback based on individual comprehension and emotional context. Current AI cannot reliably deliver this dynamic, context-sensitive human interaction or assess and respond to trainee competence in real-time at equal quality. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires in-person demonstration, hands-on correction, empathetic communication, and physical presence to teach caregiving skills; current AI cannot perform this end-to-end.rating |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | There is implicit expectation that a trained human professional (the care aide) must assess whether family members are capable of providing safe care, and liability falls on the care aide for training quality. Healthcare quality and safety standards create pressure for human sign-off on competency, restricting pure automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement mandates a specific professional for this training, but the need for physical demonstration and trust with family members creates practical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-generated training content is cheap, but the full cost of creating personalized, validated training curricula plus oversight to ensure family members achieve competency remains substantial relative to a care aide's labor. The setup and customization costs are non-trivial. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the human instructor's physical presence and supervision, so no meaningful cost savings exist for the core task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate training materials and educational content, no deployed product reliably performs end-to-end training of family members with the personalized, interactive, and emotionally-attuned instruction this task demands. Video tutorials and chatbots exist but lack the adaptability and human judgment needed for effective bedside care training. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product trains family members in hands-on bedside care; this remains a human, in-person instructional task. |
Perform housekeeping duties, such as cooking, cleaning, washing clothes or dishes, or running errands.
12CI 5–19 · exposure 8 · augmentation 25 · importance 4.2/5 · click for rater detail
Perform housekeeping duties, such as cooking, cleaning, washing clothes or dishes, or running errands.
12| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Personal care settings are predominantly low-digitization, small-firm environments with limited technology infrastructure and strong preference for direct human contact. Adoption of automation in this sector remains minimal, with most organizations in laggard positions relative to finance or tech-forward sectors. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Home care and domestic services are a low-digitization, physically embodied sector with minimal AI/robotics adoption in production settings. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While simple task-planning or scheduling aids might assist a caregiver, current AI offers minimal productivity boost for the physical execution of cooking, cleaning, and errands—the core of this task. The human remains essential, and AI augmentation opportunities are narrow and marginal. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can offer minor assistance such as recipe suggestions, scheduling errands, or smart home reminders, but it does not materially transform the physical execution of these housekeeping tasks. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Current AI systems cannot reliably perform the full range of physical housekeeping tasks (cooking, cleaning, washing clothes/dishes, errands) end-to-end. Robotic systems for these tasks remain in early stages and lack the dexterity, adaptability, and real-world reasoning needed for consistent execution at equal quality without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation of objects, cooking, cleaning, and mobility in unstructured home environments—capabilities current AI systems (software or robotics) cannot perform end-to-end reliably today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Housekeeping for personal care aides often occurs in regulated care settings (assisted living, in-home care) where direct personal contact and trust are central to the service model. Liability concerns, elder/vulnerable-person preferences for human contact, and organizational culture around personal care create substantial friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars automation, but strong human-contact preference, trust, safety concerns in someone's home, and liability for property damage or client injury create real friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Current robotic and AI-driven housekeeping solutions are significantly more expensive than the loaded wage of a personal care aide, when factoring in hardware, maintenance, software integration, and the need for human oversight and intervention for most tasks. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI/robotic substitute at scale, so cost comparison favors the human aide who can already perform these physical tasks affordably relative to any hypothetical robotic system. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed products reliably perform the complete set of housekeeping duties at scale in real-world care environments. While some narrow tasks (e.g., robot vacuums) have limited deployment, integrated systems that cook, clean dishes, launder, and run errands remain research-stage or proof-of-concept. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed consumer product performs general household cooking, cleaning, laundry, and errand-running autonomously; robotic solutions remain research/pilot-stage and narrow in scope. |
Administer bedside or personal care, such as ambulation or personal hygiene assistance.
3CI 0–5 · exposure 0 · augmentation 13 · importance 4.7/5 · click for rater detail
Administer bedside or personal care, such as ambulation or personal hygiene assistance.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Adoption of AI or robotics for personal care is minimal and slow in healthcare and home settings. The sector remains labor-intensive and highly reliant on human workers; no evidence of measurable displacement or production-scale adoption in this specific task. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Home health and personal care sectors are low-digitization, labor-intensive, and show minimal AI/robotic adoption for hands-on physical care tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | Current AI does not meaningfully augment a human performing bedside or hygiene assistance. Voice assistants or scheduling tools may help administrative aspects, but do not assist the core task of hands-on physical care. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can support scheduling, monitoring, or reminders around care routines, but offers little direct assistance to the physical act of ambulation or hygiene support itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires direct physical contact and bodily assistance (helping with walking, toileting, bathing) that current AI systems cannot perform. Physical robots lack the dexterity, safety, and reliability to handle such intimate care tasks end-to-end, and no current AI-powered system can achieve meaningful time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 1/5 | This is direct physical hands-on care (helping someone walk, bathe, toilet) requiring physical manipulation, balance support, and touch that no current AI system or robot can perform reliably or safely. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and regulatory barriers exist: licensed healthcare workers or certified aides are often legally required or expected for personal care, liability for injury is significant, and intimate contact with vulnerable populations creates high error-cost asymmetry. Consumer preference for human touch in care is also a durable social barrier. |
| Adoption barriers | claude-sonnet-5 | 4/5 | While not always formally licensed, personal care involves direct physical contact, safety risk, and liability for falls or injury, creating strong practical and organizational barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Current robotic systems capable of any form of physical assistance are expensive to acquire, maintain, and operate, making them far more costly per task than a human aide's hourly wage. The all-in cost (hardware, software, infrastructure, liability) vastly exceeds human labor cost. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute delivering this physical service, so any hypothetical AI/robotic solution would be far more expensive and less capable than a human aide today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed products reliably perform bedside or personal hygiene assistance autonomously. While some robotic platforms exist in research or limited pilots, they are not in production use for this care task in mainstream healthcare or home settings. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical ambulation or hygiene assistance for patients; robotics for personal care remains research-stage and not used at scale in production. |
Care for individuals or families during periods of incapacitation, family disruption, or convalescence, providing companionship, personal care, or help in adjusting to new lifestyles.
3CI 0–5 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail
Care for individuals or families during periods of incapacitation, family disruption, or convalescence, providing companionship, personal care, or help in adjusting to new lifestyles.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Personal care work is conducted by dispersed, low-digitization sectors (home care, small care facilities, family networks) with high regulatory oversight and deep cultural/ethical reliance on human touch. Adoption of AI automation in this domain is minimal and faces strong structural resistance. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Home care and personal care services are a low-digitization, physically-intensive sector with minimal AI agent deployment or measured displacement to date. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Conversational AI could assist with reminders, activity suggestions, or light emotional support, but the core task—physical care, adaptive judgment, and trusted presence—offers limited augmentation opportunity. AI tools remain marginal to the actual care work performed. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI tools like reminder apps, scheduling assistants, or health monitoring devices can support caregivers logistically, but offer little assistance with the core companionship and hands-on care functions. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires sustained physical presence, emotional support, personal hygiene assistance, and adaptive interpersonal judgment in response to vulnerable individuals' needs. Current AI systems lack embodied presence, physical manipulation capability, and the contextual empathy necessary for companionship and lifestyle adjustment support. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires physical presence, hands-on personal care, emotional companionship, and adaptive judgment during vulnerable life transitions—none of which current AI can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and regulatory barriers exist: most jurisdictions require licensed or credentialed human caregivers for vulnerable populations; liability and duty-of-care laws assign accountability to humans; physical care of incapacitated persons is fundamentally a human-contact requirement. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Personal care often involves licensing/certification requirements, safety and liability concerns, and strong client/family preference for human contact, especially with vulnerable populations. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Personal care aides earn modest wages (~$30k–$35k annually), and the task demands constant human presence. Even including a care worker's full loaded cost, an AI system would need to deliver equivalent physical and emotional support while maintaining 24/7 availability—economically infeasible today. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical caregiving component at all, so any cost comparison favors the human aide who can actually perform the task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product performs in-person personal care, companionship, or physical assistance tasks in real care settings. While conversational AI can provide limited emotional support remotely, it cannot replace the embodied care, hygiene support, and adaptive presence required by incapacitated individuals. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product provides physical personal care or genuine companionship for incapacitated individuals; this remains far outside current AI product capability. |
Transport clients to locations outside the home, such as to physicians' offices or on outings, using a motor vehicle.
3CI 0–5 · exposure 0 · augmentation 13 · importance 3.7/5 · click for rater detail
Transport clients to locations outside the home, such as to physicians' offices or on outings, using a motor vehicle.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Personal care aide work occurs in small, localized, non-digitized settings with low capital investment; autonomous vehicle adoption in this sector is negligible, with no production deployments for dependent client transport. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Home care and personal assistance sectors have low AI/autonomous vehicle adoption, driven by small firm structures, physical service delivery, and regulatory caution. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers minimal augmentation for this task; route planning software exists but does not meaningfully assist the core function of safely driving a dependent client to appointments or outings. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with route optimization, scheduling, or navigation assistance, but offers minimal augmentation to the core physical transport and caregiving task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires operating a motor vehicle in real-world conditions with a dependent passenger, navigating unpredictable traffic and environments, and responding to client needs—capabilities far beyond current AI systems. Autonomous vehicles remain unreliable for general public roads, and no AI system today can safely handle the full responsibility of transporting vulnerable clients. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical transportation of a vulnerable client requires a driver present in a vehicle with the client, plus care duties like assistance entering/exiting; no off-the-shelf AI system performs this end-to-end today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Transporting dependent clients via motor vehicle faces hard legal and regulatory barriers: drivers must be licensed, insured, and accountable; liability for client safety during transport is significant; and most jurisdictions require human drivers responsible for passenger welfare. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Liability, safety regulations for vulnerable populations, insurance requirements, and the need for hands-on physical assistance create strong barriers to full automation of this task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Current autonomous vehicle systems, where they exist, are expensive to deploy and maintain, with high liability and insurance costs that exceed the loaded wage of a personal care aide performing this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-only substitute, so cost comparison favors the human; any autonomous vehicle option still requires human caregiving support, adding cost rather than reducing it. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | While autonomous vehicle research exists, no deployed product reliably handles general-purpose client transportation with the safety and reliability standards required. Ride-sharing automation is nascent and not yet production-ready for dependent populations. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously transports personal care clients to appointments while providing associated caregiving; robotaxis exist in limited pilot areas but don't address the care aide role or assistance needs. |
Related occupations — Healthcare Support
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.