Community Health Workers

21-1094.00

Median wage $51,850/yr61,660 employed (US)Rank #598 of 923 scored · top 65% by substitution

Promote health within a community by assisting individuals to adopt healthy behaviors. Serve as an advocate for the health needs of individuals by assisting community residents in effectively communicating with healthcare providers or social service agencies. Act as liaison or advocate and implement programs that promote, maintain, and improve individual and overall community health. May deliver health-related preventive services such as blood pressure, glaucoma, and hearing screenings. May collect data to help identify community health needs.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution23

Exposure20

Augmentation53

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

28 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

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%21

panel mean rating 1.8/5 → substitution pressure 21/100

Technical feasibility todayw 20%18

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

Cost vs. human wagew 15%25

panel mean rating 2.0/5 → substitution pressure 25/100

Adoption barriersw 20%inverted — strong barriers lower the score32

panel mean rating 3.7/5 (barrier strength) → substitution pressure 32/100

Sector adoption velocityw 10%19

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

Task breakdown (28 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.

Refer community members to needed health services.

CI 30–80 · exposure 58 · augmentation 88 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Health systems and nonprofits are piloting AI-driven referral and care navigation, but adoption remains uneven. Community health programs are often under-resourced, digital, and slower to integrate new tools compared to hospitals or insurers; production-scale displacement is limited.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, relationship-driven field with limited AI tool adoption compared to fast-moving sectors like finance or tech.
Augmentation potential	claude-haiku-4-5-20251001	5/5	AI referral tools significantly augment CHWs by instantly surfacing appropriate services, eligibility info, and appointment availability, freeing them to focus on outreach, relationship-building, and addressing barriers to care enrollment. This is a strong augmentation use case even where full automation isn't desired.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully assist CHWs by quickly identifying relevant services, eligibility rules, and referral pathways, saving research time while the CHW manages the relationship and final decision.
Task automatability	claude-haiku-4-5-20251001	5/5	AI systems can reliably match community members to health services through intake questionnaires, symptom/need assessment, and deterministic or ML-based referral logic. This task involves information gathering and matching against service directories—both well within current AI capabilities with ≥50% time savings.
Task automatability	claude-sonnet-5	2/5	Referral involves synthesizing personal context, trust, and judgment about local resources, which AI cannot fully replicate end-to-end despite being able to draft lists of services or eligibility criteria.
Adoption barriers	claude-haiku-4-5-20251001	3/5	While referral itself has no hard licensing requirement, meaningful barriers exist: organizational reluctance to fully disintermediate human workers, trust requirements for vulnerable populations, and liability concerns if referrals are inaccurate. Care coordination often has informal accountability to the individual.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement for referrals specifically, but liability concerns, community trust needs, and reliance on personal relationships create moderate friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	5/5	AI-powered referral matching (via APIs, chatbots, or agents) costs pennies per transaction in inference and integration. Community health worker wages are $25–$40k annually; the cost ratio heavily favors AI across any volume.
Cost vs. human wage	claude-sonnet-5	2/5	While automated lookup tools are cheap, the human relationship-building, trust, and follow-up embedded in this task keep the effective cost of full automation high relative to a CHW's wage for equivalent outcomes.
Technical feasibility today	claude-haiku-4-5-20251001	4/5	Multiple deployed products (211.org referral databases, health navigation chatbots, EHR-integrated referral systems) perform this task operationally. Some systems still require human verification of complex cases or multi-condition referrals, preventing a perfect 5, but mainstream deployment is established.
Technical feasibility today	claude-sonnet-5	2/5	Chatbots and referral databases exist but are narrow in scope and not widely deployed to reliably match community members to appropriate services in real-world CHW workflows.

Contact clients in person, by phone, or in writing to ensure they have completed required or recommended actions.

CI 29–52 · exposure 38 · augmentation 63 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health sectors are typically low-digitization, small-organization environments with limited AI adoption. While some automated reminders are used, deep displacement of in-person outreach verification remains minimal in public health and community-based organizations.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and social services are a low-digitization, underfunded sector with slow AI adoption; automated reminder tools exist but are not deeply or widely integrated.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI tools can assist by tracking contacts, generating outreach templates, and flagging high-risk cases for prioritization, improving worker productivity in managing large caseloads. However, the human worker remains essential for actual relationship-building and verification of action completion.
Augmentation potential	claude-sonnet-5	4/5	AI-driven reminder systems, CRM tools, and automated outreach can significantly reduce the burden of routine follow-up contacts, letting workers focus on higher-need clients and complex cases.
Task automatability	claude-haiku-4-5-20251001	2/5	While automated systems can generate contact attempts (calls, emails, SMS), verifying completion of required actions and ensuring genuine client engagement requires human judgment and relationship-building. The task is only partially automatable without significant human oversight.
Task automatability	claude-sonnet-5	3/5	AI can automate the outreach/reminder aspect (calls, texts, emails) via automated systems, but in-person contact and nuanced follow-up conversations with vulnerable populations still require human judgment and trust-building.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Community health worker roles are often embedded in trust-based, face-to-face relationships with vulnerable populations; many organizations and regulations prioritize human contact and cultural competency. There is also organizational friction in replacing the irreplaceable interpersonal element of ensuring compliance in health equity contexts.
Adoption barriers	claude-sonnet-5	2/5	No licensing requirement blocks automated reminders, but community health work often depends on trust and personal relationships with underserved populations, creating moderate organizational and effectiveness-based friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	3/5	Automated outreach (phone systems, SMS platforms) costs significantly less per contact than human workers, but integration and failure handling require oversight, making the total cost comparison roughly equivalent when accounting for the need for human follow-up on compliance verification.
Cost vs. human wage	claude-sonnet-5	3/5	Automated messaging/reminder systems are cheap per contact, but oversight, escalation handling, and the portion requiring human relational work keep overall costs comparable rather than dramatically cheaper.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Deployed chatbots and automated reminders exist, but they cannot reliably replace in-person community health worker outreach or handle the nuanced assessment of whether actions are truly completed. Production systems exist for basic reminder calls but not for the full task.
Technical feasibility today	claude-sonnet-5	3/5	Automated reminder systems, chatbots, and IVR/text campaigns are deployed in healthcare settings today, but they have material limitations in handling complex client needs, building rapport, or addressing social determinants of health.

Maintain updated client records with plans, notes, appropriate forms, or related information.

CI 25–52 · exposure 38 · augmentation 63 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health sectors remain among the least digitized and slowest to adopt advanced automation, often operating in under-resourced settings with legacy paper or basic EHR systems. Pilot programs exist, but production-scale AI-driven record management in this sector remains rare and patchy.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and social services organizations tend to have lower digitization and slower AI tool adoption compared to fast-adopting sectors like finance or professional services.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist CHWs by drafting notes from voice or text input, suggesting form fields to complete, and flagging missing information—moderately useful productivity gains. However, the worker must still review, edit, and ensure accuracy and appropriateness, so augmentation is helpful but not transformative.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully assist by transcribing visit notes, auto-filling standard forms, and organizing client information, significantly speeding up documentation while the worker retains oversight for accuracy and judgment calls.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist with data entry and form population from structured information, it cannot reliably maintain comprehensive client records that require clinical judgment, confidentiality sensitivity, and integration of complex patient history into appropriate care plans. A human must interpret nuanced case details and ensure legal/clinical appropriateness of what gets recorded.
Task automatability	claude-sonnet-5	3/5	Structured record-keeping and note organization can largely be handled by AI transcription and templating tools, but ensuring accuracy, appropriate categorization, and integration with case plans still requires human review, so only partial time savings are achievable without significant setup.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Health records are heavily regulated under HIPAA, state regulations, and professional standards; community health workers must maintain client confidentiality and accuracy in medical documentation, which carries liability for errors. Many jurisdictions require a credentialed health professional to sign or validate records, creating a legal barrier to full automation.
Adoption barriers	claude-sonnet-5	2/5	Records may be linked to compliance or privacy regulations (HIPAA-like protections), but the task itself does not require a licensed professional’s signature, so barriers are moderate rather than strict legal requirements.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	EHR software and AI-assisted documentation tools are expensive to implement and maintain, and still require significant human time for review and correction. The all-in cost (licensing, integration, validation oversight) often exceeds the labor savings from automation, especially in resource-limited community health settings.
Cost vs. human wage	claude-sonnet-5	3/5	AI documentation tools reduce time spent on notes but still require licensing, integration with case management systems, and human verification, making the cost roughly comparable rather than dramatically cheaper.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Electronic health record (EHR) systems exist with templating and some auto-population features, but current AI cannot independently manage the full record lifecycle with confidence—judgment calls about what details matter, which forms apply, and how to synthesize notes into coherent care plans remain error-prone and require human oversight.
Technical feasibility today	claude-sonnet-5	3/5	AI scribe and EHR-assist products (e.g., ambient documentation tools) exist and are used in healthcare settings, but community health work involves varied, non-standardized forms and informal notes that reduce reliability of off-the-shelf deployment.

Advise clients or community groups on issues related to improving general health, such as diet or exercise.

CI 29–37 · exposure 25 · augmentation 75 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health organizations are typically slower to digitize and adopt AI, especially in underserved areas where community health workers are most active. Adoption remains limited to pilot projects rather than widespread production deployment.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and public health outreach sectors have low digitization and slow AI adoption relative to finance or professional services.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can assist community health workers by drafting tailored talking points, summarizing evidence-based guidelines, suggesting follow-up questions, and helping organize health information for groups—substantially boosting worker productivity while the worker maintains client relationships and contextual judgment.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully help CHWs by generating educational materials, translating health information, and personalizing advice content, boosting their efficiency while they retain the interpersonal role.
Task automatability	claude-haiku-4-5-20251001	2/5	AI can generate generic health advice on diet and exercise at scale, but lacks the contextual understanding, rapport-building, and individual adjustment needed for effective community health counseling. The task requires adapting messaging to cultural norms, literacy levels, and interpersonal trust—factors that significantly limit unattended AI performance.
Task automatability	claude-sonnet-5	2/5	General diet/exercise advice can be drafted by AI, but effective community health advising requires trust-building, cultural tailoring, and in-person rapport that current systems cannot replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Community trust, relationships, and cultural competency are primary barriers to full automation. Many clients prefer human contact, and community health work often requires professional licensure or authorization. Liability concerns around health advice and regulatory oversight of health guidance create friction against full substitution.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement in most jurisdictions, but community trust, cultural competency, and human-contact expectations create real friction against pure AI substitution.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI inference and integration cost per advice-delivery session is substantially lower than the loaded wage of a community health worker conducting counseling sessions. However, oversight and verification that advice is appropriate for the specific community remains necessary.
Cost vs. human wage	claude-sonnet-5	3/5	AI-generated generic health content is cheap to produce, but the value of CHWs lies in personalized, trust-based delivery, so cost comparison is muddled and not clearly favorable to AI once oversight is included.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Chatbots and AI health advisors exist and can deliver standard guidance, but deployed systems suffer from limited personalization, difficulty handling local/cultural contexts, and low trust adoption in real community settings. No mature production system reliably replaces a community health worker's advisory role.
Technical feasibility today	claude-sonnet-5	2/5	Health chatbots and apps exist for generic wellness tips, but no deployed product reliably substitutes for a community health worker's contextualized, relationship-based counseling at scale.

Advise clients or community groups on issues related to diagnostic screenings, such as breast cancer screening, pap smears, glaucoma tests, or diabetes screenings.

CI 25–37 · exposure 25 · augmentation 75 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Healthcare and community health organizations are slower to adopt AI for direct advisory due to liability concerns, regulatory oversight, and the cultural importance of human relationships in health promotion. Pilot projects exist, but production displacement remains limited.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, relationship-driven field with slow, uneven AI adoption compared to fast-adopting sectors like finance or professional services.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can powerfully assist community health workers by generating tailored screening reminders, providing multilingual fact sheets, summarizing clinical guidelines, and flagging high-risk individuals—substantially raising counselor productivity while the human retains relationship and judgment. This augmentation is already seeing real-world deployment.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully help CHWs by providing up-to-date screening guidelines, translating materials, and drafting educational content, boosting their efficiency while they retain the interpersonal advising role.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can provide factual information about screening guidelines and procedures, advising on personal health screening decisions requires understanding individual risk factors, cultural contexts, and motivational barriers that demand human judgment and relationship-building. Current systems lack the contextual depth and trustworthiness needed to handle this counseling role end-to-end.
Task automatability	claude-sonnet-5	2/5	While AI can generate accurate general information about screenings, the task involves trust-building, personalized motivational counseling, and community-specific cultural navigation that current AI cannot replicate end-to-end at equal quality.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Community health worker roles are embedded in trust, cultural competency, and ongoing relationships; regulatory bodies and healthcare organizations often require human touch for health advisory to vulnerable populations. Liability and organizational reluctance to replace trust-based community outreach create substantial adoption friction.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement for CHWs, but health advising carries liability concerns, community trust dynamics, and organizational reliance on human rapport that create moderate friction against AI substitution.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI-generated screening information and reminders cost a small fraction of a human community health worker's loaded wage, making automation orders of magnitude cheaper for the information-delivery component. However, the advisory relationship itself is labor-intensive by design.
Cost vs. human wage	claude-sonnet-5	2/5	AI content generation is cheap, but the human relationship, trust, and in-person community engagement components still require paid staff time, keeping overall costs comparable rather than dramatically lower.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Deployed chatbots can deliver screening information, but no production systems reliably handle the nuanced, personalized advisory aspect of this task—including assessing readiness, addressing hesitation, and adapting to language and cultural needs. Tools exist to share facts but not to counsel in the manner community health workers do.
Technical feasibility today	claude-sonnet-5	2/5	Chatbots and health information tools exist and can answer screening questions, but no deployed product reliably performs the relational advising and community outreach role of a CHW in production at scale.

Advise clients or community groups on issues related to sanitation or hygiene, such as flossing or hand washing.

CI 23–37 · exposure 25 · augmentation 50 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health sectors remain low-tech and relationship-driven, with slow digital transformation outside major health systems. Adoption of AI for direct client counseling is minimal; human workers remain the default model in most community settings.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work occurs in low-digitization, often under-resourced public health and nonprofit settings with minimal AI agent deployment in production.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist community health workers by drafting culturally appropriate materials, suggesting talking points, or generating personalized reminders for clients, but the core advising role remains human-centered and the impact is incremental rather than transformative.
Augmentation potential	claude-sonnet-5	3/5	AI can help health workers by providing educational materials, translated content, or quick reference information to support their client interactions.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can generate factual health advice about sanitation and hygiene, this task inherently requires interpersonal communication, behavior change motivation, and responsiveness to individual circumstances. Current AI cannot replicate the relationship-building and cultural sensitivity needed for effective client counseling at 50% time savings.
Task automatability	claude-sonnet-5	2/5	AI can generate hygiene education content and answer questions, but the task involves in-person trust-building, cultural sensitivity, and community engagement that current systems cannot fully replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Community health work often requires licensure, certification, and legal liability for advice given; many jurisdictions restrict health counseling to credentialed workers. Trust and cultural authority also function as barriers—communities typically prefer human advisors for sensitive health topics.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement, but community health work relies heavily on personal relationships, trust, and cultural competency that create organizational and social friction against automation.
Cost vs. human wage	claude-haiku-4-5-20251001	4/5	AI-generated hygiene guidance is nearly free to deploy at scale once built, whereas community health workers require wages and ongoing training. The cost per unit of basic health messaging delivered is orders of magnitude lower with AI.
Cost vs. human wage	claude-sonnet-5	2/5	While generating hygiene content is cheap via AI, the actual delivery requires human presence, community trust, and contextual adaptation, so all-in cost savings are limited.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Chatbots and health information systems can deliver hygiene content, but deployed products lack the contextual understanding, real-time adaptation, and human trust critical for actual community health advising. Existing systems function as information sources, not credible counselors.
Technical feasibility today	claude-sonnet-5	2/5	Chatbots and health information apps exist for basic hygiene Q&A, but no deployed product reliably substitutes for a community health worker's in-person advisory role at scale.

Collect information from individuals to compile vital statistics about the general health of community members.

CI 30–30 · exposure 25 · augmentation 50 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health sectors are typically lower-digitization, smaller-budget organizations; adoption of AI for data collection remains in pilot phase rather than mainstream production across the field.
Sector adoption velocity	claude-sonnet-5	2/5	Public health and community outreach sectors are slow AI adopters, particularly for field-based, relationship-driven data collection tasks.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by drafting intake forms, flagging missing fields, or organizing collected data, raising worker productivity on documentation; however, the core task of eliciting accurate information from community members remains human-centered.
Augmentation potential	claude-sonnet-5	3/5	AI-enabled forms, translation tools, and data entry systems can meaningfully speed up compiling and organizing collected health statistics even though collection itself remains human-led.
Task automatability	claude-haiku-4-5-20251001	2/5	Data collection from individuals requires dynamic interaction, contextual understanding, and rapport-building that current AI handles poorly; while forms could be partially automated, the nuanced follow-up, cultural sensitivity, and verification needed for accurate health statistics remain largely manual tasks.
Task automatability	claude-sonnet-5	2/5	Data collection involves in-person interviews, trust-building, and outreach in community settings that current AI cannot autonomously conduct end-to-end, though digital intake forms can partially automate data capture.
Adoption barriers	claude-haiku-4-5-20251001	3/5	HIPAA and state health privacy regulations require careful handling of personal health data, and community trust often demands human contact; however, these are organizational and regulatory constraints rather than absolute prohibitions on AI assistance.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing barrier, but community trust, language/cultural competency, and door-to-door engagement create significant organizational friction against pure automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI-driven survey and intake systems have significant setup, compliance, and oversight costs; for community health work with low-income populations, the human relationship and manual verification still undercut a major cost advantage.
Cost vs. human wage	claude-sonnet-5	2/5	Human community health workers are often low-cost already and provide trust-based access that AI tools cannot replicate without added human oversight, limiting cost savings.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Chatbots and survey tools exist but struggle with real-world health data collection requiring trust, clarification of ambiguous responses, and handling of sensitive information; no mature product reliably performs this end-to-end in community health settings at scale.
Technical feasibility today	claude-sonnet-5	2/5	Digital health surveys and chatbot intake tools exist but are not widely deployed as replacements for in-person community health worker data collection, especially in underserved populations.

Advise clients or community groups on issues related to self-care, such as diabetes management.

CI 25–34 · exposure 25 · augmentation 63 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health worker sectors are typically low-tech, nonprofit, or government-dependent with limited digital infrastructure and slow technology adoption. While some organizations pilot chatbots for triage, production-level displacement remains minimal.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and social services are a lower-digitization sector with slow AI adoption relative to finance or professional services, though some pilots of health chatbots exist.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by providing fact sheets, symptom checkers, or reminders on evidence-based protocols that a CHW then contextualizes for their client. This raises productivity on information-delivery aspects, though the core advisory work remains human-led.
Augmentation potential	claude-sonnet-5	4/5	AI can provide CHWs with quick access to updated medical information, personalized care plans, translation, and follow-up reminders, meaningfully boosting their effectiveness while they retain the advising role.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can provide factual information on diabetes management protocols, the task requires personalized assessment of individual circumstances, cultural sensitivity, and behavioral coaching that demands human judgment. Current AI cannot reliably achieve 50% time savings at equal quality for the full advising task.
Task automatability	claude-sonnet-5	2/5	While AI chatbots can provide generic diabetes self-care information, effective advising requires building trust, reading nonverbal cues, and tailoring guidance to cultural and personal context that current systems cannot fully replicate end-to-end.itness
Adoption barriers	claude-haiku-4-5-20251001	4/5	Health advice carries legal liability and regulatory scrutiny; many jurisdictions require human accountability for health guidance. Community trust and cultural competence also create strong barriers—clients often need human relationships and can spot impersonal or inappropriate advice, limiting full substitution.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing barrier for CHW advice, but liability concerns around health guidance, community trust requirements, and preference for human contact create moderate friction.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Integration of AI into community health advisory requires oversight, customization for local contexts, and human validation of outputs. The total cost is comparable to or exceeds the wage of community health workers, especially when accounting for liability and supervision needs.
Cost vs. human wage	claude-sonnet-5	3/5	AI-driven health information tools are cheap to run, but achieving comparable trust and engagement often still requires human staff time for outreach and follow-up, keeping overall cost roughly comparable.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably performs full advisory on health self-care in production settings. AI can generate educational content or answer FAQs, but clinical decision support tools and chatbots have material limitations in handling complex cases and lack the rapport-building essential to community health work.
Technical feasibility today	claude-sonnet-5	2/5	Health information chatbots and apps exist and are used for patient education, but they are not deployed as substitutes for the relational, community-embedded advising role of CHWs at scale in production.

Interpret, translate, or provide cultural mediation related to health services or information for community members.

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

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Adoption remains low in community health settings, which tend to be under-resourced, geographically dispersed, and relationship-dependent. These sectors digitize slowly and prioritize human trust and cultural authenticity over cost reduction, limiting AI agent deployment.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, relationship-driven field with slow, uneven AI tool adoption compared to sectors like finance or professional services.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI translation and cultural background lookup tools can assist community health workers in preparing for conversations or drafting materials, improving efficiency on information-gathering tasks. However, the real work—interpreting and mediating in real-time—remains human-centered, so augmentation is moderate rather than transformative.
Augmentation potential	claude-sonnet-5	4/5	AI translation and information tools meaningfully speed up the literal interpretation portion of the task, letting CHWs focus more time on relationship-based cultural mediation.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can handle basic translation of health content, genuine cultural mediation requires understanding nuanced social contexts, beliefs, and trust-building that current systems cannot reliably replicate. The task fundamentally involves human judgment about what health information means in a specific cultural context, which AI cannot do end-to-end at 50% time savings and equal quality.
Task automatability	claude-sonnet-5	2/5	AI translation tools handle literal language conversion well, but cultural mediation requires trust-building, contextual judgment, and community relationships that current AI cannot replicate end-to-end.'
Adoption barriers	claude-haiku-4-5-20251001	4/5	Strong barriers exist: health information must be accurate and trustworthy, communities often require a known human intermediary for culturally sensitive discussions, and liability for mistranslation or cultural misstep falls on the organization. Regulatory frameworks and community trust norms heavily favor human-mediated culturally competent communication.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement typically governs this role, but liability concerns around health misinformation, community trust dynamics, and preference for human relationship-based mediation create meaningful friction.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI translation and cultural guidance generation are cheap per unit, but oversight, validation, and human review of culturally sensitive health information is labor-intensive. The total cost remains comparable to or higher than paying a community health worker, especially when liability and error correction are factored in.
Cost vs. human wage	claude-sonnet-5	3/5	AI translation is cheap and fast for the language component, but the human cultural mediation component still requires paid staff time, so blended cost savings are moderate rather than dramatic.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	General-purpose translation tools and LLMs exist and can provide rough cultural context, but no deployed product reliably performs the full mediation task—which requires real-time adaptation, relationship-building, and accountability for health outcomes. Products fail on subtle cultural appropriateness and trust establishment that production healthcare systems require.
Technical feasibility today	claude-sonnet-5	2/5	Deployed translation apps (Google Translate, medical interpreter tools) work reliably for literal translation, but no product reliably performs cultural mediation, which requires nuanced understanding of community values and interpersonal trust.

Distribute flyers, brochures, or other informational or educational documents to inform members of a targeted community.

CI 14–44 · exposure 28 · augmentation 50 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	This task occurs in community-based, often under-resourced health sectors with low digitization and limited capital for technological substitution. Adoption of any physical automation remains near zero.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, high-touch field with slow AI adoption for outreach tasks, though content creation tools are creeping in.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist with identifying target locations, optimizing routes, or generating document content, but these are peripheral to the core task of physical distribution. Meaningful augmentation is limited by the task's inherent reliance on human presence.
Augmentation potential	claude-sonnet-5	4/5	AI can significantly speed up drafting, translating, and formatting educational materials, freeing workers to focus on distribution and community engagement.
Task automatability	claude-haiku-4-5-20251001	2/5	Physical distribution of materials requires in-person presence and environmental navigation that current AI cannot perform end-to-end. While content creation or decision-making about which documents to distribute could be partially automated, the core task—placing materials in hands or on doors—remains fundamentally non-automatable without specialized robotics not yet deployed at scale.
Task automatability	claude-sonnet-5	3/5	AI can generate content and even orchestrate digital distribution (email, social media), but physical flyer distribution and in-person community outreach require human presence, so only part of the task is automatable.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Community trust and interpersonal contact are core to effective health education outreach. Regulatory and organizational norms strongly prefer human community members engaging with populations, creating both cultural and institutional friction against substitution.
Adoption barriers	claude-sonnet-5	2/5	No licensing requirement, but community trust, local knowledge, and physical presence create moderate friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Human community health workers performing this task cost significantly less than any robotic or AI-driven physical distribution solution that might theoretically handle it, making automation economically unfavorable even if technically feasible.
Cost vs. human wage	claude-sonnet-5	2/5	AI can cheaply produce content, but the physical distribution component still requires paid human labor or logistics, keeping overall cost comparable to human-only approaches.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No current deployed AI system or product performs physical flyer distribution in real communities. The task requires mobile manipulation in unstructured environments, a capability that remains at research stage rather than production deployment.
Technical feasibility today	claude-sonnet-5	2/5	Tools exist for designing materials and automating digital sends, but no deployed product handles the physical distribution or targeted community canvassing aspect reliably.

Identify or contact members of high-risk or otherwise targeted groups, such as members of minority populations, low-income populations, or pregnant women.

CI 25–30 · exposure 25 · augmentation 50 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Adoption of AI-assisted identification is emerging in large health systems, but actual automated contact deployment is minimal. Most community health work remains low-digitization and human-centric; regulatory and trust constraints slow substitution.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and social services sectors are slow adopters of AI tools, particularly for outreach to vulnerable or marginalized populations where trust is paramount.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by flagging high-risk individuals from data, generating candidate lists, and suggesting outreach strategies, allowing CHWs to prioritize and focus on relationship-building and personalized contact. This raises worker efficiency on the identification side while leaving the critical human contact task intact.
Augmentation potential	claude-sonnet-5	3/5	AI can help analyze demographic/health data to identify at-risk groups and support outreach logistics (e.g., scheduling, translation), meaningfully aiding but not replacing the human-centered contact work.
Task automatability	claude-haiku-4-5-20251001	2/5	Identifying members of targeted groups from existing data (databases, registries, demographic records) is partially automatable via filtering and segmentation, but contacting them reliably requires human judgment about cultural context, trust-building, and appropriate messaging. Current AI cannot handle the full contact workflow end-to-end with equal quality.
Task automatability	claude-sonnet-5	2/5	Identifying targets via data analysis could be AI-assisted, but actually contacting and engaging vulnerable populations requires trust-building, cultural competence, and human relationship skills that AI cannot replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Significant barriers exist: HIPAA and privacy regulations restrict automated contact of protected health information; community trust and cultural sensitivity often legally or practically require a licensed/trained human; liability for inappropriate outreach or data misuse; and organizational policy typically mandates human judgment in contacting vulnerable populations.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement specifically for this task, but privacy regulations (HIPAA), community trust dynamics, and preference for human contact in sensitive outreach create meaningful friction.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI tools for data segmentation are relatively cheap, but the contact and engagement portion still requires human community health workers, who are essential for trust and cultural competency. The overall cost savings are modest since labor-intensive outreach remains required.
Cost vs. human wage	claude-sonnet-5	2/5	While data screening tools are cheap, the actual outreach and trust-building work still requires human labor, so overall cost savings are limited relative to the full task.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Data-driven identification tools exist in healthcare systems, but outreach and contact initiation remain largely manual. No deployed product reliably handles both identification and meaningful contact at scale; existing systems support segmentation only, leaving contact execution to humans.
Technical feasibility today	claude-sonnet-5	2/5	Some products exist for population health analytics and outreach automation (SMS/robocalls), but reliable identification and meaningful contact with hard-to-reach populations remains largely manual and relationship-driven in practice.

Teach classes or otherwise disseminate medical or dental health information to school groups, community groups, or targeted families or individuals, in a manner consistent with cultural norms.

CI 25–30 · exposure 25 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health worker sectors are typically lower-digitization, community-embedded settings with strong preference for human relationships and cultural continuity. Adoption of AI-led health education remains pilot-stage; most production systems still rely on human CHWs for trust and cultural authenticity.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and public health outreach sectors are generally slow adopters of AI compared to finance or professional services, with pilots more common than production deployment.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can significantly assist CHWs by generating multilingual materials, translating content into accessible formats, preparing culturally-informed talking points, and handling administrative follow-up, freeing the human educator to focus on engagement, trust-building, and real-time responsiveness to the specific audience.
Augmentation potential	claude-sonnet-5	4/5	AI can significantly help CHWs by drafting culturally adapted materials, translating content, and generating lesson plans, meaningfully boosting productivity while the human remains the primary deliverer.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can generate health information content and scripts, delivering health education that is culturally appropriate, responds to live questions, and builds trust requires human presence and real-time adaptation. Current AI systems cannot reliably perform the interpersonal and cultural calibration aspects end-to-end, though they could assist with preparation.
Task automatability	claude-sonnet-5	2/5	While AI can generate culturally-tailored health content and even present via video, the live, adaptive, trust-building delivery to specific community groups requires human presence and real-time cultural attunement that current AI cannot fully replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Health information delivery is subject to regulatory oversight (e.g., FDA, state health boards), liability concerns regarding accuracy and harm, and strong cultural and social expectations that health education be delivered by trusted human figures. Legal and professional standards often require human accountability for health information.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement, but strong reliance on community trust, cultural competence, and in-person relationship-building creates significant organizational and social friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI tools for generating health content and subtitles are inexpensive, but when integrated with oversight, cultural review, and platform costs, the all-in cost approaches that of a junior health educator, especially given the quality and trust requirements of health information.
Cost vs. human wage	claude-sonnet-5	2/5	AI can cheaply generate content, but the in-person facilitation, trust-building, and cultural adaptation still require human labor, keeping overall costs comparable to or only modestly below human-only delivery.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed AI systems reliably perform culturally-sensitive health education delivery at scale in production. AI can assist with content generation and translation, but live teaching and audience engagement remain human-dependent. Chatbots and video content are emerging but do not meet the reliability standard for classroom or community health education delivery.
Technical feasibility today	claude-sonnet-5	2/5	Chatbots and AI-generated materials exist for health education, but no deployed product reliably conducts culturally-sensitive live classes or community outreach at scale in place of a human CHW.

Advise clients or community groups on issues related to social or intellectual development, such as education, childcare, or problem solving.

CI 25–30 · exposure 20 · augmentation 50 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health sectors are typically under-resourced, less digitized, and slower to adopt automation than information or finance sectors. While some use of AI-assisted tools is emerging, production-level deployment of AI for primary advisory roles remains limited.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work occurs in low-digitization, often underfunded public health and social service settings where AI adoption is slow and mostly limited to pilot programs.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist community health workers by drafting resource summaries, suggesting educational materials, or helping organize information for case review. However, the core advisory function—listening, empathizing, and personalizing guidance—remains primarily human-centered, limiting transformative augmentation.
Augmentation potential	claude-sonnet-5	3/5	AI can help CHWs prepare materials, translate information, or suggest resources on education/childcare topics, providing moderate assistance while the human remains central to trust-based delivery.
Task automatability	claude-haiku-4-5-20251001	2/5	This task requires nuanced understanding of individual circumstances, cultural context, and interpersonal judgment to deliver appropriate advice on sensitive social or developmental matters. While AI can retrieve generic information about education or childcare options, tailoring advice to a specific client's needs and building the trust necessary for effective counseling remain primarily human functions.
Task automatability	claude-sonnet-5	2/5	This task requires trust-building, contextual judgment, and personalized advising within community relationships, which current AI cannot replicate end-to-end despite being able to draft informational content.
Adoption barriers	claude-haiku-4-5-20251001	3/5	Community health work is less regulated than clinical medicine, so legal barriers to AI substitution are moderate. However, organizational trust, client preference for human contact, liability concerns around developmental advice, and the importance of relationship-building create meaningful adoption friction.
Adoption barriers	claude-sonnet-5	3/5	No licensing typically required for CHWs, but strong reliance on interpersonal trust, community embeddedness, and liability concerns around advice on childcare/education create moderate structural friction against automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI chatbots are cheap to run, but the setup, customization, validation, and human oversight required to deploy them safely in a community health context add significant cost. The human cost of poor advice (harm to vulnerable clients) makes total cost of ownership higher than a straightforward wage comparison.
Cost vs. human wage	claude-sonnet-5	2/5	While AI content generation is cheap, the actual value delivered—trusted, culturally-attuned advising in community contexts—still requires expensive human labor with no comparable AI substitute, so effective cost per equivalent-quality outcome favors humans.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed products reliably perform end-to-end advisory on social and intellectual development at the quality expected in real community health settings. Chatbots can provide information but lack the contextual understanding, empathy, and accountability required for genuine counseling.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product substitutes for a community health worker's in-person, relationship-based advising on childcare or education; chatbot tools exist only as narrow informational aids, not as reliable advising products.

Advise clients or community groups to ensure parental understanding of the importance of childhood immunizations and how to access immunization services.

CI 25–30 · exposure 25 · augmentation 63 · importance 3.3/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health worker roles remain rooted in relationship-based, in-person work in underserved areas. Digitization and AI adoption in public health outreach are slower than in white-collar professional services, with most programs still relying on traditional human deployment.
Sector adoption velocity	claude-sonnet-5	2/5	Public health and community health sectors are generally slow adopters of AI tools for direct client engagement, with most digitization focused on records rather than frontline advising.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist a human community health worker by drafting talking points, summarizing immunization schedules, identifying local service providers, and translating materials—raising efficiency. However, the human must deliver the trust-building and persuasion, so augmentation is meaningful but not transformative.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully support CHWs via multilingual translation, generating tailored educational materials, reminders, and FAQs, improving efficiency while the CHW remains the primary trusted communicator.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can generate informational content about immunizations, the task requires building trust, assessing individual barriers, and adapting messaging to specific client contexts—elements that demand human judgment and relationship-building. Current systems could draft educational materials but cannot reliably handle the interpersonal persuasion and cultural sensitivity needed end-to-end.
Task automatability	claude-sonnet-5	2/5	The core work involves building trust, in-person or relational communication, and responsiveness to community-specific concerns and cultural context, which current AI cannot fully replicate; chatbots can deliver static information but not the personalized advocacy CHWs provide.》
Adoption barriers	claude-haiku-4-5-20251001	4/5	Health communication, especially to vulnerable populations and regarding medical decisions, faces regulatory scrutiny and liability concerns. Communities often prefer and trust familiar human advisors; health authorities may require licensed or certified personnel to deliver immunization counseling. Organizational and legal friction is substantial.
Adoption barriers	claude-sonnet-5	3/5	No licensing requirement strictly mandates a human for this advising task, but strong community trust dynamics, cultural competency needs, and public health outreach norms create moderate organizational and trust-based friction against pure automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI information systems are cheap to operate, but the oversight and human correction needed to ensure safe, culturally competent health advice offsetts cost savings. A human community health worker's loaded wage is modest to moderate; the all-in cost of AI with adequate oversight remains comparable or higher.
Cost vs. human wage	claude-sonnet-5	2/5	While automated messaging/chatbot tools are cheap per interaction, they don't achieve equivalent outcomes to trusted human advising, so effective cost-per-quality-outcome remains comparable to or worse than human labor.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably performs this task autonomously. Chatbots exist for immunization information, but they lack the contextual understanding, trust-building capability, and ability to navigate parental concerns that community health workers provide. Real-world deployment in this role remains absent.
Technical feasibility today	claude-sonnet-5	2/5	Deployed chatbots and health information apps exist for immunization FAQs, but no production system reliably replaces the relational, community-embedded advising role of a CHW.

Develop plans or formal contracts for individuals, families, or community groups to improve overall health.

CI 25–30 · exposure 25 · augmentation 63 · importance 3.3/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health worker programs are often in under-resourced, smaller organizations with low digitization; while health systems explore AI documentation tools, production adoption of AI-generated formal health plans/contracts remains limited and cautious in this sector.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, relationship-driven field with limited AI adoption in production settings; the sector overall has been slow to deploy AI agents for client-facing planning.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist by generating initial plan templates, organizing family health data, suggesting evidence-based interventions, and streamlining contract language—genuinely useful aids that can improve a worker's efficiency—but the core relationship-building and personalized risk assessment remain human-driven.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully help draft plan templates, summarize health data, and suggest goals, boosting a health worker's efficiency while they retain responsibility for personalization and relationship building.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can draft template health plans and contracts, this task requires individualized assessment, understanding of family dynamics, cultural context, and negotiation—elements that demand human judgment and cannot be fully automated end-to-end with 50% time savings at equal quality without extensive human oversight.
Task automatability	claude-sonnet-5	2/5	AI can draft template health improvement plans but tailoring to individual/family circumstances, trust-building, and negotiating commitments requires human judgment and relational context that current AI cannot autonomously replicate end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Health contracts and care plans often require licensed health professional sign-off, may have legal implications under state health codes, involve vulnerable populations with informed-consent requirements, and carry liability risk if automated plans prove inadequate—creating meaningful regulatory and legal barriers.
Adoption barriers	claude-sonnet-5	3/5	No licensing mandate strictly requires a human to write these plans, but community trust, cultural competency needs, and organizational protocols create meaningful friction against full automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI-assisted drafting saves some labor, but the cost of oversight, legal review, integration with health systems, and human validation remains substantial; combined with community health workers' relatively modest wages, the all-in AI cost is not yet significantly lower.
Cost vs. human wage	claude-sonnet-5	2/5	AI drafting assistance is cheap, but the overall task still requires substantial human labor for engagement, negotiation, and follow-up, so total cost savings versus a human worker are modest.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably creates legally binding health contracts or formal care plans independently; existing AI tools can assist with templates or documentation but require substantial human validation, clinical review, and customization for real-world deployment.
Technical feasibility today	claude-sonnet-5	2/5	No deployed product independently creates and finalizes personalized health contracts with clients; existing tools only support drafting or documentation within a human-led workflow.

Assist families to apply for social services, including Medicaid or Women, Infants, and Children (WIC).

CI 23–30 · exposure 25 · augmentation 63 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Adoption is minimal. Community health work remains relationship-driven and low-digitization; social service agencies move slowly on automation; regulatory compliance and fraud concerns limit experimentation. Public data shows negligible deployment of AI in this role.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and social services sectors are historically slow AI adopters due to funding constraints, low digitization, and reliance on interpersonal trust with vulnerable populations.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can usefully assist by auto-populating forms, flagging eligibility gaps, summarizing requirements, and translating materials, enabling workers to focus on outreach and relationship-building. However, the human must verify all outputs and handle the sensitive final submission steps.
Augmentation potential	claude-sonnet-5	4/5	AI can meaningfully help CHWs by pre-filling forms, summarizing eligibility rules, translating documents, and flagging next steps, significantly speeding parts of the workflow while the CHW retains the relational role.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI could help draft application materials or identify eligibility criteria, the task requires navigating complex, jurisdiction-specific social service rules, verifying family circumstances, and handling sensitive personal information. Current systems cannot reliably handle the full workflow—eligibility determination, form completion, submission, and follow-up—without substantial human oversight, falling well short of 50% time savings at equal quality.
Task automatability	claude-sonnet-5	2/5	AI can help draft or explain forms and eligibility criteria, but navigating case-specific documentation, in-person verification, and agency-specific bureaucratic hurdles still requires substantial human effort and trust-building.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Strong barriers exist: social service agencies have strict identity-verification, fraud-prevention, and record-keeping requirements; many jurisdictions legally require a qualified human representative or the applicant themselves to certify truthfulness; liability exposure for incorrect applications is high; and many low-income families prefer or require human assistance due to language, literacy, or digital access barriers.
Adoption barriers	claude-sonnet-5	3/5	No strict licensing requirement for CHWs, but many families need trusted personal relationships, in-person assistance, and translation/cultural navigation that create real adoption friction.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI systems require significant setup, compliance oversight, and human verification to handle social service applications safely. The cost of integration, testing, regulatory review, and continuous human oversight likely exceeds or matches the loaded wage of a community health worker in most settings.
Cost vs. human wage	claude-sonnet-5	2/5	While AI can cheaply generate informational content, the labor-intensive parts (accompanying families, verifying documents, liaising with caseworkers) still require paid human staff, keeping costs comparable to human-only delivery.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably automates end-to-end social service applications. Some tools assist with form filling or eligibility screening, but they operate in narrow contexts and require human verification. Production systems capable of handling multi-state Medicaid and WIC rules with high accuracy do not exist at scale.
Technical feasibility today	claude-sonnet-5	2/5	Chatbots and benefit-navigation tools exist (e.g., benefits screening apps) but are narrow in scope and don't reliably handle the full application process, follow-ups, or exceptions across varied local agencies.

Advise clients or community groups on issues related to risk or prevention of conditions, such as lead poisoning, human immunodeficiency virus (HIV), prenatal substance abuse, or domestic violence.

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

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health work remains concentrated in resource-limited, primarily human-delivered settings with strong preference for in-person, culturally trusted advisors. Digital health adoption is slow in this sector, with pilots common but production deployment of AI counseling systems rare and typically limited to information provision rather than counseling.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and social services are a low-digitization sector with limited AI adoption in direct client-facing advisory roles, though some digital health education tools are piloted.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can usefully augment community health workers by preparing evidence-based fact sheets, drafting educational materials, and flagging risk assessment frameworks, but the core counseling—listening, rapport, crisis response, and personalized advice—remains squarely human-driven. Augmentation is meaningful but partial.
Augmentation potential	claude-sonnet-5	3/5	AI can help CHWs by providing up-to-date information, translation, and educational materials to support their conversations, though the core advising interaction remains human-led.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can provide factual information about risk factors and prevention strategies, the task requires tailored, contextual advice sensitive to individual circumstances, cultural backgrounds, and immediate safety concerns. Current AI systems lack the nuanced judgment and trust-building essential to effective health counseling, making end-to-end automation infeasible at parity with human counselors.
Task automatability	claude-sonnet-5	2/5	AI can provide general health information but cannot fully replace the trust-building, cultural competency, and situational judgment needed to advise vulnerable individuals on sensitive risks like domestic violence or HIV.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Significant barriers exist: community health workers often operate under clinical oversight, legal liability for incorrect health advice is substantial, trust and human contact are central to effectiveness with vulnerable populations, and many interventions (domestic violence safety planning, prenatal substance abuse counseling) involve mandatory reporting obligations that require human judgment and accountability.
Adoption barriers	claude-sonnet-5	4/5	Sensitive topics like domestic violence, HIV, and substance abuse carry high liability and require human judgment, cultural trust, and often mandated reporting responsibilities that legally and practically require human involvement.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI information systems are cheaper per query, but oversight, liability, and the need for human escalation in sensitive cases (suicide risk, abuse disclosure) make the all-in cost per meaningful counseling session comparable to or higher than human community health workers, especially in safety-critical contexts.
Cost vs. human wage	claude-sonnet-5	2/5	While AI content generation is cheap, the human oversight, community trust-building, and liability management needed for sensitive health advising keep effective cost comparable to or only modestly cheaper than a CHW.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed products reliably perform comprehensive health risk counseling at the quality required for vulnerable populations. Chatbots can deliver generic information, but real-world counseling demands responsiveness to trauma, crisis detection, and culturally competent guidance—areas where current systems show material error rates and narrow applicability.
Technical feasibility today	claude-sonnet-5	2/5	Chatbots and health information apps exist and are used for basic health education, but no deployed product reliably replaces the in-person, relationship-based advising CHWs provide to at-risk community members.

Monitor nutrition of children, elderly, or other high-risk groups.

CI 20–30 · exposure 20 · augmentation 50 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health worker sectors are typically lower-digitization, resource-constrained environments in public health and rural settings. Adoption of advanced AI monitoring systems remains limited; most progress is in data collection support rather than autonomous decision-making.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, high-touch field sector with historically slow AI adoption, mostly limited to pilot programs using basic digital data collection tools rather than AI-driven monitoring.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can meaningfully assist by automating dietary logging, flagging nutritional deficiencies from intake data, and generating visual reports on trends. However, the human worker remains essential for behavioral counseling, building trust with vulnerable populations, and contextual intervention decisions.
Augmentation potential	claude-sonnet-5	3/5	AI can meaningfully assist by analyzing collected data (e.g., growth charts, dietary logs), sending reminders, and flagging risk patterns, improving efficiency of the human worker without replacing direct observation.
Task automatability	claude-haiku-4-5-20251001	2/5	Nutritional monitoring requires interpreting dietary intake, health metrics, and contextual factors that demand nuanced judgment. AI can assist with data logging and pattern detection, but the integration of clinical observation, behavioral assessment, and individualized intervention planning remains largely dependent on human judgment and presence.
Task automatability	claude-sonnet-5	2/5	Monitoring nutrition requires in-person observation, physical measurements (weight, growth, anemia signs), and interpersonal trust-building that current AI cannot perform end-to-end; AI can support data logging and flagging but not replace the core surveillance activity.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Nutritional monitoring of vulnerable populations carries high liability risk, requires trust-based human relationships, and often involves clinical judgment that may be subject to healthcare regulations. The human-contact and care dimensions create substantial organizational and legal barriers to full automation.
Adoption barriers	claude-sonnet-5	3/5	No hard licensing requirement blocks AI involvement, but the vulnerable populations (children, elderly, high-risk) create liability and trust concerns, and health monitoring often ties to protocols requiring human accountability.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Current AI tools for nutrition monitoring (dietary apps, health tracking software) require significant human oversight, validation, and clinical interpretation, making the all-in cost competitive with or exceeding a community health worker's labor in real-world deployment.
Cost vs. human wage	claude-sonnet-5	2/5	AI tools (apps, chatbots) are cheap per interaction, but they cannot replace the physical home visits, measurements, and relationship-based assessment that comprise the actual task, so cost comparison favors humans for the core work.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No deployed product reliably performs end-to-end nutritional monitoring for vulnerable populations. While nutrition tracking apps and dietary analysis tools exist, they lack the contextual, relational, and clinical oversight necessary to substitute for a community health worker's integrated assessment and intervention.
Technical feasibility today	claude-sonnet-5	1/5	There are no deployed products that autonomously monitor at-risk individuals' nutrition status in the field; nutrition tracking apps exist but require human data entry and clinical judgment, not autonomous monitoring.

Provide feedback to health service providers regarding improving service accessibility or acceptability.

CI 20–30 · exposure 20 · augmentation 63 · importance 3.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health work remains rooted in relationship and trust; healthcare organizations are slowly adopting data analytics but not AI-driven feedback intermediation. Adoption is confined to pilots and measurement tools, not displacement of the feedback provider role itself.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, relationship-driven sector with slow AI adoption in frontline community engagement roles compared to office-based professional services.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can meaningfully assist by analyzing feedback patterns, identifying themes across multiple providers, and drafting summary documents, allowing community health workers to focus on deeper engagement and stakeholder communication. This augmentative use is already emerging in some health systems.
Augmentation potential	claude-sonnet-5	3/5	AI can help organize, summarize, and draft structured feedback reports from community input, improving efficiency of communication with providers even though the core listening/assessment work remains human.
Task automatability	claude-haiku-4-5-20251001	2/5	AI can help compile and summarize feedback data, but providing meaningful, contextually grounded feedback about health service accessibility requires understanding community dynamics, cultural nuance, and stakeholder relationships that current systems handle poorly. End-to-end automation with equal quality is not achievable today.
Task automatability	claude-sonnet-5	2/5	This requires synthesizing lived community experience, relational trust, and contextual judgment about cultural acceptability that AI cannot independently generate; AI could help draft or organize feedback but not perform the core task of gathering and interpreting community sentiment.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Health service improvement feedback often requires trusted human intermediaries and accountability; legal/regulatory frameworks around healthcare quality improvement may require human sign-off and authentic community voice, not algorithmic intermediation. Organizational trust in AI-mediated feedback is also low.
Adoption barriers	claude-sonnet-5	3/5	No licensing barrier specifically bars this, but organizational reliance on trusted community relationships and provider accountability structures create moderate friction against AI substitution.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI tools for data analysis and summarization are cheap, but integrating them into authentic community feedback workflows requires significant human oversight and validation, making the all-in cost comparable to or higher than direct human feedback provision.
Cost vs. human wage	claude-sonnet-5	2/5	While AI text tools are cheap, the actual value comes from community trust-building and field observation that AI cannot replicate, so cost comparison favors humans for the substantive part of the task.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	No production systems reliably perform this task independently. Sentiment analysis and data aggregation tools exist, but they cannot replicate the judgment-dependent, relationship-mediated work of a community health worker gathering and delivering constructive feedback to providers.
Technical feasibility today	claude-sonnet-5	1/5	No deployed products autonomously perform community liaison feedback synthesis to providers; this remains a human relational function with no production-scale substitute.

Identify the particular health care needs of individuals in a community or target area.

CI 20–25 · exposure 20 · augmentation 50 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health sectors remain relatively under-digitized and operate in resource-constrained settings with limited AI adoption. While health systems are digitizing, community-based primary assessment remains heavily reliant on human workers; pilot programs are uncommon and production deployment is rare in this space.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work occurs in under-resourced, low-digitization settings (public health, nonprofit, community organizations) where AI adoption for frontline outreach remains slow and pilot-stage at best.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist community health workers by synthesizing aggregate health data, identifying high-risk populations, or suggesting assessment frameworks based on prior cases. However, augmentation is limited because the core task—direct engagement and individualized need assessment—requires human presence and judgment.
Augmentation potential	claude-sonnet-5	3/5	AI can meaningfully assist by analyzing community health data, surveys, and demographic patterns to help identify priority needs, but the human must still conduct outreach and validate findings on the ground.
Task automatability	claude-haiku-4-5-20251001	2/5	AI can assist with data aggregation and pattern recognition in health records or survey responses, but identifying individual health care needs requires understanding complex social, cultural, and contextual factors specific to each person. Current systems cannot reliably assess needs end-to-end without substantial human oversight and field presence.
Task automatability	claude-sonnet-5	2/5	This requires in-person community engagement, trust-building, observation of social context, and interviewing individuals, which AI cannot autonomously conduct end-to-end today. AI can help analyze data once collected but cannot replace the human field identification work.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Community health workers often operate under licensing/certification requirements and serve vulnerable populations where liability and consent concerns are high. Legal and regulatory frameworks governing health assessment, combined with requirements for in-person cultural competency and trust-building, create strong adoption barriers to full automation.
Adoption barriers	claude-sonnet-5	4/5	Effective needs assessment depends on community trust, cultural competence, and often licensed/certified CHW roles with organizational and funder requirements for human contact, creating substantial barriers to automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	Data processing and analytics tools exist but require significant setup, data standardization, and human review. The cost of AI infrastructure, integration, and necessary human oversight remains comparable to or exceeds employing a community health worker, especially given the need for field validation.
Cost vs. human wage	claude-sonnet-5	2/5	The core need-identification work still requires paid human labor for community presence, trust, and interviews, so AI only reduces costs on the data-synthesis side rather than replacing the labor-intensive fieldwork.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	Existing health analytics and NLP tools can process aggregate health data and flag risk populations, but no deployed product reliably identifies individual care needs at the granularity and accuracy required by community health workers. Production systems lack the community knowledge and interpersonal assessment capability this task demands.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product independently identifies individual or community health needs through direct outreach; this remains a human relational and field-based task with AI at best playing an analytics support role.

Perform basic diagnostic procedures, such as blood pressure screening, breast cancer screening, or communicable disease screening.

CI 5–32 · exposure 13 · augmentation 63 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	3/5	Screening interpretation tools (AI-assisted image review) are seeing moderate adoption in healthcare settings, but the procedure performance itself remains manual. Adoption is pilots-and-integration stage, not deep displacement of the procedural work.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work is a low-digitization, high physical-presence field with minimal AI agent deployment for hands-on screening tasks.
Augmentation potential	claude-haiku-4-5-20251001	4/5	AI can substantially assist CHWs by providing real-time decision support during screening (e.g., flagging abnormal readings, suggesting follow-up actions, aiding interpretation of results), enabling them to work more efficiently and confidently while they retain the procedural and judgment role.
Augmentation potential	claude-sonnet-5	3/5	AI can assist with interpreting readings, flagging risk levels, documentation, and triage decisions once physical measurements are taken by the human worker.
Task automatability	claude-haiku-4-5-20251001	2/5	While AI can assist with some screening interpretation (e.g., image analysis for breast cancer), the physical performance of procedures like blood pressure measurement, palpation, and specimen collection cannot be automated by current systems. Partial automation of analysis post-collection is possible but does not meet the 50% time-saving threshold for the full task.
Task automatability	claude-sonnet-5	1/5	This task requires hands-on physical measurement (cuffs, palpation, sample collection) that current AI cannot physically perform; AI cannot replace the manual/physical execution of screening procedures.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Significant legal and liability barriers exist: many screening procedures must be performed by a licensed clinician or certified health worker under regulations (e.g., clinical laboratory standards, medical device oversight). Patient contact and informed consent are often required by law.
Adoption barriers	claude-sonnet-5	3/5	While not always requiring a licensed clinician, community health screenings often occur under health program protocols, in-person trust relationships, and physical contact requirements that resist automation, though barriers are lower than for licensed medical procedures.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	AI analysis tools can be cost-effective per scan or test, but the procedural component (blood pressure cuff, specimen collection, patient interaction) remains human-dependent. Total cost savings are modest when procedure execution is not automatable.
Cost vs. human wage	claude-sonnet-5	1/5	AI cannot substitute for the physical act of screening, so there is no meaningful AI cost basis to compare; human labor remains required for hands-on measurement.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	AI products exist for screening image interpretation (radiology, pathology), but diagnostic procedures themselves require physical interaction and human clinical judgment. No deployed system performs the hands-on procedural work reliably; analysis tools are narrow and require human oversight.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product autonomously performs physical health screenings in community settings; existing tools only interpret data after a human collects it (e.g., wearable BP monitors still need human application/interpretation).

Provide basic health services, such as first aid.

CI 0–16 · exposure 5 · augmentation 38 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	2/5	Community health programs are typically in lower-digitization settings with limited tech infrastructure; adoption of AI tools remains spotty and mostly experimental rather than at production scale.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work involving physical care is a low-digitization, high-touch field with minimal AI/robotic adoption for direct physical intervention tasks.
Augmentation potential	claude-haiku-4-5-20251001	3/5	AI can assist with triage protocols, symptom checking, and decision support (e.g., when to escalate), helping community health workers make better initial judgments, though human assessment and care remain central.
Augmentation potential	claude-sonnet-5	2/5	AI can provide reference guidance, decision-support checklists, or training materials for first aid, but offers limited real-time assistance during the physical act itself.
Task automatability	claude-haiku-4-5-20251001	1/5	First aid requires real-time physical assessment, manual intervention (bandaging, CPR, splinting), and immediate clinical judgment in unpredictable situations. Current AI cannot perform these hands-on interventions or reliably assess patients in person.
Task automatability	claude-sonnet-5	1/5	Providing first aid requires physical presence, manual dexterity, and real-time hands-on intervention that current AI systems cannot perform, being purely digital/software-based.
Adoption barriers	claude-haiku-4-5-20251001	4/5	Liability concerns are substantial: improper first aid can cause serious harm, and legal/regulatory frameworks typically require a qualified human to assess and treat patients. Organizational and community trust in human care providers also creates friction.
Adoption barriers	claude-sonnet-5	5/5	Direct physical care and first aid often require certification, hands-on human presence, and carry significant liability if performed incorrectly, creating hard barriers to automation.
Cost vs. human wage	claude-haiku-4-5-20251001	2/5	While AI decision support has low marginal cost, community health workers are already low-wage, and the capital, integration, and human oversight required to deploy AI in resource-constrained settings often exceeds the savings.
Cost vs. human wage	claude-sonnet-5	1/5	AI cannot substitute for the physical labor involved, so there is no viable cost comparison—human presence is mandatory, making AI infeasible rather than cheaper or costlier in a meaningful sense.
Technical feasibility today	claude-haiku-4-5-20251001	2/5	AI can provide first aid guidance via chatbots or decision trees, and some screening tools exist, but deployed systems remain narrow in scope and cannot replace human assessment and physical care delivery in real emergency contexts.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product administers physical first aid; this remains entirely outside the scope of software or robotic products in production today.

Conduct home visits for pregnant women, newborn infants, or other high-risk individuals to monitor their progress or assess their needs.

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

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	This task is performed by human community health workers in low-digitization, relationship-dependent sectors (public health, community care). Current adoption trends show no movement toward AI-driven home visit automation; the sector remains resistant to displacement of direct human contact.
Sector adoption velocity	claude-sonnet-5	2/5	Community health and home-visiting programs are typically under-resourced, low-digitization environments with limited AI deployment beyond scheduling or documentation tools.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could provide minor assistance through pre-visit data summaries, appointment scheduling, or post-visit documentation support, but these are peripheral to the core task of conducting the visit itself. The human worker remains the irreplaceable locus of care delivery.
Augmentation potential	claude-sonnet-5	3/5	AI can assist with visit scheduling, risk triage, documentation, translation, and follow-up reminders, but the core in-home assessment remains human-driven.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires in-person physical presence, direct observation of health status in a home environment, and nuanced interpersonal assessment of vulnerable individuals. AI systems cannot conduct home visits or perform the physical and relational work necessary to monitor pregnant women, newborns, or assess complex social and health needs in real time.
Task automatability	claude-sonnet-5	1/5	This task requires physical presence, hands-on observation, and in-person trust-building in a home setting, none of which current AI systems can perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Home visits to vulnerable populations (pregnant women, newborns, high-risk individuals) are legally and ethically required to be conducted by trained human health workers. Regulatory requirements, duty of care, informed consent, and the need for licensed oversight create hard barriers to any AI substitution.
Adoption barriers	claude-sonnet-5	4/5	Health monitoring of vulnerable populations involves liability, privacy, and often licensing/certification requirements for community health workers, plus an inherent human-contact requirement for physical home visits.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI cannot perform this task at all, making cost comparison moot. The task is inherently human-centric and location-dependent, with no plausible automation pathway that would create a cost advantage.
Cost vs. human wage	claude-sonnet-5	1/5	AI cannot substitute for the physical visit at all, so there is no viable AI cost basis to compare against the human wage for this specific task.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI product can substitute for in-person home visits. While remote monitoring tools and telehealth exist, they cannot replace the direct assessment, physical examination capability, and trusted human presence that this task fundamentally requires.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product conducts physical home visits or performs the embodied assessment and rapport-building this task requires.

Advocate for individual or community health needs with government agencies or health service providers.

CI 0–7 · exposure 0 · augmentation 38 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Community health advocacy is embedded in trust-based, relationship-intensive work in underresourced sectors with low digital maturity. There is no evidence of AI adoption for this core function in real deployments.
Sector adoption velocity	claude-sonnet-5	2/5	Community health work is a low-digitization, relationship-driven, often under-resourced sector with limited AI adoption for interpersonal advocacy tasks.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist by drafting advocacy materials, aggregating data on health disparities, or summarizing policy documents, but the core act of advocating—persuading, negotiating, representing—remains fundamentally human and cannot be substantially delegated to AI.
Augmentation potential	claude-sonnet-5	3/5	AI can help draft letters, summarize case histories, research policies, or prepare talking points, meaningfully aiding preparation even though the advocacy interaction itself remains human-led.
Task automatability	claude-haiku-4-5-20251001	1/5	Advocacy requires sustained relationship-building, negotiation, understanding of individual circumstances and agency politics, and strategic judgment about when and how to push for change. Current AI cannot authentically represent constituents or negotiate with government agencies on their behalf.
Task automatability	claude-sonnet-5	1/5	Effective advocacy requires building trust, navigating politics, negotiating in person, and representing lived community experience—capabilities current AI cannot autonomously execute end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Advocacy often requires legal standing, professional credentials, and formal representation of constituents. Many health service interactions and government negotiations legally or practically require a human representative accountable to the community and bound by professional ethics.
Adoption barriers	claude-sonnet-5	4/5	Advocacy often requires trusted human relationships, legal standing, and accountability to communities and agencies, creating strong organizational and trust-based barriers to automation, though not a strict licensing requirement.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The task's value lies in human relationships, legitimacy, and accountability. An AI system attempting this would either fail to accomplish the goal or require substantial human oversight, making it more expensive than direct human advocacy.
Cost vs. human wage	claude-sonnet-5	1/5	There is no viable AI substitute performing this function, so cost comparison favors the human worker entirely; any AI use is just a minor supplement to labor cost, not a replacement.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system operates as an advocate in production settings. This task fundamentally requires human credibility, legal standing, and the trust of both community members and institutional decision-makers, which AI cannot establish.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product independently advocates with government agencies or health providers on behalf of individuals or communities; this remains a human relational and political function.

Transport or accompany clients to scheduled health appointments or referral sites.

CI 0–5 · exposure 0 · augmentation 25 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Community health work is labor-intensive, localized, and concentrated in underserved areas with low tech adoption. Autonomous transport remains nascent and not deployed in this occupational context.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work is a low-digitization, physically-embedded human service sector with minimal AI/agent deployment for this specific task.
Augmentation potential	claude-haiku-4-5-20251001	2/5	Navigation and appointment scheduling tools provide minor assistance, but AI offers limited augmentation for the core task of physically accompanying and supporting clients during transport.
Augmentation potential	claude-sonnet-5	2/5	AI can assist with scheduling, route optimization, or reminders around appointments, but offers minimal enhancement to the core physical accompaniment task itself.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires physical presence, navigation in real-world settings, and human-to-human care (accompanying vulnerable clients). Current AI systems cannot physically transport people or provide the interpersonal support essential to the task.
Task automatability	claude-sonnet-5	1/5	This is a physical transportation and accompaniment task requiring bodily presence in a vehicle and interpersonal support; no AI system can perform physical transport.」
Adoption barriers	claude-haiku-4-5-20251001	5/5	Strong barriers exist: direct human contact is legally and practically required (duty of care, liability, vulnerable populations), and many jurisdictions have regulations governing transport of clients in health contexts.
Adoption barriers	claude-sonnet-5	4/5	Liability, safety, and trust concerns around transporting vulnerable clients (e.g., medical needs, mobility issues) create strong barriers to any automated substitution beyond human drivers.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Any theoretical automation would require autonomous vehicles with significant infrastructure and liability costs, far exceeding the wage of a community health worker providing this service today.
Cost vs. human wage	claude-sonnet-5	1/5	AI has no capability to substitute for physical transport, so there is no viable AI cost comparison; human labor remains the only option.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI product can physically accompany and transport clients to health appointments. While ride-sharing and navigation exist separately, they do not address the core care and accompaniment component.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product physically transports or accompanies people; this remains entirely human-performed, sometimes aided by rideshare apps but not AI automation of the task itself.

Attend community meetings or health fairs to understand community issues or build relationships with community members.

CI 0–5 · exposure 0 · augmentation 25 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Community health organizations operate in under-digitized, relationship-intensive sectors with minimal AI adoption. The core value of CHW roles is human presence and trust; adoption pressure is low because automation does not map onto the actual work.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work is a low-digitization, relationship-driven field with minimal AI agent deployment for in-person engagement tasks.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could marginally assist with pre-meeting research or post-meeting documentation and trend analysis, but the core task of attending and relationship-building leaves little room for AI augmentation. The human must be fully present and engaged.
Augmentation potential	claude-sonnet-5	2/5	AI could help summarize meeting notes, track community issues, or prepare talking points beforehand, but offers little assistance during the actual in-person engagement.
Task automatability	claude-haiku-4-5-20251001	1/5	Attending meetings and building relationships require physical presence, interpersonal dynamics, and contextual social understanding that current AI systems cannot perform autonomously. This task is fundamentally about human-to-human engagement and trust-building, which AI cannot replicate end-to-end today.
Task automatability	claude-sonnet-5	1/5	This task requires physical presence, in-person relationship-building, and real-time social engagement that current AI systems cannot perform end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Community health work is inherently relational and legally/ethically requires human community health workers to establish trust and provide culturally competent engagement. Regulatory frameworks and organizational mandates require licensed or credentialed humans to perform this outreach work.
Adoption barriers	claude-sonnet-5	4/5	Community trust, cultural competency, and in-person rapport are core to the CHW role and are effectively unautomatable without a human presence, creating a strong structural barrier.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	AI has no cost advantage here because the task requires human attendance and presence. Any AI-assisted analysis would be supplementary, not a replacement, so the loaded human wage remains the dominant cost.
Cost vs. human wage	claude-sonnet-5	1/5	AI has no viable way to substitute for physical attendance and relationship-building, so cost comparison favors the human by default.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI system can independently attend community meetings or build genuine relationships with community members. While AI can assist with information gathering or post-meeting analysis, it cannot substitute for the human presence and relational work that defines this task.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product attends meetings or builds trust-based community relationships in person; this is fundamentally a human presence task.

Administer immunizations or other basic preventive treatments.

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

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	This task involves direct patient contact and clinical responsibility; adoption of automation is negligible because legal and safety requirements mandate human presence and decision-making at the point of care.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work involves in-person, physical healthcare delivery in a sector with low digitization for hands-on clinical tasks, showing minimal AI displacement.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist with ancillary tasks like vaccination record lookup, patient identification, or post-visit documentation, but offers minimal real-time support during the core act of administering treatment itself.
Augmentation potential	claude-sonnet-5	2/5	AI can support scheduling, patient education, or tracking immunization records, but offers minimal assistance to the actual physical act of administering a treatment.
Task automatability	claude-haiku-4-5-20251001	1/5	Administering immunizations and preventive treatments requires physical contact with patients, sterile technique, observation for adverse reactions, and personalized clinical judgment. Current AI systems cannot perform physical medical procedures or reliably identify subtle signs of medical distress.
Task automatability	claude-sonnet-5	1/5	Administering immunizations requires physical, hands-on manipulation of needles and patient bodies, which current AI systems cannot perform since they lack physical embodiment for this task.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Medical licensure, malpractice liability, legal scope-of-practice requirements, and often explicit state regulations mandate that a qualified human (nurse, physician, or authorized health worker) must physically administer immunizations and assess patient response.
Adoption barriers	claude-sonnet-5	5/5	Administering immunizations typically requires certified/licensed personnel, involves bodily contact, and carries significant liability and safety regulations that legally require human execution.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	The fully-loaded cost of a community health worker performing this task is already low; AI cannot substitute for the essential human labor component of physical medical administration and cannot achieve cost parity.
Cost vs. human wage	claude-sonnet-5	1/5	AI has no cost-effective way to perform this physical task at all, so it cannot be cheaper than a human worker who can actually administer treatment.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed AI product performs vaccine administration or injection-based preventive treatments. This is a hands-on clinical procedure that remains entirely dependent on trained human practitioners.
Technical feasibility today	claude-sonnet-5	1/5	No deployed AI product administers physical injections or preventive treatments; this remains purely a human physical task.

Report incidences of child or elder abuse, neglect, or threats of harm to authorities, as required.

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

Panel ratings per dimension for this task
Dimension	Model	Rating	Rationale
Sector adoption velocity	claude-haiku-4-5-20251001	1/5	Adoption of AI to replace this task is negligible because it is legally prohibited. Community health workers operate in regulated, human-facing roles with deep accountability requirements.
Sector adoption velocity	claude-sonnet-5	1/5	Community health work is a low-digitization, in-person, community-embedded field with minimal AI agent deployment for safeguarding decisions.
Augmentation potential	claude-haiku-4-5-20251001	2/5	AI could assist by flagging potential patterns or generating documentation templates, but the core judgment, decision-making, and legal responsibility remain entirely with the human reporter. Augmentation potential is limited because the task is already straightforward and high-stakes.
Augmentation potential	claude-sonnet-5	2/5	AI can help with documentation, checklists, or reminders about reporting protocols, but offers little assistance for the core judgment of detecting and deciding to report abuse.
Task automatability	claude-haiku-4-5-20251001	1/5	This task requires human judgment to interpret complex social situations, assess credibility of reports, and make consequential decisions about legal obligations that trigger mandatory reporting laws. AI cannot reliably perform the investigative and observational work required to substantiate abuse or neglect.
Task automatability	claude-sonnet-5	1/5	This requires in-person judgment, trust-building, direct observation of vulnerable individuals, and legally mandated reporting responsibility that cannot be delegated to software end-to-end.
Adoption barriers	claude-haiku-4-5-20251001	5/5	Mandatory reporting laws in all U.S. jurisdictions legally require a human professional (the community health worker) to make and file the report personally. The human is legally accountable for the decision and filing, creating an absolute barrier to substitution.
Adoption barriers	claude-sonnet-5	5/5	Mandatory reporting laws require a specific accountable human to identify and report abuse/neglect, with legal liability for failure to do so, making this a hard-barrier task.
Cost vs. human wage	claude-haiku-4-5-20251001	1/5	Automation is not feasible, so cost comparison is moot. However, if attempted, the liability exposure and error costs far exceed the labor savings of the task itself.
Cost vs. human wage	claude-sonnet-5	1/5	AI cannot substitute for the human observation, relationship, and legal accountability involved, so there is no viable AI cost basis to compare against the human wage.
Technical feasibility today	claude-haiku-4-5-20251001	1/5	No deployed product performs this task autonomously in production. Mandatory reporting is a legal and ethical requirement that demands human professional judgment and legal accountability; it cannot be delegated to AI systems.
Technical feasibility today	claude-sonnet-5	1/5	No deployed product independently identifies abuse/neglect in the field and files mandated reports; this remains a human judgment and legal-responsibility task.

Related occupations — Community & Social 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.