Nannies
39-9011.01Median wage $34,980/yr518,910 employed (US)Rank #844 of 923 scored · top 91% by substitution
Care for children in private households and provide support and expertise to parents in satisfying children's physical, emotional, intellectual, and social needs. Duties may include meal planning and preparation, laundry and clothing care, organization of play activities and outings, discipline, intellectual stimulation, language activities, and transportation.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution13
Exposure12
Augmentation35
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
19 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
11%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%11
panel mean rating 1.4/5 → substitution pressure 11/100
Technical feasibility todayw 20%13
panel mean rating 1.5/5 → substitution pressure 13/100
Cost vs. human wagew 15%9
panel mean rating 1.4/5 → substitution pressure 9/100
Adoption barriersw 20%inverted — strong barriers lower the score22
panel mean rating 4.1/5 (barrier strength) → substitution pressure 22/100
Sector adoption velocityw 10%7
panel mean rating 1.3/5 → substitution pressure 7/100
Task breakdown (19 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.
Keep records of play, meal schedules, and bill payment.
84CI 70–97 · exposure 87 · augmentation 75 · importance 3.8/5 · click for rater detail
Keep records of play, meal schedules, and bill payment.
84| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Childcare and family services sectors are increasingly digitized; major platforms (Brightwheel, Kinly) report rapid uptake among childcare providers, and consumer calendar/expense tools see broad adoption in household contexts. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Household/childcare services are a low-digitization, small-scale sector where formal AI tool adoption is slow and inconsistent despite tool availability. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted record-keeping directly augments nanny productivity by auto-filling schedules, reminding about bill payments, and organizing meal logs, allowing the nanny to focus on child interaction rather than manual paperwork. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered apps significantly ease and speed up logging, reminders, and payment tracking, meaningfully boosting a nanny's efficiency on this administrative task. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Record-keeping for play schedules, meals, and bills is a straightforward data-entry and organizational task well-suited to current AI. Calendar management, expense tracking, and schedule logging can be fully automated with calendar apps, expense trackers, and note systems that meet the ≥50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Record-keeping of schedules, meal logs, and bill payments is a structured data-entry/tracking task easily handled by apps, spreadsheets, or AI-assisted scheduling tools with minimal human input beyond initial data capture. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While caregiving contexts may involve parental preferences for human oversight and some data-privacy sensitivity, there are no legal barriers or licensing requirements preventing automation of these administrative records. Adoption friction is light—mainly organizational inertia. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, legal, or safety barrier prevents using apps or software to track meals, play, and payments; it's a purely administrative function. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Cloud-based calendar, note, and expense-tracking tools cost pennies per user per month, orders of magnitude cheaper than the loaded hourly wage of a nanny ($15–25/hour or more), making automation vastly more cost-effective. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Digital record-keeping and bill-pay automation tools cost only a few dollars monthly versus paying a human for this administrative sub-task, making AI-assisted tools far cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products like Google Calendar, Notion, Microsoft 365, and specialized childcare software (e.g., Brightwheel, Kinderlime) reliably perform these record-keeping functions in production today, handling scheduling, meal logs, and payment tracking at scale. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Numerous deployed apps (childcare tracking apps, budgeting/bill-pay tools) already reliably handle logging and payment records in production, though nannies must still input raw observations. |
Help develop or monitor family schedule.
72CI 44–100 · exposure 62 · augmentation 88 · importance 3.5/5 · click for rater detail
Help develop or monitor family schedule.
72| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Family scheduling automation is already deeply adopted across consumer and professional sectors; calendar AI and voice assistants are ubiquitous, integrated into smartphones, and widely used for family coordination in households and schools. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Domestic childcare is a low-digitization, informal sector with slow uptake of AI tools compared to professional office environments. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI scheduling assistants meaningfully augment nanny productivity by automating schedule tracking, generating alerts, handling conflicts, and reducing manual coordination overhead while the nanny remains in charge of childcare decisions and responsiveness. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI calendar assistants and reminder apps can meaningfully help a nanny track appointments, activities, and family logistics, improving efficiency while the nanny remains in control. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Scheduling and monitoring family calendars is a routine, well-structured task that AI systems can fully automate today. Calendar management, conflict detection, and reminder generation meet the 50%-time-saving bar using off-the-shelf tools like Google Calendar integrations, AI assistants, and scheduling bots. |
| Task automatability | claude-sonnet-5 | 2/5 | AI calendar/scheduling tools can help organize a family schedule, but nannies must integrate real-time child needs, family preferences, and physical presence, limiting full automation to well under the 50% threshold for the whole task.》 |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No legal, regulatory, or authorization barriers prevent automated scheduling; family schedules are private administrative data without licensure requirements. Adoption is purely voluntary and frictionless. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing, legal, or liability barrier preventing AI-assisted scheduling tools from being used to help manage family calendars. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI calendar and scheduling systems cost pennies per month for consumer software or minutes of LLM inference for enterprise integration, orders of magnitude cheaper than a nanny's loaded wage (~$30–50k/year or $15–25/hour for this single task component). |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Scheduling software is cheap, but a nanny's schedule task is a minor part of a broader in-person role, so cost comparison is mixed rather than clearly favoring AI. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Multiple mature products reliably handle family schedule management in production: Google Assistant, Apple Siri, calendar apps with AI conflict resolution, and specialized family coordination tools all demonstrate reliable performance at scale with minimal error rates. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Consumer scheduling apps and AI assistants (e.g., calendar bots) exist and are used by some families, but they are not reliable substitutes for the judgment and personal coordination a nanny provides in practice. |
Work with parents to develop and implement discipline programs to promote desirable child behavior.
29CI 5–52 · exposure 33 · augmentation 63 · importance 4.5/5 · click for rater detail
Work with parents to develop and implement discipline programs to promote desirable child behavior.
29| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Childcare remains a highly relationship-dependent, low-tech sector where parents prioritize human nanny judgment over algorithmic discipline frameworks; adoption of AI for this specific task is minimal despite general parenting-app penetration. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Childcare and domestic care work is a low-digitization, high-touch sector with minimal AI adoption in caregiving decisions or behavioral management. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can substantially assist nannies by drafting behavior plans, suggesting evidence-based strategies, and tracking behavioral progress, allowing the nanny to focus on relationship and real-time execution rather than research and planning. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI chatbots and parenting apps can offer suggested discipline strategies or behavioral frameworks that a nanny could reference, providing some assistance without executing the interpersonal work. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can generate discipline frameworks, behavioral strategies, and implementation plans based on child development research and parental circumstances with minimal human setup. However, the final ~20-30% requires human judgment on family dynamics, cultural context, and real-time behavioral feedback that AI cannot fully capture, achieving substantial but not complete automation. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires in-person relationship building, real-time behavioral observation of a specific child, and ongoing collaborative negotiation with parents—none of which current AI can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Parents retain strong preference for human relationship-building and trust in discipline decisions; there is also implicit liability if an AI-generated program fails or causes harm, creating organizational and legal friction against full substitution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement exists for nannies, but strong parental trust, in-person presence, and child safety/liability concerns create substantial practical barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-generated discipline frameworks are cheap to produce, but effective implementation requires human nanny judgment, parent consultation, and iterative refinement; the all-in cost remains comparable to or higher than hiring an experienced human nanny for this component. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this task, so no meaningful cost comparison exists; a human nanny remains the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Products exist (parenting apps, chatbots offering discipline advice) that demonstrate basic capability, but they operate at narrow scope and rely heavily on human interpretation; no mature production system reliably implements end-to-end discipline programs without significant parental oversight and adaptation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product designs and implements child-specific discipline programs in coordination with parents and a live-in caregiver context; this is purely research/theoretical territory for AI. |
Supervise and assist with homework.
19CI 9–30 · exposure 17 · augmentation 63 · importance 4.1/5 · click for rater detail
Supervise and assist with homework.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare and nanny roles remain stubbornly human-centric; adoption of AI for unsupervised homework help is minimal and limited to supplementary tools. The sector has low digitization and strong cultural/legal resistance to removing the human from the supervision role. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Domestic childcare is a low-digitization, in-person sector with minimal AI agent deployment; adoption is mostly limited to educational software used alongside a caregiver. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | A nanny could use an AI tutor or homework helper as an assistant tool to explain concepts or generate practice problems, raising productivity on some homework tasks. However, the nanny's core function—supervision, motivation, and judgment—remains human-driven. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI homework helpers and tutoring chatbots meaningfully assist nannies in explaining concepts and checking answers, improving the quality of homework help while the nanny remains present and responsible. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Homework supervision and assistance requires real-time judgment about a child's understanding, adaptive explanations tailored to their learning style, and the ability to motivate and redirect—contextual, relational tasks that current AI cannot reliably execute end-to-end in live settings. Chatbots can answer questions but cannot replicate the continuous, responsive presence a nanny provides. |
| Task automatability | claude-sonnet-5 | 2/5 | AI tutoring tools can help explain homework content, but 'supervise' requires physical presence, attention monitoring, and behavioral management that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Parents have strong preference for human judgment in childcare; duty-of-care and liability concerns mean a child cannot legally be left unsupervised with only AI. Legal and regulatory frameworks, parental trust requirements, and the irreducible need for a responsible adult present create near-absolute barriers. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement for homework help itself, but parental trust, in-person childcare needs, and child safety/supervision expectations create real friction against full substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Nanny wages are typically $15–25/hour or higher; AI tutoring tools cost dollars per month at scale. However, the full cost of integration, monitoring for safety/appropriateness, and addressing failures in a childcare context likely approaches or exceeds the loaded cost of part-time supervision. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI tutoring assistance is cheap per query, but since a human must still be present to supervise, the overall task cost is not meaningfully reduced by AI alone. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | LLMs can provide tutoring-like explanations and answer academic questions, but no deployed product reliably supervises a child's homework session, adjusts difficulty on the fly, or manages behavioral/motivational issues in real time. Homework apps exist but require human oversight; they do not replace the nanny role. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tutoring apps (e.g., Khan Academy's Khanmigo, chatbots) reliably assist with homework content, but no deployed product provides the supervisory/childcare component of this task. |
Shop for groceries, clothing, and other items needed for children's care.
18CI 5–31 · exposure 13 · augmentation 50 · importance 2.9/5 · click for rater detail
Shop for groceries, clothing, and other items needed for children's care.
18| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While general e-commerce adoption is high, the specific use case of AI agents autonomously shopping for children's needs remains in early pilot phases with minimal production displacement in household/nanny contexts. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Childcare and household shopping are low-digitization, physical-world tasks with minimal AI adoption in this sector. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist nannies by generating shopping lists, comparing prices, and identifying items meeting specified criteria, meaningfully reducing time spent on planning and research while the human retains final purchasing decisions. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with list-making, price comparison, and online ordering suggestions, moderately improving efficiency while the nanny still executes the shopping. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with shopping list generation and price comparison, the task requires human judgment on child-appropriate selections, quality assessment, and physical store navigation/purchasing, which remain difficult to automate end-to-end today. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical shopping requires selecting, purchasing, and transporting goods, which is a physical-world task current AI cannot perform end-to-end without human or robotic execution. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Parents retain strong preferences for human judgment on children's products and care quality; legal and liability concerns around autonomous purchasing decisions for minors create organizational friction and oversight requirements. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates a human for shopping, but trust, judgment about children's needs, and physical execution create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Current AI-assisted shopping solutions (including delivery service markups and integration overhead) are more expensive than a human nanny shopping directly, especially when accounting for error correction and oversight labor. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-only substitute for physical shopping, so cost comparison favors the human who can actually complete the task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed products reliably perform full grocery and clothing shopping for children autonomously; existing shopping assistants require extensive human oversight and cannot handle the sensory evaluation and contextual decision-making required. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product independently shops for children's groceries and clothing; at most AI can generate lists or place orders via apps controlled by a human. |
Help prepare and serve nutritionally balanced meals and snacks for children.
12CI 5–19 · exposure 13 · augmentation 38 · importance 4.7/5 · click for rater detail
Help prepare and serve nutritionally balanced meals and snacks for children.
12| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Adoption of autonomous food preparation in childcare is virtually nonexistent. The sector remains highly labor-intensive and relationship-driven, with families preferring human oversight of nutrition and feeding practices for their children. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare and domestic work sectors show minimal AI/robotics adoption for physical caregiving tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI meal-planning assistants and nutritional analysis tools can help nannies design balanced menus and track dietary needs, moderately improving task efficiency. However, the core physical and interactive components remain human-driven. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with meal planning, recipes, or nutritional balance suggestions, but offers little assistance with the actual physical preparation and serving. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate meal plans and provide nutritional guidance, the physical execution of food preparation and serving requires human manipulation of tools and ingredients. Current AI systems cannot autonomously handle food safety, dietary restrictions in real-time, or serve meals to children with the oversight needed. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical meal prep and serving for children requires hands-on presence, dexterity, and real-time supervision that no current AI system can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Food safety regulations, child welfare standards, and parental preference for direct human care create substantial barriers. In many jurisdictions, childcare providers are legally responsible for meal quality and child safety during feeding, which cannot be delegated to automated systems. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Direct childcare involves safety, liability, and trust concerns; parents and regulations generally require a responsible human present for food prep and feeding of children. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of robotic food preparation systems, if they existed at scale, would far exceed the wage of a nanny. AI meal-planning software is cheap but does not eliminate the human labor needed for actual cooking and serving. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing the physical labor, so comparing cost per task-equivalent is not meaningful—human labor is the only option and thus effectively cheaper than any hypothetical robotic alternative. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed AI can suggest balanced meal plans and nutritional content, but no current product autonomously prepares, cooks, and serves meals to children. Robotic systems for food preparation exist only in narrow laboratory or specialized settings, not in homes or childcare facilities at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically prepares or serves food to children; this remains purely a human physical-care task. |
Perform housekeeping and cleaning duties related to children's care.
10CI 5–15 · exposure 0 · augmentation 13 · importance 4.1/5 · click for rater detail
Perform housekeeping and cleaning duties related to children's care.
10| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Adoption of autonomous cleaning in household childcare settings is minimal; most families rely on human nannies. The domestic sector has low digitization and high preference for human presence and judgment. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Domestic childcare and housekeeping is a low-digitization, physical, small-scale employment sector with essentially no AI/robotic adoption in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide scheduling reminders or cleaning checklists, but offers minimal augmentation for the core physical cleaning task that defines the role; human nannies must remain present for supervision and safety. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers negligible assistance for physical cleaning and tidying tasks tied to child care; smart home devices provide only marginal, indirect help. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Housekeeping and cleaning duties require physical manipulation in unstructured home environments with fragile items and child safety concerns. Current robots cannot reliably perform general household cleaning or organize spaces while maintaining child-safe conditions. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical task requiring hands-on cleaning, tidying, and childcare-related chores; no current AI system can physically perform housekeeping duties. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: parents require trust and human judgment for child safety oversight, liability concerns around autonomous systems near children, and regulatory/insurance gaps for robotic childcare assistance create organizational friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing barrier specifically covers cleaning, but the task is embedded in a trust-based, in-home caregiving role with strong parental preference for human presence. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Household cleaning robots that exist are expensive capital purchases (thousands of dollars) with significant maintenance costs, vastly exceeding the hourly labor cost of nanny services in most markets. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for physical cleaning labor, so cost comparison favors the human by default; any robotic solution would be far more expensive than a nanny's wage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI systems can autonomously perform end-to-end housekeeping and cleaning in typical homes. Robotics for household cleaning remain research-stage or extremely limited in scope and reliability. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical housekeeping and childcare-related cleaning; robotics for such unstructured domestic tasks remain research-stage. |
Teach and perform age-appropriate activities, such as lap play, reading, and arts and crafts, to encourage intellectual development of children.
7CI 0–14 · exposure 8 · augmentation 25 · importance 4.8/5 · click for rater detail
Teach and perform age-appropriate activities, such as lap play, reading, and arts and crafts, to encourage intellectual development of children.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare and nanny services remain labor-intensive, low-digitization sectors with minimal AI adoption; regulatory barriers, trust concerns, and the irreducibility of human attachment and supervision make automation unlikely even in early-adopter contexts. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Childcare is a low-digitization, physical-presence sector with essentially no AI adoption for hands-on activities. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can assist by suggesting activity ideas, reading recommendations, or developmental milestones, but the core task of in-person teaching, play, and rapport-building benefits only marginally from AI tools; the human nanny's presence and responsiveness remain essential. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can suggest age-appropriate activity ideas or reading materials in advance, but offers minimal real-time assistance during the actual physical interaction with the child. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate age-appropriate activity ideas and reading suggestions, the core task requires real-time, responsive interaction, physical presence, and emotional attunement to individual children's needs. Current AI systems cannot replicate the adaptive, in-person teaching and play engagement that drives intellectual development. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, hands-on play, and real-time human interaction with a child, which no current AI system can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal, regulatory, and safety barriers protect this role: childcare requires licensed or authorized caregivers in most jurisdictions, parents have a fundamental preference for trusted human caregivers, and liability and duty-of-care requirements legally mandate human oversight of children. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Parental trust, safety liability, and the fundamentally in-person nature of childcare create strong barriers to any automated substitution, though no formal licensing is typically required for nannies. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems cannot yet replace the labor cost of a nanny, as the task requires continuous human presence, supervision, and responsive interaction that no current technology can provide at scale. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical, supervisory task, so cost comparison favors the human by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs this task end-to-end; AI literacy applications exist but they are supplements to human instruction, not substitutes. The task fundamentally requires a human caregiver present to observe, adapt, and interact with children in real-time. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs in-person childcare activities like lap play or supervised arts and crafts; this remains purely a human physical caregiving task. |
Meet regularly with parents to discuss children's activities and development.
5CI 5–5 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail
Meet regularly with parents to discuss children's activities and development.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare and nanny services remain predominantly offline, relationship-driven, and resistant to automation; adoption of AI in this context is minimal given the personal trust requirements. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Domestic childcare is a low-digitization, highly personal service sector with minimal AI adoption for interpersonal caregiving tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI might assist with transcribing notes or summarizing child activities for documentation, but the core task of discussing development requires human judgment and interpersonal presence that AI augmentation cannot substantially enhance. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help a nanny organize notes, track milestones, or draft summaries beforehand, offering modest support, but the actual meeting and communication remain unaided by AI. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires understanding complex family dynamics, child developmental nuances, and interpersonal communication with parents—all heavily dependent on context, emotional intelligence, and trust-building that current AI cannot replicate autonomously. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires an in-person relational conversation between caregiver and parent grounded in the nanny's direct observation of the child; AI cannot substitute for the human presence and trust involved.the task is inherently interpersonal. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Parents typically require direct human contact with their child's caregiver to build trust and confidence; substituting AI would face strong customer and cultural preference for human interaction in matters involving child welfare. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Strong reliance on personal trust, direct observation, and parental preference for human judgment about their child creates high friction against automation, though no formal licensing requirement exists. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Meaningful automation of this task would require human-level conversational AI and relationship management, which remains expensive relative to the cost of a human nanny conducting brief regular meetings. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this interaction, so cost comparison favors the human by default; any AI note-taking aid is marginal to the core task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably conducts parent-nanny meetings or replaces this face-to-face discussion. AI cannot establish the rapport, answer follow-up questions dynamically, or adapt to parental concerns in real time. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs parent-caregiver child development discussions; this is a live human interaction, not a research or product target. |
Instruct children in safe behavior, such as seeking adult assistance when crossing the street and avoiding contact with unsafe objects.
3CI 0–5 · exposure 5 · augmentation 25 · importance 4.9/5 · click for rater detail
Instruct children in safe behavior, such as seeking adult assistance when crossing the street and avoiding contact with unsafe objects.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare remains a low-digitization, highly interpersonal, regulation-heavy sector; adoption of automation for core safety functions is minimal and legally constrained. Sector-wide adoption patterns show reliance on licensed or vetted human caregivers. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare and nanny services are a low-digitization, high-touch physical sector with minimal AI adoption for direct supervisory tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with generating safety materials or reminders, but the core task—modifying children's behavior through instruction and enforcement—requires human presence and judgment that AI augmentation minimally enhances in practice. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI tools (apps, reminders, educational content) could supplement teaching materials about safety, but the core supervisory instruction and modeling remains manual with limited AI assistance. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Instructing children in safe behavior requires ongoing monitoring, responsiveness to individual child understanding, adaptive teaching based on developmental stage, and real-time environmental judgment—capabilities that current AI systems cannot execute reliably in physical, unsupervised contexts. No off-the-shelf system can replace a present caregiver enforcing safety rules across dynamic situations. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires ongoing, real-time physical presence, supervision, and in-person modeling of safe behavior with children, which current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Child safety instruction is intrinsically a human-contact task with high liability and regulatory oversight; legal guardianship and duty of care requirements mandate human accountability. Regulatory and organizational barriers make full substitution infeasible. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Childcare requires direct physical supervision and trust; parents and regulations expect a responsible human adult present at all times for child safety. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of AI systems (surveillance, content delivery, oversight) plus the liability of failure in safety instruction far exceeds the cost of human care. Safety instruction cannot be cost-offset because the human supervisor must remain present anyway. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical presence needed, so there is no viable cost comparison as a replacement; any hypothetical solution would still require human oversight, making AI not cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate safety content and scripts, deployed systems lack the embodied presence and real-time adaptation needed to instruct children in behavioral safety during actual dangerous moments (street crossing, object avoidance). No production system reliably delivers this task end-to-end. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product provides in-person childcare supervision and safety instruction; this remains a physical, relational task outside AI's current scope. |
Remove hazards and develop appropriate boundaries and rules to create a safe environment for children.
3CI 0–5 · exposure 0 · augmentation 25 · importance 4.9/5 · click for rater detail
Remove hazards and develop appropriate boundaries and rules to create a safe environment for children.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare remains one of the most human-dependent sectors with minimal AI deployment. Regulatory requirements, liability concerns, and parental preference for human caregivers create structural resistance to automation in this domain. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare is a low-digitization, physical-presence sector with essentially no AI adoption for hazard removal or supervision. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could marginally assist by providing hazard checklists or rule templates, but the core task—perceiving hazards in real time, making judgment calls, and enforcing boundaries—remains fundamentally human. The assistance is modest and secondary to human decision-making. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could provide checklists or hazard-awareness tips via apps, but it offers minimal real-time assistance for physically inspecting and childproofing a space or enforcing rules. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time situational awareness, dynamic risk assessment, and immediate physical intervention in a home or childcare environment. Current AI systems cannot perceive physical environments independently, identify emerging hazards in real time, or enforce boundaries through physical presence or social authority with children. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence to identify hazards in a real home environment and ongoing judgment to enforce rules with actual children; no AI system can perform this end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Childcare is heavily regulated and legally requires responsible human adults present and accountable. Liability for child injuries is severe, and parents fundamentally expect human judgment and presence for safety. Legal and insurance frameworks mandate human accountability. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Direct child supervision carries high liability, trust, and safety expectations; parents overwhelmingly require a trusted human physically present, creating strong organizational and social barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI cannot perform this task at any cost today, making cost comparison moot. A human nanny's loaded wage is the only realistic option for ensuring child safety through active hazard management and rule enforcement. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical caregiving task, so AI cost is not comparable—human labor is the only option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product can autonomously perform hazard removal or boundary-setting with children. These require embodied presence, judgment under uncertainty, and the ability to adapt rules to individual child development—capabilities that exist only in research prototypes if at all. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical child-safety-proofing or in-person rule enforcement; this remains entirely outside current AI product capability. |
Perform first aid or cardiopulmonary resuscitation (CPR) when required.
3CI 0–5 · exposure 5 · augmentation 25 · importance 4.9/5 · click for rater detail
Perform first aid or cardiopulmonary resuscitation (CPR) when required.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This is a core safety-critical, physically-dependent task in childcare. No automation adoption is occurring because the task cannot be automated; human nannies remain legally and practically required. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Childcare and in-home caregiving is a low-digitization, highly physical sector with essentially no adoption of AI for emergency physical response tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can provide real-time CPR guidance (compression rate, hand placement via video) to assist a human performer, but the assistance is limited to coaching and knowledge prompts, not replacing human judgment or physical action under time pressure. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could provide instructional support (e.g., an app talking a nanny through CPR steps) but offers minimal transformation of the actual physical task performance. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | CPR and first aid require real-time physical intervention on a human body—compressing a chest, opening an airway, positioning a patient. Current AI systems cannot physically act in the world and cannot perform these interventions end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires immediate physical intervention on a child in a medical emergency, which no current AI system can perform since it lacks a physical embodiment capable of administering first aid or CPR. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Performing CPR and first aid on a child requires legal authority, medical training certification, and duty-of-care responsibility. Liability, licensing requirements, and the necessity of a trained human present create hard barriers to any non-human substitution. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Emergency physical care requires immediate human presence and, in many jurisdictions, certified CPR/first-aid training; liability and safety concerns make this a hard barrier to any automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI cannot replace the physical act of CPR or first aid, so cost comparison is not meaningful. A robot capable of performing these tasks at human competence would be extremely expensive; guidance systems do not substitute for the human performer. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI alternative capable of performing this physical task, so cost comparison is moot—only a present human can act, making AI infinitely more 'expensive' in the sense of being impossible. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI can guide CPR via audio/video instruction (e.g., dispatch systems), but this is human-assisted guidance, not autonomous performance. No deployed AI system performs CPR or first aid independently; such tasks remain research-stage or tele-guidance prototypes. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical first aid or CPR; this remains purely a human physical-response task with no robotic or AI substitute in production. |
Regulate children's rest periods and nap schedules.
3CI 0–5 · exposure 0 · augmentation 25 · importance 4.8/5 · click for rater detail
Regulate children's rest periods and nap schedules.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task occurs in household and small childcare settings with minimal digitization; the nature of childcare work remains centered on direct human supervision and physical care, with no meaningful AI adoption in this domain. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare is a low-digitization, high-touch physical sector with essentially no AI adoption for direct caregiving tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide modest assistance through reminder systems or scheduling tools, but the core task of physically implementing and monitoring nap schedules requires human judgment and presence, limiting augmentation value. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI apps can help schedule nap times, track patterns, or send reminders, but they offer only marginal support to the actual caregiving task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Regulating children's rest periods requires real-time physical presence, responsiveness to individual child needs, and dynamic adjustment to unpredictable behavior. Current AI cannot physically implement or enforce nap schedules or respond to a child's resistance or medical needs. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires continuous in-person physical presence, observation, and real-time judgment about a specific child's needs, which no current AI system can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and ethical barriers exist: caregivers have a duty of care obligation to supervise children, and parents expect licensed or trusted human supervision during rest periods. Liability concerns and child welfare regulations create hard barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Direct physical care of children involves strong safety, trust, and in-person supervision requirements that create a high barrier to any non-human substitution, though not a formal licensing requirement everywhere. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI cannot perform this task at any cost, as it fundamentally requires human physical presence and real-time responsiveness to a vulnerable population. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical caregiving task, so AI cost is not comparable; the human is the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously regulate children's rest periods and nap schedules; this requires in-person supervision, comfort measures, and judgment that no current AI system can provide reliably or safely. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product manages physical childcare routines like putting children down for naps or monitoring their rest in real time. |
Model appropriate social behaviors and encourage concern for others to cultivate development of interpersonal relationships and communication skills.
3CI 0–5 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail
Model appropriate social behaviors and encourage concern for others to cultivate development of interpersonal relationships and communication skills.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare sectors remain heavily reliant on in-person human labor with minimal AI displacement. Parental preference, regulatory requirements, and the irreducible need for human supervision mean adoption velocity is extremely low. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare is a low-digitization, physically embodied, small-scale service sector with minimal AI adoption for this kind of task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI might marginally assist by suggesting social activities or providing reference material on child development, but it offers minimal direct augmentation to a nanny's core work of modeling behaviors and cultivating relationships through lived interaction. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI apps or media could supplement some social-skill learning content, but the nanny's core act of modeling behavior and relational engagement is barely aided by current tools. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time emotional attunement, modeling complex social behaviors, and responding authentically to a child's developmental needs—capabilities that current AI systems cannot perform end-to-end. AI cannot sustain the consistent, embodied social presence necessary for genuine relationship-building with children. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires an embodied, ongoing human presence to model behavior for a child in real-time physical and emotional contexts, which current AI cannot replicate or perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal, regulatory, and social barriers protect this role: parents and childcare regulations require licensed or screened human caregivers to supervise children, and liability concerns around unsupervised AI interaction with minors create hard constraints on substitution. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Childcare involves strong parental trust, safety, and often licensing/background-check norms plus an inherent human-contact requirement for socialization and safeguarding, creating high barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Even if AI could technically assist with this task, the cost of deploying, monitoring, and ensuring appropriateness would likely exceed the value delivered, and human nannies remain far more cost-effective for the full scope of social modeling and relationship cultivation. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute providing this service, so cost comparison favors the human by default since AI cannot deliver the output at all. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product reliably performs this task in production. While chatbots can discuss social skills, they cannot model behavior, read emotional cues in real time, or form genuine interpersonal relationships—the core of what this task demands. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product provides in-person social modeling and relationship-building for children; this remains far outside current AI product capability. |
Organize and conduct age-appropriate recreational activities, such as games, arts and crafts, sports, walks, and play dates.
3CI 0–5 · exposure 5 · augmentation 38 · importance 4.6/5 · click for rater detail
Organize and conduct age-appropriate recreational activities, such as games, arts and crafts, sports, walks, and play dates.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nanny work is low-digitization, typically individual-household employment with limited institutional adoption of automation; families strongly prefer live human caregivers for child safety and developmental interaction. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare is a low-digitization, physically-embedded sector with essentially no AI adoption for this activity. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by suggesting activities, tracking age-appropriateness, and providing craft or game ideas, helping a nanny plan more efficiently; however, the core execution remains dependent on human judgment and presence. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can suggest activity ideas, craft plans, or games in advance, offering modest planning assistance, but cannot help during live execution. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time responsiveness to individual children's developmental stages, safety judgment, and dynamic adaptation to changing moods and interests—all deeply dependent on in-person interaction and nuanced judgment that current AI cannot perform end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, hands-on supervision, and real-time interpersonal engagement with children that no current AI system can perform end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Child safety, duty of care, and supervision laws create hard legal barriers—a responsible adult must always be physically present and accountable for child welfare; no jurisdiction permits unsupervised AI-led childcare. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Direct physical supervision of children involves safety, liability, and trust requirements that create near-absolute barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI tools for activity planning are cheap but represent only a small fraction of the nanny's work; full automation would still require paying a human to be physically present with the children, making AI cost-prohibitive relative to the loaded wage. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so cost comparison favors the human by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can suggest age-appropriate activities or generate craft ideas via chatbots, no deployed system can independently organize and conduct recreational activities with children; any such system would require continuous human supervision and intervention. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product conducts in-person childcare recreational activities; this remains entirely outside current AI product scope. |
Assign appropriate chores and praise targeted behaviors to encourage development of self-control, self-confidence, and responsibility.
3CI 0–5 · exposure 0 · augmentation 25 · importance 4.5/5 · click for rater detail
Assign appropriate chores and praise targeted behaviors to encourage development of self-control, self-confidence, and responsibility.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This is a task embedded in personal, in-home childcare that depends on relationship, presence, and parental authority. Sectors employing nannies show minimal AI adoption because human presence and trust are non-negotiable. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Childcare/nanny work is a low-digitization, physically embodied, small-scale employment sector with minimal AI adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | A nanny might use AI for suggestions on child development strategies or chore ideas, but the core work—observing behavior, choosing when and how to praise, and building a child's confidence—remains deeply human. Augmentation potential is limited because judgment and emotional attunement cannot be meaningfully offloaded. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could offer suggestions on age-appropriate chores or positive reinforcement strategies via apps, but it cannot meaningfully assist during the live behavioral interaction itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time behavioral observation, relationship-based judgment about individual child development, and emotionally calibrated responsiveness—none of which current AI can perform autonomously. AI cannot reliably assess which chores suit a particular child's developmental stage or deliver praise in ways that meaningfully reinforce self-control and confidence in an ongoing relationship. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires real-time in-person interaction, physical presence, and relational judgment with a specific child that no current AI system can perform end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Parents and guardians have strong legal responsibility for child rearing and development; direct human care, judgment, and relationship are culturally and legally essential. No parent would delegate core behavioral guidance and emotional reinforcement to an automated system. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Childcare involves strong safety, trust, and in-person supervisory requirements; parents overwhelmingly require a present, trusted human caregiver for children. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI has no meaningful capability to perform this task, so cost comparison is not applicable; human nannies remain the only option. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute delivering this in-person care task, so cost comparison favors the human by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can independently assign chores and deliver targeted behavioral praise to children in a way that produces documented developmental outcomes. Current systems lack the continuous embodied presence, real-time behavior observation, and relationship history required for this task. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs in-person childcare behavioral guidance; this remains entirely outside the scope of current AI products. |
Instruct and assist children in the development of health and personal habits, such as eating, resting, and toilet behavior.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.8/5 · click for rater detail
Instruct and assist children in the development of health and personal habits, such as eating, resting, and toilet behavior.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare remains predominantly human-staffed and shows minimal AI displacement. The sector has low digital maturity, high regulatory scrutiny, and strong cultural preference for human caregivers; adoption of any automation remains negligible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare is a low-digitization, high-touch physical sector with essentially no AI/agent adoption for hands-on tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide marginal assistance via reminders about meal times or developmental milestones, or educational content about healthy habits, but cannot materially augment the core instruction and hands-on assistance work. The human remains fully responsible for execution. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI apps can offer scheduling reminders, developmental tips, or habit-tracking suggestions to nannies, but this offers only marginal assistance to the core physical task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task fundamentally requires real-time physical interaction, emotional responsiveness, and behavioral intervention with children—functions no current AI system can perform. AI cannot feed, toilet, bathe, or physically comfort a child, nor can it provide the consistent human presence necessary for habit formation in young children. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, hands-on supervision, and real-time responsive care of a child's bodily needs and behavior, which current AI cannot perform at all end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and liability barriers protect this task: childcare workers are typically subject to background checks, training requirements, and direct duty-of-care obligations. Parents and regulators have high error-cost sensitivity, and jurisdictions impose strict standards on who may supervise children's safety and development. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Direct physical childcare, especially around toileting and safety-sensitive routines, requires trusted human presence, background-checked caregivers, and parental comfort with in-person supervision, creating strong practical and safety barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems have no cost advantage for this task because they cannot perform the core childcare function at all. The loaded wage of a nanny far exceeds any conceivable AI infrastructure cost, but that comparison is moot when AI cannot substitute for the actual work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so AI cost is not comparable—human labor is the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs childcare instruction and assistance tasks end-to-end. While AI can provide parenting advice or reminders, it cannot execute the actual care and behavioral guidance this task demands, which requires physical presence and adaptive real-time interaction. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically instructs or assists children with eating, resting, or toileting; this remains purely a human caregiving function. |
Observe children's behavior for irregularities, take temperature, transport children to doctor, or administer medications, as directed, to maintain children's health.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.5/5 · click for rater detail
Observe children's behavior for irregularities, take temperature, transport children to doctor, or administer medications, as directed, to maintain children's health.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Childcare is a low-digitization, high-human-contact sector with strong preference for human caregivers, regulatory constraints on automation, and significant liability exposure. Adoption of AI in this domain remains negligible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | In-home childcare is a low-digitization, physically-embedded sector with essentially no AI displacement occurring. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could potentially assist with logging observations (e.g., a caregiver describes behavior and AI helps flag concerning patterns) or medication reminders, but these are peripheral to the core task and require the human to remain fully responsible for judgment and action. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help with reminders for medication schedules or symptom-checking information, but offers minimal assistance for the core physical monitoring and caregiving actions. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time physical observation of children, direct physical contact for temperature-taking and medication administration, transportation, and judgment about behavioral irregularities in a dynamic, safety-critical context. Current AI cannot perform any of these elements end-to-end without human presence. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, hands-on caregiving, and real-time judgment about a child's health and safety that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | This task has hard legal and regulatory barriers: administering medications requires licensure or explicit parental/medical authorization; transporting minors involves liability and duty-of-care requirements; observing children's health and determining when medical intervention is needed involves both professional standards and parental trust that legally and practically require a responsible human present. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Direct physical care of children, medication administration, and health monitoring carry strong liability, safety, and often legal/licensing expectations requiring a responsible human caregiver. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The loaded cost of a nanny is substantial, but the infrastructure needed for an AI system to perform physical health monitoring, medication administration, and transportation (robotics, sensing, legal liability coverage) would exceed the cost of human childcare by orders of magnitude. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical caregiving task, so AI cost is not comparable to human wage for the same output. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously observe children's behavior for health irregularities, physically measure temperature, administer medications, or transport children. These require embodied presence, physical interaction, and real-time human judgment in unstructured environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product administers medication, transports children, or physically checks temperatures; this remains purely a human physical-care task. |
Transport children to schools, social outings, and medical appointments.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail
Transport children to schools, social outings, and medical appointments.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Nanny services remain in low-digitization household-service sectors with minimal AI adoption. Parents retain strong preference for human supervision of children, and the task involves physical presence in high-liability contexts where automation adoption has been negligible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Domestic childcare and personal transport of minors is a low-digitization, physical-world sector with essentially no AI displacement occurring. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Route-planning apps and real-time traffic information assist human nannies moderately, but they do not substantially transform the core transportation-supervision task. The human remains fully responsible for driving and child oversight, limiting meaningful productivity gain. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with scheduling, route planning, or reminders for appointments, but offers minimal assistance to the core physical task of transporting and supervising a child. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Transportation of children requires physical vehicle operation, route navigation, real-time decision-making under uncertainty, and supervision of minors—none of which current AI systems can perform end-to-end. While route planning is automatable, the core task (driving + childcare supervision) remains entirely human-dependent. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, driving, and real-time childcare supervision that current AI systems cannot perform end-to-end; no software substitute exists for physically transporting a child. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Hard legal and moral barriers exist: parents are liable for child safety during transport, childcare licensing regulations typically require a trained human supervisor present, and no jurisdiction permits unsupervised child transport by AI. These are not merely regulatory friction but fundamental legal requirements. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Child safety, custody, and duty-of-care laws combined with parental trust and liability concerns create hard requirements for a trusted human caregiver to be present. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Even if autonomous vehicles were available, the integration, liability coverage, insurance, and oversight costs would far exceed the loaded wage of a nanny ($25–40k/year in most markets). Current autonomous vehicle operating costs are multiples higher than human labor. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-only solution, so any comparison is moot; even robotaxis require human oversight and are not designed for unsupervised child transport, making AI substitution more costly or infeasible. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs child transportation autonomously; self-driving vehicles remain in limited pilot stage without proven safety at scale, and no system handles the child supervision component. This remains a research-stage capability with significant liability and safety barriers. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product physically transports children; autonomous vehicles are not authorized or trusted to independently supervise and transport unaccompanied minors. |
Related occupations — Personal Care & Service
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.