Adult Basic Education, Adult Secondary Education, and English as a Second Language Instructors
25-3011.00Median wage $61,540/yr37,310 employed (US)Rank #460 of 923 scored · top 50% by substitution
Teach or instruct out-of-school youths and adults in basic education, literacy, or English as a Second Language classes, or in classes for earning a high school equivalency credential.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution28
Exposure24
Augmentation63
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
37 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
0%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%23
panel mean rating 1.9/5 → substitution pressure 23/100
Technical feasibility todayw 20%27
panel mean rating 2.1/5 → substitution pressure 27/100
Cost vs. human wagew 15%31
panel mean rating 2.3/5 → substitution pressure 31/100
Adoption barriersw 20%inverted — strong barriers lower the score39
panel mean rating 3.4/5 (barrier strength) → substitution pressure 39/100
Sector adoption velocityw 10%22
panel mean rating 1.9/5 → substitution pressure 22/100
Task breakdown (37 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.
Assign and grade class work and homework.
59CI 51–67 · exposure 58 · augmentation 88 · importance 4.2/5 · click for rater detail
Assign and grade class work and homework.
59| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | K-12 and adult education sectors show moderate adoption of automated grading tools, with LMS integration common in higher ed and some adult programs; however, many institutions remain reliant on manual grading due to pedagogical preference and comfort levels. Adoption is faster in formal higher education and corporate training, but slower in underfunded public adult education. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors are under-resourced and slower to adopt AI tools compared to corporate or well-funded K-12/higher-ed systems, with pilots more common than full deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assists instructors by handling objective-assessment scoring and providing draft rubric feedback on essays, freeing instructor time for deeper feedback and intervention. Teachers report meaningful productivity gains when using AI-assisted grading for routine assignments, though they remain responsible for quality assurance and personalized commenting. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI substantially assists instructors by auto-generating practice exercises, providing instant feedback on grammar/writing, and pre-scoring homework, freeing time for personalized instruction. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Current AI systems can fully automate multiple-choice and short-answer grading, and can handle routine homework assignment, reducing time by 50%+ on objective tasks. However, grading essay-based work, open-ended writing, and assignments requiring nuanced pedagogical judgment remain partially manual, requiring human review of AI scoring for consistency and fairness. |
| Task automatability | claude-sonnet-5 | 4/5 | AI can grade objective and even many written assignments (e.g., ESL grammar exercises, short essays) with rubric-based scoring, and can generate assignments automatically, saving substantial teacher time though human spot-checking is still common. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Educational institutions have discretion over grading automation, but significant institutional friction exists: parent and student preferences for human feedback, teacher union concerns, accreditation requirements for human judgment in some contexts, and liability risk if AI-assigned grades are challenged. These are not hard legal barriers but do slow adoption. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates human grading for adult basic/ESL education, though instructors typically retain final grade authority and may need to verify AI output for fairness and accuracy. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Inference cost for LMS-integrated automated grading is near-zero per assignment, and marginal oversight cost is modest compared to an instructor's loaded labor for manual grading of 20–30 student submissions per assignment. AI grading is substantially cheaper, though integration and quality-assurance overhead reduce the advantage slightly below an order of magnitude. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated grading and assignment tools cost a fraction of instructor grading time per assignment, especially at scale across many students and repeated exercises. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Learning management systems with automated grading for objective assessments are deployed at scale in educational institutions; tools like Turnitin and Canvas grade multiple-choice/short-answer reliably. However, AI essay grading and subjective assignment evaluation still show material error rates and require instructor oversight, limiting production reliability for the full task scope. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like Grammarly, AI-based LMS grading tools, and essay-scoring systems (e.g., Turnitin's AI features) are deployed in some adult education settings, but reliability varies especially for open-ended or ESL-specific language nuances. |
Maintain accurate and complete student records as required by laws or administrative policies.
54CI 40–67 · exposure 55 · augmentation 75 · importance 4.1/5 · click for rater detail
Maintain accurate and complete student records as required by laws or administrative policies.
54| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Educational institutions have adopted student information systems widely, but record maintenance remains largely manual or human-overseen rather than AI-driven; adoption of autonomous AI record management in schools is still limited. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Educational institutions, especially in adult and continuing education, adopt digital administrative tools at a moderate pace, often slower than corporate sectors due to funding constraints and legacy systems. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems substantially assist instructors by automating data entry, flagging missing fields, formatting records, and generating compliance reports, meaningfully reducing the time instructors spend on manual record maintenance while maintaining human oversight. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered administrative tools can significantly reduce time spent on data entry, flagging errors, and generating required reports, meaningfully augmenting instructor productivity on this task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with data entry and formatting student records, maintaining legal and policy-compliant completeness requires understanding context-specific administrative requirements and manual verification of student information accuracy. Current systems cannot reliably ensure compliance with diverse institutional policies without human review. |
| Task automatability | claude-sonnet-5 | 4/5 | Record-keeping is largely structured data entry and compliance documentation, which current systems (student information systems, forms with AI-assisted data entry) can largely automate, though some human verification remains needed for accuracy and edge cases. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions face regulatory requirements (FERPA, state education laws) and internal audit obligations that legally necessitate institutional control and human accountability over student records; these create substantial legal and compliance barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Legal and administrative compliance requirements mean some human oversight/certification of records is typically required, though the underlying data entry and maintenance work itself faces no strict licensing barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Learning management and student information systems are relatively low-cost and available at scale, making the AI-assisted approach significantly cheaper than paying instructors to manually maintain all records from scratch. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated record-keeping software is inexpensive relative to instructor time spent on paperwork, though licensing, integration with legacy systems, and compliance oversight add some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Products for student information systems exist and handle record-keeping at scale, but they typically require human data entry, verification, and policy-specific configuration rather than autonomous end-to-end automation of the maintenance task itself. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Student information systems and administrative software already handle record maintenance reliably in many educational institutions, though full end-to-end automation without human review is less common in adult education settings specifically. |
Prepare reports on students and activities as required by administration.
52CI 39–65 · exposure 50 · augmentation 75 · importance 3.6/5 · click for rater detail
Prepare reports on students and activities as required by administration.
52| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions are slower to adopt AI-driven administrative automation than other sectors; most adoption remains in LMS features (gradebook, roster export) rather than production agent-based report generation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction is a modestly digitized, often underfunded sector with slower AI tool adoption compared to corporate or tech-forward fields. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist by auto-populating data fields, suggesting narrative prompts, and formatting compliance documentation, allowing instructors to focus on interpreting results and writing reflective commentary rather than data entry and basic synthesis. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can substantially speed up drafting of narrative reports, summarizing student data and activities, while the instructor reviews and finalizes for accuracy and compliance. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can draft routine report sections (attendance, grades, basic demographics) with significant setup and templates, but cannot independently assess student progress or pedagogical outcomes with sufficient reliability to meet the 50% time-saving threshold without substantial human review and correction. |
| Task automatability | claude-sonnet-5 | 4/5 | Report writing from structured data (attendance, grades, progress notes) is a well-suited generative task; AI can draft most of the text quickly given inputs, though instructors must supply and verify data. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Institutions have established reporting procedures and regulatory compliance requirements (FERPA, accreditation) that create procedural friction, and instructors retain responsibility for the accuracy and integrity of student records, limiting full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks AI-assisted report drafting, but instructors remain accountable for accuracy and administrative sign-off, creating mild organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI tools (LMS reporting, document automation) may reduce report-writing time modestly, but the cost of integration, oversight, and error-correction is roughly comparable to an instructor's time spent on this administrative task. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating draft reports via an LLM is very cheap compared to instructor time spent compiling and writing narrative reports, though some oversight time remains. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Tools exist to auto-generate attendance summaries and basic record exports, and some LMS systems offer reporting dashboards, but they require extensive human curation to meet institutional and administrative standards, and nuanced student progress narratives still require instructor judgment. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | AI writing assistants and some LMS reporting tools are used to draft summaries and progress reports today, but full end-to-end automated administrative reporting integrated into school systems is not universal or fully reliable. |
Provide information, guidance, and preparation for the General Equivalency Diploma (GED) examination.
49CI 39–59 · exposure 42 · augmentation 75 · importance 3.7/5 · click for rater detail
Provide information, guidance, and preparation for the General Equivalency Diploma (GED) examination.
49| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | GED prep is increasingly digitized (many community colleges and programs now offer hybrid or online options), and AI tools are being piloted in adult education, but adoption remains spotty. The sector is moderately digitized, pilots are common, but true end-to-end AI replacement of instruction is rare; most use AI as a supplementary resource within instructor-led or structured program contexts. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and workforce development sectors are historically slow adopters of new technology, with limited resources and infrequent large-scale AI deployment compared to corporate sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI significantly augments GED instructors by generating unlimited practice problems, providing instant explanations, offering 24/7 accessibility for students, and freeing instructor time for high-touch diagnosis and motivation. This enables instructors to focus on identifying struggling students and providing personalized support rather than content delivery, substantially raising overall program productivity. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can meaningfully help instructors by generating practice tests, explaining difficult subject matter, and tracking student progress, enhancing instructor efficiency while human guidance remains central. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can generate practice problems, model solutions, and study guides for GED prep at scale, but cannot replicate the adaptive feedback, motivational guidance, and assessment of individual learning gaps that effective GED instruction requires. The task involves diagnosis of student-specific weaknesses and targeted remediation that demands human judgment and iterative interaction. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate GED study plans, explain content, and provide practice questions, but personalized guidance, motivation, and pacing for adult learners still require significant human involvement, especially for struggling students.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | GED preparation is not formally licensed (instructors are not required to be credentialed in the way physicians are), but institutional and regulatory friction exists: schools and programs must maintain accountability for student success, students often need structured enrollment and financial aid access, and employers/institutions may require evidence of participation in recognized programs rather than self-study or AI-only preparation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically for GED tutoring, though adult education programs often require certified instructors for funding/accreditation purposes, creating some institutional friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-powered GED prep (via chatbots or online platforms) costs a small fraction of instructor-led courses or tutoring. A subscription-based AI system costs roughly $10–50/month versus $500–3000+ for instructor-led prep, making AI approximately 10–30× cheaper on a per-course basis when scaled. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI tutoring tools are cheap per use, but integrating them with human oversight, assessment, and learner support still requires paid staff time, keeping costs roughly comparable for full task delivery. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed AI tutoring products (e.g., ChatGPT, specialized edtech platforms) can provide GED content, practice questions, and explanations reliably, but they lack the enrollment verification, official pacing, accountability structures, and institutional integration that formal GED preparation programs require. Most such AI is used as a supplement rather than as the primary preparation pathway. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Adaptive learning platforms and AI tutors (e.g., Khan Academy, GED-specific prep apps) exist and are used in some programs, but reliability across diverse adult learners and full guidance counseling is uneven. |
Establish clear objectives for all lessons, units, and projects and communicate those objectives to students.
47CI 37–56 · exposure 42 · augmentation 75 · importance 4.3/5 · click for rater detail
Establish clear objectives for all lessons, units, and projects and communicate those objectives to students.
47| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions adopt learning management and AI writing tools slowly, and most K-12 and adult education settings have not yet integrated AI for lesson planning at scale. Adoption remains pilot-stage or limited to early-adopter institutions. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult/ESL education is a modestly digitized, underfunded sector with slower AI tool adoption compared to corporate or tech-forward industries, though some pilots exist in the broader ed-tech space. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at generating objective drafts, rephrasing for clarity, and rapidly iterating multiple versions, allowing instructors to focus on judgment and alignment with student needs rather than drafting from scratch. This meaningfully raises instructor productivity while keeping them in control. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI is widely useful for brainstorming, drafting, and refining lesson objectives, saving instructors substantial planning time while they retain control over final content and delivery. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can draft lesson objectives and communicate them in written form, but establishing *clear* objectives requires understanding student cohort needs, skill levels, and learning outcomes—human pedagogical judgment is essential. Current systems can accelerate drafting but cannot replace the instructional design judgment needed. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can draft lesson objectives aligned with curricula and standards quickly, but tailoring them to specific student populations, contextualizing communication, and adapting in-class delivery still requires human judgment.dramatically reduces drafting time though not the full task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Instructors are typically required by educational institutions, accreditors, and professional standards to author and own learning objectives; there is no licensing requirement preventing AI use, but institutional policy, pedagogical autonomy, and accountability for student outcomes create strong organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human write objectives, but instructors are professionally expected to own instructional design and there's institutional preference for teacher-authored plans tied to accreditation standards. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-assisted objective-drafting costs pennies per lesson, while an instructor's time spent writing objectives from scratch costs tens of dollars. Even with overhead, AI assistance is substantially cheaper than unassisted human labor. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating draft objectives via AI costs a fraction of instructor planning time, though the instructor must still review and communicate them, so it's not a full order-of-magnitude saving on the whole task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | AI writing tools (Claude, ChatGPT) can generate lesson objectives and emails communicating them, and some LMS platforms integrate AI assistance; however, no mature product reliably handles the full context-dependent task of establishing objectives *for* a specific class without instructor review and refinement. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like lesson-planning assistants and LMS tools (e.g., ChatGPT-based planners, Curipod) are used by teachers today to generate objectives, but reliability varies and human review is standard practice. |
Use computers, audio-visual aids, and other equipment and materials to supplement presentations.
47CI 30–64 · exposure 47 · augmentation 88 · importance 3.6/5 · click for rater detail
Use computers, audio-visual aids, and other equipment and materials to supplement presentations.
47| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions, particularly adult ESL and basic education programs, tend to be conservative adopters of automation. Budget constraints, legacy infrastructure, and institutional resistance to replacing instructor discretion in pedagogical decisions slow adoption relative to corporate or information-intensive sectors. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Education sector adoption of AI content-creation tools is growing steadily but is uneven across adult education programs, which vary widely in tech resources and digitization. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist instructors by generating slide drafts, recommending multimedia resources, organizing materials by learning objective, and suggesting presentation structures. These tools can materially reduce preparation time and expand the range of supplementary materials available while instructors retain full control over pedagogical decisions and classroom delivery. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI dramatically speeds up creation of visual aids, quizzes, and multimedia content, letting instructors focus more on delivery and student interaction. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can help generate or organize digital content and suggest multimedia selections, the task of integrating and deploying these tools in real time to supplement live instruction requires human judgment about pacing, audience engagement, and pedagogical fit. Current systems cannot manage the interactive, responsive aspects of presentation supplementation that effective teaching demands. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate slides, audio-visual content, and select supplementary materials quickly, but the instructor still must integrate and operate these tools live in the classroom setting. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational settings face regulatory requirements around curriculum alignment, accessibility standards (WCAG, ADA), and institutional approval of materials. Additionally, instructors often must personally validate pedagogical appropriateness and student safety, creating both organizational and legal barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks use of AI-assisted materials; some institutional policies or accessibility standards may add minor friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The cost of integrating AI systems, maintaining content libraries, ensuring compliance with educational standards, and oversight by instructors is comparable to or potentially exceeds the cost of instructors spending time selecting and organizing materials themselves, given the relatively modest labor cost of this preparatory work. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | AI tools for generating slides, images, and supplementary media cost a fraction of the instructor time it would take to create these from scratch. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Tools exist to generate presentation slides, suggest relevant audio-visual content, and organize materials, but deployed educational products that autonomously manage supplementary media in live classroom settings with consistent reliability remain limited. Most implementations require significant human oversight and manual integration. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Products like presentation generators, AI-assisted content creation tools, and multimedia authoring platforms are widely deployed and reliably used by educators today. |
Select and schedule class times to ensure maximum attendance.
41CI 30–52 · exposure 38 · augmentation 63 · importance 3.6/5 · click for rater detail
Select and schedule class times to ensure maximum attendance.
41| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adult education institutions, often under-resourced with legacy systems, adopt scheduling automation slowly; most continue manual or semi-manual processes, with adoption concentrated in larger, better-funded districts rather than widespread production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education programs, often community-based or under-resourced, tend to have slower technology adoption compared to corporate or tech-forward sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by analyzing enrollment data, identifying attendance patterns, and generating candidate schedules for review, meaningfully reducing the time administrators spend on manual scheduling while they retain decision-making authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted scheduling tools can meaningfully help instructors and administrators identify optimal time slots based on attendance patterns and student data, improving decision quality without replacing human judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Scheduling class times involves constraint satisfaction and data analysis that AI can partially support, but selecting times for maximum attendance requires understanding institutional policies, instructor availability, and enrollment patterns that typically necessitate human judgment and final approval. |
| Task automatability | claude-sonnet-5 | 3/5 | Scheduling optimization based on student availability data is a well-structured logic problem that scheduling software can largely handle, though it requires gathering and interpreting human input about preferences and constraints.dare |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Educational institutions have policies, union agreements, and administrative review requirements around scheduling; while not legally mandated automation barriers, institutional friction and the need for human sign-off on final schedules provide moderate protection. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates a human perform scheduling, though institutional policies and stakeholder coordination create moderate organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current scheduling software requires integration, customization, and ongoing human review to ensure constraints are met, making the all-in cost comparable to or potentially higher than having an administrator manually schedule classes. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Scheduling software licenses are relatively cheap compared to administrative staff time, but integration with institutional systems and instructor input still requires human coordination, keeping costs comparable rather than drastically lower. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While scheduling tools exist, few deployed products reliably optimize class times specifically for adult education attendance without human oversight; most systems require significant manual configuration and validation of their recommendations. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Scheduling and calendar-optimization tools exist and are used in educational administration, but they typically require human oversight to interpret institutional constraints, room availability, and instructor preferences. |
Prepare and administer written, oral, and performance tests and issue grades in accordance with performance.
41CI 34–48 · exposure 42 · augmentation 75 · importance 3.8/5 · click for rater detail
Prepare and administer written, oral, and performance tests and issue grades in accordance with performance.
41| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption remains limited in K–12 and adult education sectors compared to higher-education learning analytics. Most instructors use basic LMS tools for objective grading but retain manual processes for subjective assessment; deep automation is rare and often viewed skeptically by educators and institutions. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors, often under-resourced public/nonprofit programs, show slower AI tool adoption compared to corporate training or higher ed institutions with more digitization. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI meaningfully assists by generating test items, organizing item banks, auto-scoring objective components, and flagging grading patterns—freeing instructors to focus on performance and oral assessment. These tools demonstrably raise productivity while the instructor retains decision-making authority over grades. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can significantly assist with generating test items, rubrics, and first-pass grading/feedback, letting instructors focus on nuanced assessment and student-specific accommodations. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate test items and score objective components (multiple choice, short answer with rubrics), it cannot reliably assess performance tests, oral exams requiring nuanced judgment, or issue grades that reflect institutional standards without human oversight. The task requires contextual knowledge of student progress and institutional policy that AI currently cannot fully replace. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate test questions, rubrics, and even score written/oral responses with reasonable accuracy, but performance-based assessment and final grading judgment still require human oversight, especially for ESL speech evaluation nuances and accommodations. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educators bear professional and legal accountability for grades, which affects student transcripts, financial aid, and progression. Most institutions require credentialed instructors to sign off on final grades, and many have policies against full automation of assessment—creating organizational and regulatory friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Grading and certification often require instructor accountability and institutional grading policies; while no license is legally required to grade, accreditation and academic integrity norms create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI-assisted test generation and objective scoring can reduce administrative time, making costs roughly comparable to human labor when accounting for tool licensing, API costs, and necessary human oversight of grades. Significant human review requirements prevent dramatic cost reduction. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Test creation and grading via AI tools can be cheap per use, but the need for human review of oral/performance components and grade issuance keeps blended costs closer to comparable rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Products exist for automated test generation and objective scoring (learning management systems, automated grading tools), but they struggle with subjective assessment, oral performance evaluation, and grade-weighting decisions that reflect institutional policy. Production deployments are narrow in scope and typically require substantial human review. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like AI-based quiz generators, automated essay scoring, and speech assessment tools (e.g., for language proficiency) exist and are used in some ed-tech platforms, but they're not universally reliable across diverse adult learner populations or performance-based tests. |
Prepare objectives and outlines for courses of study, following curriculum guidelines or requirements of states and schools.
41CI 25–56 · exposure 38 · augmentation 88 · importance 3.6/5 · click for rater detail
Prepare objectives and outlines for courses of study, following curriculum guidelines or requirements of states and schools.
41| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Education is a traditionally laggard sector in AI adoption. While some districts pilot AI-assisted tools, production deployment of autonomous curriculum planning remains uncommon; adoption is slow due to institutional conservatism and risk-aversion around educational standards. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors are lower-digitization, resource-constrained environments where AI tool adoption for lesson planning is emerging but not yet widespread or systematic. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist instructors by generating initial outlines, suggesting learning objectives aligned with frameworks, and offering structural templates. Educators report substantial productivity gains when using AI as a drafting partner, though human judgment on pedagogy and compliance remains essential. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI is highly useful for brainstorming, drafting, and structuring course objectives and outlines, significantly speeding up the planning process while the instructor retains final judgment and compliance responsibility. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate draft course outlines and objectives quickly, the task requires meaningful pedagogical judgment, curriculum alignment with state standards, and adaptation to specific student populations. Current systems lack reliable understanding of nuanced state compliance and educational best practices needed for equal-quality output. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can draft course objectives and outlines quickly given curriculum standards, but aligning to specific state/school requirements and learner needs typically requires human review and customization, limiting full end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Schools must verify curriculum compliance with state education standards and local district policy; administrators and instructional leaders typically sign off on curricula. Liability concerns and regulatory oversight of educational standards create meaningful friction against full substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks AI-assisted planning, though schools may require instructor sign-off and adherence to state curriculum mandates, creating light oversight friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI inference is cheap, but integration requires educator oversight, validation against state standards, and revision to ensure quality. The total cost per usable curriculum product remains comparable to or exceeds having an instructor draft it directly, especially given required human review. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating a draft outline via AI costs cents in inference versus hours of instructor planning time, making it substantially cheaper even after factoring in review time. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No production system reliably performs this task end-to-end for educational institutions. AI can assist with outline generation, but educators report that outputs require substantial manual revision to meet state requirements and pedagogical soundness, making autonomous deployment rare. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like ChatGPT, curriculum-planning tools, and LMS AI features are used by educators to draft outlines, but they are assistive rather than autonomously producing compliant, finalized curricula at scale. |
Register, orient, and assess new students according to standards and procedures.
39CI 25–52 · exposure 38 · augmentation 75 · importance 3.5/5 · click for rater detail
Register, orient, and assess new students according to standards and procedures.
39| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adult education institutions are typically lower-digitization, budget-constrained organizations; while some use basic online registration, AI-driven orientation and assessment adoption remains in pilot phase in this sector. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education is a modestly digitized, often underfunded public-sector-adjacent field with slower technology adoption compared to corporate training or higher ed." |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist by pre-filling registration forms, generating candidate assessment items, suggesting diagnostic insights, and auto-flagging anomalies—allowing instructors to focus on dialogue, relationship-building, and final placement judgment more efficiently. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted intake systems, automated placement tests, and data dashboards meaningfully speed up registration and assessment while instructors retain the orientation and interpretive components. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with administrative registration (data entry, form completion) and generate initial assessments, the human-centered aspects—orientation requiring conversational responsiveness, social-emotional calibration, and nuanced diagnostic assessment for ESL placement—require trained instructor judgment and cannot currently be reliably automated end-to-end at 50% time savings with equal quality. |
| Task automatability | claude-sonnet-5 | 3/5 | Registration and standardized assessment components (intake forms, placement testing, scoring) can largely be automated, but orientation involving human rapport-building and contextual judgment about student needs resists full automation." |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational standards, accreditation requirements, and accountability frameworks typically mandate that instructors document and validate student placement and initial assessment, creating regulatory and institutional friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human perform registration or assessment, though some funding/regulatory frameworks (e.g., adult ed grant compliance) require documented human-administered procedures. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI solutions for registration (form automation) and basic placement tests are cheap, but they require instructor oversight to ensure accuracy and cultural appropriateness; the total cost including fallback human review approaches the cost of direct instructor-led registration. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated registration/testing platforms reduce staff time significantly, but licensing, integration, and human oversight for orientation keep costs roughly comparable to partial human staffing in many programs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Enrollment management systems can automate form intake, and some LMS platforms offer baseline placement tests, but no deployed product reliably performs the full task (registration + meaningful orientation + valid adaptive assessment) without substantial human intervention and oversight. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products for online registration, adaptive placement testing, and automated scoring exist and are used in adult education programs, but integrated end-to-end orientation-and-assessment workflows still require human coordination. |
Select, order, and issue books, materials, and supplies for courses or projects.
39CI 25–52 · exposure 38 · augmentation 63 · importance 3.2/5 · click for rater detail
Select, order, and issue books, materials, and supplies for courses or projects.
39| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions are slow to adopt automation in procurement and materials management; most still rely on manual requisition processes and human librarian/instructor judgment. Digitization of education is advancing, but procurement automation remains immature in K-12 and adult education sectors. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education programs are generally under-resourced and slower to adopt AI tools compared to corporate or tech-forward sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could usefully assist instructors by recommending materials based on course objectives, student level, and past usage, or by flagging inventory availability and price comparisons. However, instructors must retain final judgment over educational content and pedagogical fit, limiting augmentation to decision support rather than transformative productivity gain. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist in searching curricula-aligned materials, comparing suppliers, and drafting order lists, saving instructor time while they retain final decision-making. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could partially assist with inventory selection and ordering workflows, the task requires contextual judgment about course needs, student levels, and project requirements that vary substantially per course and institution. Current systems lack the domain knowledge and real-time visibility into educational contexts to execute end-to-end procurement autonomously at scale. |
| Task automatability | claude-sonnet-5 | 3/5 | Selecting appropriate materials benefits from AI recommendation and search, but ordering and physically issuing supplies requires system integration and physical handling that AI cannot fully do end-to-end.wingman |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions have formal purchasing policies, budget authority requirements, and approval chains that legally and procedurally restrict autonomous material procurement. Many districts require human sign-off on curriculum choices and vendor orders, creating hard organizational and compliance barriers. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for this administrative task, though institutional purchasing rules and budget approval processes create some friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Integration with existing institutional procurement systems, content databases, and approval workflows would require substantial customization. The labor cost of an instructor selecting materials is modest relative to AI infrastructure and integration overhead for reliable autonomous ordering. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-assisted catalog search and ordering tools can reduce time spent on selection, but human oversight and physical logistics keep costs roughly comparable rather than dramatically cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No mainstream educational procurement product reliably automates book/material selection and issuance for ESL or adult education without significant human review. While some LMS and inventory tools exist, they do not independently select or order materials—humans must still evaluate educational fit and approve purchases. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Procurement and inventory software with AI-assisted recommendations exist and are used in schools, but the full workflow (selection, ordering, physical distribution) still requires human coordination. |
Advise students on internships, prospective employers, and job placement services.
38CI 30–46 · exposure 33 · augmentation 75 · importance 3.0/5 · click for rater detail
Advise students on internships, prospective employers, and job placement services.
38| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adult education and ESL instruction remain relatively less digitized sectors with slower AI adoption compared to tech and finance. Institutions are mostly in pilot or early exploration phases of AI for student advising rather than at production scale for job placement automation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult/continuing education is a lower-digitization sector with slow AI adoption for personalized advising functions, though generic career tools are spreading slowly. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist instructors by rapidly surfacing employer information, identifying relevant internship matches, and curating job placement resources, freeing instructors to focus on personalized career counseling and student motivation. This augmentation is particularly valuable given the resource constraints many adult education programs face. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can effectively help generate resume drafts, interview prep materials, job search strategies, and research on employers, significantly aiding instructors who retain the relational and advisory role. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can partially automate parts of this task such as researching employer information, matching students to potential internships via databases, and providing general job placement resources. However, advising students requires judgment about individual student fit, motivation, and circumstances—outcomes that demand meaningful human discretion and personalized interpretation that AI cannot reliably replicate end-to-end with quality parity. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires personalized knowledge of the student's skills, local labor market, and relationship-building with employers, which current AI cannot fully replicate end-to-end despite being able to draft generic advice. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Educational institutions often have preferences for human advising relationships and pastoral care; some regulations or accreditation standards may implicitly or explicitly expect human-led student support. However, no strict legal mandate prevents AI from assisting with job placement information, creating moderate but real friction to full replacement. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically for career advising, but institutional expectations of personal mentorship, trust-building, and liability for guidance quality create moderate friction to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-driven job matching and employer research costs are very low per interaction compared to the loaded hourly wage of a human instructor performing research and advising work. The asymmetry strongly favors AI for the information-retrieval and matching components of the task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI chat costs are low, the human oversight, relationship management, and local employer network knowledge required keep the effective cost of a fully AI-driven service comparable to or higher than an instructor's marginal time on this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI products can research jobs and employers and generate initial job matches, deployed systems lack the personalized advising capability and contextual understanding needed for reliable real-world deployment in educational settings. Existing chatbots and job-matching tools exist but have material limitations in understanding individual student circumstances and career readiness. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Chatbots and career-advice tools exist but are not deployed as reliable substitutes for individualized advising in adult education settings; they remain supplementary at best. |
Prepare for assigned classes and show written evidence of preparation upon request of immediate supervisors.
37CI 16–59 · exposure 30 · augmentation 88 · importance 3.7/5 · click for rater detail
Prepare for assigned classes and show written evidence of preparation upon request of immediate supervisors.
37| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions are slower to adopt automation compared to tech or finance sectors. While some schools experiment with AI content tools, they remain pilot-stage; production-scale displacement of preparation work is minimal and hesitant. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Education sector adoption of AI for lesson planning is growing steadily with many pilot programs and tool integrations, but formal institutional policies and training lag behind faster-adopting sectors like tech and finance. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist instructors by generating initial lesson outlines, drafting activity descriptions, or summarizing content, allowing instructors to focus on adaptation and pedagogical strategy. This augmentation is already occurring in education but remains human-centered. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI tools substantially speed up drafting lesson plans, generating materials, and organizing preparation evidence, letting instructors focus on customization and adapting content to student needs while remaining in control. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | While AI can help generate lesson content or outline materials, the task fundamentally requires human judgment about pedagogical approach, student needs assessment, and curriculum alignment—all deeply tied to instructional experience. Full end-to-end automation with 50% time savings at equal quality is not achievable with current systems. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can draft lesson plans, materials, and preparation documentation quickly, but tailoring to specific class needs, student levels, and institutional requirements still requires human judgment and review, so only partial time savings accrue end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions have institutional norms, accreditation standards, and supervisor expectations that tie preparation approval to the instructor's professional judgment and accountability. Liability and quality assurance requirements create friction against pure AI substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | There's no licensing requirement mandating a human write lesson plans, but supervisors often require personalized, evidence-based documentation tied to specific student progress and curriculum standards, creating some organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | An instructor's preparation time is part of their salary; AI assistance tools cost money but do not eliminate the human labor entirely. The cost-benefit remains unfavorable because supervisory-approved preparation still requires significant instructor oversight and revision. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating a lesson plan draft via AI costs a fraction of a cent to a few cents in compute versus the teacher's hourly wage for the same prep work, though human review time still adds some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI tools can assist in drafting lesson plans and study materials, but no deployed product reliably performs the full task of preparing classes to supervisor standards. What exists (chatbots generating generic outlines) falls short of the contextual knowledge needed for effective classroom preparation. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like ChatGPT, lesson-planning tools, and LMS-integrated AI assistants are used by teachers today to draft lesson plans and materials, but reliability varies and outputs need instructor verification and customization. |
Review instructional content, methods, and student evaluations to assess strengths and weaknesses, and to develop recommendations for course revision, development, or elimination.
36CI 25–47 · exposure 33 · augmentation 75 · importance 3.5/5 · click for rater detail
Review instructional content, methods, and student evaluations to assess strengths and weaknesses, and to develop recommendations for course revision, development, or elimination.
36| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions, especially adult/ESL programs, digitize slowly and remain skeptical of outsourcing curriculum assessment to AI. While some analytics tools exist, production-scale adoption of AI for course evaluation remains rare; pilots are more common than deployments. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors are historically slow adopters of AI tools compared to corporate L&D or higher-ed edtech, with limited production-scale curriculum analytics deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist instructors by rapidly synthesizing student feedback, identifying common gaps in evaluation data, and suggesting revision options for human review. This augmentation can save instructors significant time on data preparation and initial idea generation while preserving instructional authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist by summarizing evaluation data, flagging trends, and drafting revision options, substantially speeding up the instructor's review process while they retain final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist with analyzing student evaluations and flagging content gaps, but the task fundamentally requires human judgment about pedagogical effectiveness, learner needs, and course strategy. No current system reliably performs end-to-end assessment with the nuanced understanding needed for adult education contexts at 50% time savings. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can analyze student performance data, summarize evaluations, and draft revision recommendations, but synthesizing pedagogical judgment about curriculum fit for adult learners still requires human oversight for validity and context. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Institutional barriers are moderate-to-strong: educational institutions typically require faculty judgment on curriculum, accreditation bodies may mandate instructor involvement in course review, and legal/contractual obligations often specify that instructional decisions rest with qualified educators rather than automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human perform this specific evaluative task, though institutional accreditation processes and administrative sign-off create some procedural friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Oversight costs for AI recommendations are substantial—instructors must verify outputs for pedagogical soundness and institutional fit. The total cost of AI inference, integration, and required human review approaches or exceeds the cost of a skilled instructor performing the review. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-assisted data aggregation and drafting can cut analysis time significantly, but human curriculum experts still must validate and contextualize outputs, keeping overall cost savings moderate rather than dramatic. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No production systems demonstrably perform this full task reliably. While LLMs can summarize student feedback and suggest content revisions, they lack the domain expertise and contextual understanding to assess instructional methods or make defensible course elimination recommendations at scale. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some LMS analytics and AI tools surface performance trends and generate reports, but no mature deployed product autonomously performs full curriculum review and revision recommendations in adult education settings. |
Observe and evaluate students' work to determine progress and make suggestions for improvement.
34CI 25–43 · exposure 30 · augmentation 75 · importance 4.4/5 · click for rater detail
Observe and evaluate students' work to determine progress and make suggestions for improvement.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Education, particularly adult education, remains a relatively low-digitization sector with significant institutional inertia and instructor autonomy. While some LMS systems include AI grading modules, production deployment at scale for evaluation and feedback remains limited; most adoption is still pilot or trial-phase. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors are relatively slow adopters of AI tools compared to corporate or tech-driven sectors, with pilots more common than widespread production deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI provides strong augmentation here: it can flag errors, surface patterns across a class, generate draft feedback, and highlight outliers, allowing instructors to focus on nuanced suggestion-making and one-on-one coaching rather than mechanical evaluation. This assistive role is already materially improving instructor productivity in deployed tools. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI writing feedback tools, plagiarism checkers, and language assessment platforms meaningfully assist instructors in efficiently reviewing student work and generating improvement suggestions, even though final evaluation and contextual judgment remain human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can evaluate written work and identify grammatical/factual errors at scale, but holistic assessment of progress requires understanding individual learning trajectories, motivation, and readiness for advancement—nuanced judgment that AI cannot reliably perform end-to-end. Current systems lack the contextual depth needed for meaningful improvement suggestions tailored to adult learners' circumstances. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can grade objective assessments and even provide feedback on writing, but observing in-class performance, participation, and holistic student progress requires human presence and contextual judgment that current systems cannot fully replicate. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions face regulatory and accreditation requirements for instructor oversight, and instructors often hold teaching credentials or professional standing tied to the judgment required for this task. Liability concerns around student progress assessments, combined with stakeholder (parents, students, institutions) expectations that a human make summative judgments, create strong friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement mandates a human evaluate every piece of student work, but institutional accreditation standards, teacher accountability, and student support norms create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference for bulk evaluation of assignments is substantially cheaper than instructor time per student, especially for initial pass/grading. Integration and oversight add cost, but the cost ratio still favors automation by a meaningful margin for large cohorts. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI tools can cheaply grade certain assignments, but comprehensive evaluation combining observation, contextual understanding, and tailored suggestions still requires substantial human oversight, keeping costs comparable to human labor for full task completion. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | LLM-based essay graders and rubric-scoring tools exist and are deployed in some educational settings, but they produce material error rates on non-standard writing, struggle with culturally or linguistically diverse responses, and rarely generate pedagogically sound suggestions. Products work narrowly and require significant human oversight. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Products like automated essay scoring or language-learning apps exist and provide some evaluative feedback, but they are narrow in scope and don't reliably assess overall classroom progress or adult learners' diverse needs. |
Prepare and implement remedial programs for students requiring extra help.
32CI 25–39 · exposure 25 · augmentation 75 · importance 3.7/5 · click for rater detail
Prepare and implement remedial programs for students requiring extra help.
32| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Education, particularly adult basic and ESL instruction, has been a relatively slow adopter of autonomous AI systems. Most deployments remain pilot-stage or supplementary; displacement of instructors in remedial roles is minimal outside of narrow supplemental tutoring contexts. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL programs are typically under-resourced, public-sector-adjacent, and slower to adopt sophisticated AI tools compared to corporate or tech-forward sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly augment instructors by generating personalized practice materials, analyzing student performance data to identify gaps, drafting lesson content, and providing real-time language feedback. These tools can substantially increase instructor productivity while the teacher retains responsibility for diagnosis, motivation, and adaptation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can generate customized practice materials, diagnostic quizzes, and progress tracking that meaningfully help instructors design and refine remedial interventions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate draft lesson plans and remedial materials, the task fundamentally requires understanding individual student learning gaps, adapting instruction in real-time, and providing personalized motivation—all of which demand sustained human judgment. AI tools can assist with content generation but cannot independently deliver the full program at the required quality level. |
| Task automatability | claude-sonnet-5 | 2/5 | Designing and implementing remedial programs requires diagnosing individual student needs, adapting pedagogy, and building rapport, which current AI cannot fully replace end-to-end despite being able to draft materials or suggest exercises. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions often have accreditation and accountability requirements that mandate qualified, credentialed instructors for remedial programs. There are also strong institutional and student expectations that remedial help involves human guidance, plus liability concerns around AI-only instruction for vulnerable populations. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing barrier prevents AI-assisted remedial tools, but institutional policies, funding requirements (e.g., accreditation, state adult ed standards) and preference for instructor oversight create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | An AI tutoring system requires significant upfront setup, customization, and ongoing monitoring to ensure quality. The all-in cost per student outcome remains comparable to or higher than direct instructor delivery, especially when accounting for the need for human oversight and intervention. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-based tutoring tools can be cheap per student interaction, but the instructor's diagnostic and program-design work still requires human time, so overall cost savings are moderate rather than dramatic. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Educational platforms with some AI tutoring capabilities exist, but they typically handle narrow domains and lack the adaptability needed for comprehensive remedial program implementation across diverse adult learners. No mature, production-scale system reliably replaces an instructor's diagnostic and motivational role in remedial education. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Adaptive learning platforms exist and are used in some settings, but they handle narrow skill practice rather than the full cycle of assessment, program design, and delivery for at-risk adult learners. |
Prepare materials and classrooms for class activities.
30CI 25–35 · exposure 25 · augmentation 63 · importance 4.2/5 · click for rater detail
Prepare materials and classrooms for class activities.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions have adopted AI for some content generation but have not moved toward automated classroom preparation systems. Most adoption remains in content design; physical classroom setup remains instructor-driven in practice. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction is a modestly digitized, resource-constrained sector where AI tool adoption for lesson planning is emerging but not yet widespread or deeply embedded. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist by drafting materials, suggesting activity layouts, and generating accessibility accommodations, improving instructor productivity in planning. However, the final assembly and classroom setup still depend heavily on human judgment and physical presence. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can significantly speed up drafting worksheets, quizzes, and lesson content tailored to student levels, meaningfully boosting instructor productivity in the material-preparation portion of this task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could help generate lesson materials (worksheets, reading lists), the physical preparation of classrooms—arranging furniture, setting up equipment, organizing supplies—requires on-site manual labor that current AI cannot perform. Material generation alone does not meet the 50% time-saving bar for the full task. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can help generate worksheets and lesson content, but physically arranging classrooms and preparing physical materials requires human presence and manual action that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Instructors are typically required to be present and responsible for their classroom environment by organizational policy and professional standards. Schools also expect human judgment about accessibility, safety, and student-specific needs, creating friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks material prep, though instructors are generally expected to personally prepare their own classroom and materials as part of professional practice, creating mild organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI material-generation tools are inexpensive, but the physical classroom setup still requires paid staff labor. The hybrid nature of the task (some automatable content creation, irreducible physical work) keeps overall cost comparable to or higher than a human instructor preparing their own space. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply produce some materials, but the physical setup portion still requires paid human labor, so overall cost savings versus a human doing the whole task are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI can draft educational materials and suggest classroom layouts, but no deployed product reliably handles the complete task end-to-end. Classroom preparation requires physical presence and contextual judgment about student needs that deployed systems struggle to match consistently. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Content-generation tools (e.g., lesson plan generators) exist and are used by educators, but no deployed product handles the physical classroom-setup component, limiting reliability across the full task. |
Train and assist tutors and community literacy volunteers.
30CI 30–30 · exposure 25 · augmentation 63 · importance 3.5/5 · click for rater detail
Train and assist tutors and community literacy volunteers.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adult education and nonprofit literacy sectors digitize more slowly than corporate training. While some nonprofits have piloted AI chatbots, sustained, production-level AI trainer replacement in community literacy programs remains limited and primarily experimental. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and community literacy programs are typically under-resourced, non-profit-driven, and slow to adopt AI tools compared to tech-forward sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist instructors by generating customized training modules, providing suggested feedback templates, or answering administrative questions. This support could raise instructor productivity, though the human instructor remains essential for real relationship-building and adaptive coaching with volunteers. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist by generating training curricula, practice materials, feedback templates, and answering tutor questions, boosting instructor efficiency while humans still lead training sessions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task involves interpersonal mentoring, skill assessment, and adaptive coaching—areas where current AI systems lack the judgment to handle end-to-end at quality parity. AI could help draft training materials or organize resources, but cannot replace the contextual feedback, relationship-building, and real-time assessment adjustments that tutors need. |
| Task automatability | claude-sonnet-5 | 2/5 | Training tutors and volunteers requires interpersonal facilitation, live modeling of teaching techniques, and responsive coaching that AI cannot fully replicate end-to-end today, though some content prep could be automated.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Institutional and regulatory barriers are moderate: schools and nonprofits may require human instructors to sign off on volunteer training; there is also organizational friction around adoption of AI for sensitive adult learning contexts. However, no strict legal requirement mandates human-only trainer delivery. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement mandates a human specifically, but volunteer trust-building, mentorship, and community relationship dynamics create significant organizational friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Creating bespoke training for community literacy volunteers requires domain expertise and customization. The cost of AI systems capable of meaningful trainer-like interaction plus human oversight would likely exceed the hourly cost of a qualified adult education instructor, especially given the specialized domain. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Human trainers still need to deliver most of the hands-on coaching and relationship-building, so AI only reduces some prep costs while the core labor cost remains largely intact. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate generic training content and answer factual questions, no deployed product reliably trains and assists tutors at scale with the nuance required for adult literacy contexts. LLM-based tutoring aids exist but have documented limitations in understanding learner needs and instructor contexts. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tools exist to generate training materials or scripted modules, but no deployed product reliably conducts full volunteer training and mentoring programs autonomously in real organizations. |
Participate in publicity planning, community awareness efforts, and student recruitment.
30CI 25–35 · exposure 25 · augmentation 63 · importance 3.1/5 · click for rater detail
Participate in publicity planning, community awareness efforts, and student recruitment.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions adopt marketing automation slowly, and adult education programs are among the least digitized sectors. Recruitment remains heavily relationship-driven and manual, with minimal evidence of AI agent displacement in this domain. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education programs are typically under-resourced, non-profit or public-sector entities with low digitization and slow AI tool adoption compared to corporate marketing functions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by drafting recruitment content, suggesting outreach strategies, and managing social media scheduling, which could raise instructor productivity in planning phases. However, the core recruitment and community-building work requires the human to remain central. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist by drafting outreach materials, social media content, flyers, and translated messaging for ESL populations, boosting instructor productivity in planning tasks. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with content generation and campaign planning, this task fundamentally involves human judgment about community needs, relationship-building, and persuasion that requires cultural sensitivity and authentic engagement. Recruiting students and building community awareness demand trust-building that AI cannot perform end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can draft flyers, social posts, and outreach copy but the actual community engagement, relationship-building, and event participation require human presence and judgment that current AI cannot substitute for end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong organizational and human-contact barriers exist: institutions typically require instructors to own community relationships and represent the program authentically, and there is inherent preference for human-to-human recruitment in educational contexts. Liability concerns around misrepresentation also create friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks AI assistance in publicity, though community trust-building and in-person recruitment favor human involvement, creating moderate organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI can reduce content creation costs, but the integration, oversight, and fact-checking needed for sensitive community messaging, combined with the irreplaceable human relationship work, makes total cost comparable to or higher than direct human effort for this task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply generate promotional drafts, but the human labor for community outreach, meetings, and recruitment events still dominates cost, keeping overall savings modest. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Chatbots and content tools can draft marketing materials and social media posts, but deployed AI products lack the ability to navigate community relationships, conduct face-to-face recruitment, and adapt messaging based on real community feedback at scale. Most real-world recruitment still relies heavily on human outreach. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Marketing content-generation tools exist and are used broadly for drafting materials, but no deployed product manages the full recruitment/publicity coordination and community relationship work reliably. |
Adapt teaching methods and instructional materials to meet students' varying needs, abilities, and interests.
29CI 25–34 · exposure 25 · augmentation 75 · importance 4.2/5 · click for rater detail
Adapt teaching methods and instructional materials to meet students' varying needs, abilities, and interests.
29| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | K–12 and adult education sectors adopt technology slowly; most classroom instruction remains teacher-led and relationship-dependent, with AI integration confined to narrow administrative or content-generation pilots rather than deep production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL programs are often under-resourced public/nonprofit sectors with slower technology adoption compared to corporate training or higher ed, so AI tool uptake remains at the pilot stage. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist by generating multiple instructional explanations, suggesting scaffolding strategies, flagging struggling students, and producing differentiated worksheets—allowing teachers to remain in control while raising their adaptive capacity. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully help instructors generate multiple versions of materials, translate concepts across proficiency levels, and suggest differentiated activities, significantly speeding up prep while the teacher retains judgment over delivery. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can help generate variant lesson materials and flag student learning patterns, but adapting teaching methods in real-time requires understanding individual learner psychology, classroom dynamics, and instructional judgment that current systems cannot reliably perform end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Adapting materials involves nuanced, real-time judgment about individual students' emotional, cognitive, and social needs that current AI cannot reliably assess or act on end-to-end; AI can help draft variants but cannot autonomously execute the full adaptive teaching cycle at equal quality.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Teachers are credentialed professionals; institutional governance, accreditation standards, and professional norms require human educators to own instructional design and student assessment decisions, creating substantial legal and organizational barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement blocks AI-assisted material adaptation, but instructional accountability, learner trust, and accreditation standards for adult/ESL education create moderate organizational friction against full delegation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | While AI can lower the marginal cost of generating materials, the instructor's loaded wage includes deep domain expertise, student relationship management, and real-time adaptive judgment that AI assistance does not displace significantly enough to be cheaper all-in. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI tools for generating differentiated materials are cheap per-item, but factoring in teacher oversight, contextualization, and verification for adult learners with diverse literacy levels, the net cost savings are moderate rather than dramatic. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Generative AI can draft alternative explanations and learning resources, but no deployed product reliably assesses student needs, tailors instructional methodology, and validates adaptation effectiveness in live classroom settings at scale. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some adaptive-learning products exist (e.g., differentiated worksheet generators, leveled reading tools) but they are narrow and don't handle the full spectrum of ability, interest, and motivational adaptation reliably in live classrooms. |
Instruct students individually and in groups, using various teaching methods, such as lectures, discussions, and demonstrations.
29CI 25–34 · exposure 25 · augmentation 75 · importance 4.2/5 · click for rater detail
Instruct students individually and in groups, using various teaching methods, such as lectures, discussions, and demonstrations.
29| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | K–12 and higher education adoption of AI in instruction has been cautious and slow; primary use is administrative or supplemental. Adult education and ESL programs remain heavily instructor-led with limited AI integration in core instruction, though online tutoring is growing. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult/ESL education sector has low digitization and funding constraints; AI adoption is mostly limited to pilot tutoring apps rather than systemic classroom deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist instructors by generating lesson outlines, creating practice materials, providing instant grammar or pronunciation feedback, and suggesting differentiation strategies. This frees instructors to focus on engagement, motivation, and personalized guidance—a high-value augmentation while the instructor remains central. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist instructors with lesson planning, generating discussion prompts, personalized practice materials, and language exercises, enhancing teaching efficiency while the instructor still leads. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate lecture content, demonstrations, and discussion prompts, the core interpersonal elements—responsive feedback, individualized pacing, real-time classroom management, and adaptive teaching to mixed student needs—require human presence and judgment. AI cannot yet replicate sustained group instruction with meaningful student interaction at scale. |
| Task automatability | claude-sonnet-5 | 2/5 | AI tutoring tools can deliver content and practice exercises, but live classroom instruction involving group dynamics, discussion facilitation, and adaptive in-person teaching remains beyond full automation today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Legal and regulatory barriers are moderate to strong: many jurisdictions require certified teachers for accredited education; institutional accreditation and funding models mandate human instructors; liability and duty of care rest with licensed educators. Student enrollment and continued attendance depend on perceived human engagement. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Adult education programs often require certified instructors and accreditation standards, and students value human interaction and mentorship, creating moderate institutional and credentialing friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI content generation and tutoring platforms reduce some preparation work, but integrating them into a classroom, managing student engagement, and handling diverse learner needs still demands skilled human instructors. Per hour of actual instruction, the blended cost is comparable to or higher than hiring teachers. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-based supplemental tutoring is cheap per interaction, but replacing an instructor's full multimodal teaching role still requires human oversight and blended delivery, keeping costs comparable overall. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI tutoring systems exist and can deliver content or run drills, but deployed products typically cover narrow topics and lack the adaptive responsiveness, motivational dynamics, and depth of support needed for mixed-ability adult education or ESL instruction. Most live educational delivery still requires human instructors. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tutoring products (e.g., Duolingo, Khanmigo) exist for individual practice, but no deployed system reliably conducts group instruction with discussions and demonstrations at scale in classrooms. |
Observe students to determine qualifications, limitations, abilities, interests, and other individual characteristics.
28CI 25–30 · exposure 25 · augmentation 50 · importance 4.3/5 · click for rater detail
Observe students to determine qualifications, limitations, abilities, interests, and other individual characteristics.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Education remains a laggard sector in AI automation. While some schools use dashboards for flagging at-risk students, systematic replacement of instructor observation with AI assessment is rare and faces cultural and regulatory resistance in most districts. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction, especially in community and nonprofit settings, show slow and uneven AI adoption compared to corporate or tech-sector environments, with pilots more common than production deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered dashboards and analytics can assist instructors by surfacing attendance patterns, grade trends, and flagging students who may need support, reducing time spent on data compilation. However, the core task of nuanced observation still requires the instructor in the loop. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools like adaptive assessments, language proficiency testers, and learning analytics can meaningfully assist instructors in identifying student needs and progress, complementing but not replacing direct observation. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can analyze some quantitative student data (test scores, attendance) and flag patterns, but meaningful observation requires sustained interaction, emotional/social cue detection, and contextual judgment about motivation and barriers that current systems handle poorly. Full end-to-end automation with 50% time savings at equal quality is not achievable today. |
| Task automatability | claude-sonnet-5 | 2/5 | Observation of students to gauge abilities, interests, and limitations relies heavily on in-person social perception, rapport, and contextual judgment that current AI cannot substitute for end-to-end, though some data-driven inference (quiz results, engagement metrics) can be automated. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions have strong preferences for human instructor judgment on student characteristics and needs, accreditation standards often require documented instructor assessment, and liability concerns around automated student profiling create organizational friction. Parents and students expect human observation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing barrier specifically blocks AI-assisted assessment, but the human-contact nature of adult ESL instruction and pedagogical norms around teacher-student relationships create moderate organizational and trust-based friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Basic learning analytics tools cost money and require instructor oversight to interpret, making the total cost approach or exceed the marginal cost of an instructor spending time on direct observation. No clear cost advantage exists. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While analytics dashboards are cheap to run, they only capture a narrow slice of what 'observation' entails; achieving equivalent insight would require significant human oversight, making all-in cost comparable to or higher than a teacher's embedded observation during instruction. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some learning management systems include basic analytics dashboards that flag struggling students, but no deployed product reliably captures the holistic qualitative assessment of individual characteristics, learning styles, and emotional/social factors that skilled instructors perform. Products exist at research/pilot stage only. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some adaptive learning platforms track performance metrics and flag skill gaps, but no deployed product reliably replicates a teacher's holistic observation of student characteristics like motivation, anxiety, or learning style in real classroom settings. |
Plan and conduct activities for a balanced program of instruction, demonstration, and work time that provides students with opportunities to observe, question, and investigate.
21CI 13–30 · exposure 17 · augmentation 63 · importance 4.2/5 · click for rater detail
Plan and conduct activities for a balanced program of instruction, demonstration, and work time that provides students with opportunities to observe, question, and investigate.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Despite enthusiasm for educational technology, actual replacement of classroom instruction by AI agents remains rare and mostly pilot-stage; most adoption centers on supplementary tools (tutoring, assessment) rather than core lesson planning and conduct. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction is a modestly digitized sector with slow, uneven AI adoption, mostly limited to supplementary tools rather than replacing instructional planning and delivery. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI tools can assist instructors by generating activity ideas, drafting lesson plans, and analyzing student data, raising planning efficiency; however, the live teaching and facilitation aspects remain firmly human-led, limiting transformative potential. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully help instructors brainstorm activities, create materials, and differentiate instruction, significantly boosting planning productivity even though delivery remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Planning and conducting instructional activities requires real-time responsiveness to student needs, dynamic adjustment of pedagogy, and judgment about when to pivot methods—capabilities that current AI cannot reliably replicate in a live classroom setting. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can help draft lesson plans and activities, but planning and conducting a balanced, responsive classroom program requires in-person facilitation, adaptation to student reactions, and pedagogical judgment that current AI cannot fully replace end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Most jurisdictions require licensed, credentialed educators to lead instruction; organizational, legal, and parental expectations strongly favor human teachers for direct student contact and accountability, creating substantial adoption friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no formal licensing barrier exists for lesson planning itself, teaching often requires certification and direct human presence for conducting classroom activities and interaction, creating moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A human instructor's loaded wage is competitive with or lower than the cost of building and maintaining an AI agent capable of live classroom facilitation, supervision, and differentiated instruction across diverse learners. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI tools can cheaply assist with content generation, but the human conducting activities, managing group dynamics, and interacting with students remains necessary, keeping overall costs comparable to human labor for the full task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate lesson plans and activity suggestions, no deployed product reliably conducts the full instruction cycle (demonstration, real-time facilitation, student engagement) in an actual classroom; current systems lack embodied presence and adaptive responsiveness. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Products like lesson-planning assistants and AI tutors exist, but no deployed system autonomously plans and conducts a full balanced classroom instruction cycle reliably at scale. |
Plan and supervise class projects, field trips, visits by guest speakers, contests, or other experiential activities, and guide students in learning from those activities.
21CI 16–25 · exposure 20 · augmentation 50 · importance 3.1/5 · click for rater detail
Plan and supervise class projects, field trips, visits by guest speakers, contests, or other experiential activities, and guide students in learning from those activities.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Education sectors are slow to adopt automation for core instructional duties; while administrative tools are widely used, displacement of supervision and experiential-learning guidance remains minimal. Most adoption is in planning aids, not substitution of the instructor role itself. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education programs are slow adopters of AI generally, and this hands-on supervisory task sees essentially no automation adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist by suggesting activities, drafting reflection prompts, organizing logistics, and providing post-activity analysis, but the instructor remains essential for live supervision and responsive guidance. Moderate augmentation potential in preparation and reflection phases. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with planning logistics, generating discussion questions, or drafting reflection prompts to help students learn from activities, even though it cannot supervise the activities themselves. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Planning project logistics and creating activity frameworks could be partially automated, but the core requirement—supervising students during activities and guiding real-time learning—demands human presence and judgment. AI cannot meaningfully replace the interactive guidance and adaptive feedback that experiential learning requires. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can help brainstorm project ideas or logistics text, but planning, supervising, and physically guiding students through field trips and guest-speaker events requires in-person coordination and judgment AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: education regulation typically requires qualified, in-person instructors to supervise and guide student learning experiences; liability for student safety during field trips and experiential activities creates legal accountability that cannot be transferred to AI; parental and institutional expectations reinforce human-instructor requirement. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Student supervision, safety, and liability concerns during off-site or guest activities create strong institutional and legal requirements for direct human oversight. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for activity planning are relatively cheap, but they address only a fraction of task cost. The human instructor's labor for supervision, adaptation, and guided reflection remains irreplaceable, making the all-in cost still heavily weighted toward the human role. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the human presence, supervision, and liability required, so there's no meaningful cost substitution possible for the core task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI can assist with generating project ideas and activity schedules, but no deployed system reliably supervises students or facilitates learning from field experiences in real classroom contexts. Prototypes exist for activity planning, but they lack the responsive, contextual guidance this task demands. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product supervises field trips or in-person experiential learning activities; this remains fundamentally a human coordination and safety task. |
Collaborate with other teachers and professionals in the development of instructional programs.
20CI 10–30 · exposure 17 · augmentation 63 · importance 3.5/5 · click for rater detail
Collaborate with other teachers and professionals in the development of instructional programs.
20| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Educational institutions, particularly those serving adult basic education and ESL populations, adopt technology slowly and remain skeptical of AI-led curriculum decisions. Human educator collaboration on program design remains the entrenched norm. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Education sector, especially adult and ESL instruction, has historically slow and uneven AI adoption for curriculum design collaboration, with pilots more common than production use. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by generating initial curriculum drafts, organizing research on best practices, or synthesizing feedback from multiple teachers, thereby saving time on documentation and idea compilation while educators focus on final judgment and alignment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist by generating draft materials, aligning content to standards, and suggesting activities, boosting productivity for teachers involved in collaborative program design. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Collaborative program development requires judgment, stakeholder alignment, and creative synthesis of diverse educational philosophies and expertise. AI cannot replicate the interactive negotiation, consensus-building, and contextual decision-making that defines effective instructional program collaboration. |
| Task automatability | claude-sonnet-5 | 2/5 | Collaborative curriculum development involves interpersonal negotiation, institutional context, and shared professional judgment that current AI cannot fully replicate end-to-end, though it can draft materials to support the process.”, , |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational program development typically requires licensed educators' professional judgment and institutional sign-off; accountability for curriculum quality rests with qualified educators. Regulatory and professional standards expect human educators to own instructional design decisions. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing barrier prevents AI assistance, but organizational norms, accreditation requirements, and the inherently social nature of collaboration create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of AI systems designed to coordinate multi-stakeholder program development, combined with required human oversight and refinement, exceeds the cost of teachers directly collaborating, especially given the quality mismatches in current systems. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Since the task fundamentally requires human interaction and consensus-building among staff, AI cannot substitute for the labor cost of collaboration, only supplement it modestly. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can draft curriculum documents or summarize existing frameworks, no deployed system reliably performs end-to-end collaborative development where multiple professionals must align on pedagogy and implementation. Drafting support exists but falls short of genuine collaborative engagement. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tools exist to help draft lesson plans or suggest curriculum ideas, but no deployed product actually performs the collaborative, multi-stakeholder aspects of program development reliably. |
Conduct classes, workshops, and demonstrations to teach principles, techniques, or methods in subjects, such as basic English language skills, life skills, and workforce entry skills.
19CI 9–30 · exposure 17 · augmentation 63 · importance 4.1/5 · click for rater detail
Conduct classes, workshops, and demonstrations to teach principles, techniques, or methods in subjects, such as basic English language skills, life skills, and workforce entry skills.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Adult education and ESL instruction occur primarily in schools, nonprofits, and community colleges—sectors that are laggards in AI adoption due to funding constraints, regulatory oversight, and organizational conservatism. No widespread displacement or production deployment of AI-led instruction in this domain is evident. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction occur in under-resourced public and nonprofit sectors with historically slow technology adoption and limited AI integration in classrooms. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist instructors through automated grading, lesson-plan generation, vocabulary drilling platforms, and personalized feedback on student writing, raising instructor productivity on certain task components. However, the core act of conducting live instruction with adaptive engagement remains squarely human-led. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can generate lesson plans, personalized practice materials, translation aids, and supplementary exercises that meaningfully enhance instructor productivity and learner practice outside class time. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time pedagogical interaction, classroom management, and adaptive teaching based on student engagement and comprehension—capabilities that current AI systems cannot reliably replicate end-to-end in a classroom setting. While AI can generate lesson content, it cannot conduct a live class with meaningful student interaction at 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Live classroom facilitation, workshop management, and real-time responsiveness to diverse adult learners require in-person presence, adaptive judgment, and interpersonal skills that current AI cannot replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | This task has hard regulatory and institutional barriers: instructors teaching in accredited adult education programs, secondary education, and ESL typically must hold appropriate teaching credentials, licenses, or certifications. Educational institutions also have legal and ethical obligations to provide human instruction, and students often prefer human teachers for language and life skills. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing mandate strictly requires a human instructor, but funding models, accreditation requirements, and learner preference for in-person support create meaningful friction against full substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI systems (chatbots, tutoring platforms) have nontrivial inference and integration costs, and they still require substantial human oversight, lesson design, and student support to supplement what they deliver. The total cost remains comparable to or exceeds the cost of a human instructor, especially when factoring in student dropout and learning loss. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI content generation is cheap, replacing live instruction requires human facilitation, technology infrastructure, and support staff, so all-in costs remain comparable to or only modestly below instructor costs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably conducts full classes or workshops with the responsiveness, emotional intelligence, and real-time adaptation that this task demands. AI can support lesson planning and content creation, but production systems do not replace the instructor in actual classroom execution. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tutoring products (e.g., Duolingo, language apps) exist for self-paced practice but no deployed system conducts full classes or workshops with group management and live demonstrations reliably. |
Confer with other staff members to plan and schedule lessons that promote learning, following approved curricula.
19CI 7–30 · exposure 13 · augmentation 63 · importance 3.2/5 · click for rater detail
Confer with other staff members to plan and schedule lessons that promote learning, following approved curricula.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions have been slow to adopt automation for core instructional planning. Most adoption remains at the individual lesson-drafting level (assistive), with collaborative curriculum planning still centered on human educators. Organizational culture and regulatory environment limit velocity. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult/ESL education is a modestly digitized sector with slow, uneven AI adoption in instructional planning workflows, mostly limited to pilot tools rather than embedded scheduling/coordination systems. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by generating draft lesson ideas, suggesting scheduling options, or summarizing curriculum requirements, helping instructors prepare more efficiently for staff meetings. However, the core collaborative and decision-making work remains human-centered, limiting transformation of the full task. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can meaningfully help draft lesson plans, suggest curriculum-aligned activities, and summarize discussions, enhancing efficiency even though the conferring itself remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Conferring with staff and planning lessons requires nuanced interpersonal coordination, curriculum interpretation, and contextual decision-making about learning outcomes. Current AI systems cannot reliably conduct real-time collaborative meetings, adapt to institutional norms, or make curriculum choices that reflect collective professional judgment. |
| Task automatability | claude-sonnet-5 | 2/5 | This task centers on interpersonal coordination and collaborative decision-making with colleagues, which AI cannot genuinely conduct end-to-end, though it can support planning materials.mt |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions have strong professional norms, accreditation oversight, and curriculum approval processes that typically require human instructor judgment and signoff. Liability and quality assurance expectations create material friction against full substitution of instructor planning. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing barrier specifically prevents AI involvement, but institutional norms of staff collaboration, curriculum compliance processes, and human coordination create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The time and integration cost of AI tools for lesson planning assistance would likely exceed the value for a task that fundamentally requires human judgment and coordination. Staff time is already allocated to planning; AI would add overhead rather than reduce labor cost. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply generate lesson plan drafts, but the core task—meeting and conferring with colleagues—still requires paid staff time regardless of AI availability, limiting cost savings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can draft lesson plans or suggest schedules in isolation, no deployed product reliably handles the collaborative, context-aware negotiation and decision-making this task requires. Some tools assist with individual lesson drafting, but staff conferencing and real-time schedule coordination remain primarily human-driven. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for the actual conferring and consensus-building among staff; this remains a human social process requiring real-time negotiation and relationship context. |
Prepare students for further education by encouraging them to explore learning opportunities and to persevere with challenging tasks.
14CI 7–21 · exposure 5 · augmentation 50 · importance 4.2/5 · click for rater detail
Prepare students for further education by encouraging them to explore learning opportunities and to persevere with challenging tasks.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | K–12 and adult education sectors show slow, cautious adoption of AI; pilots in learning analytics exist, but the irreplaceability of human mentorship in encouraging persistence means deployment remains limited and defensive rather than transformative. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education sectors have lower digitization and AI adoption compared to fast-moving professional/finance sectors, with pilots for tutoring tools but limited deployment for motivational mentorship. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist instructors by identifying at-risk students, suggesting targeted interventions, or generating discussion prompts, materially raising instructor productivity in outreach and monitoring, though the human remains essential for authentic encouragement. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can help instructors find learning resources, generate personalized materials, or track student progress, indirectly supporting encouragement efforts even though the human relationship is primary. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task fundamentally requires sustained interpersonal engagement, emotional support, and adaptive encouragement tailored to individual student psychology. Current AI cannot autonomously build the relational trust, model perseverance, or deliver the nuanced motivational interventions that define this work. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires building motivation, relationships, and personalized encouragement grounded in human trust and mentorship, which AI cannot execute end-to-end today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Educational institutions face regulatory requirements around student welfare, duty of care, and accountability that implicitly require qualified human educators to maintain learner engagement and progress; liability and student wellbeing concerns create substantial friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing barrier specifically blocks AI use, but strong human-contact and trust requirements, plus institutional norms around teacher-student mentorship, create real friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Integration of AI for motivational support (chatbots, analytics dashboards) still requires significant human oversight and supplementation; the all-in cost remains comparable to or higher than direct instructor engagement for equivalent motivational outcomes. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Since AI cannot substitute for the human relationship and motivational component, there is no meaningful AI cost comparison—the human is required. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate generic motivational scripts or identify students at risk of dropout via learning analytics, no deployed system reliably performs the core task of encouraging persistence through authentic interaction and personalized mentorship at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously performs the relational motivational coaching and perseverance-building this task describes; this remains inherently a human interpersonal function. |
Observe and evaluate the performance of other instructors.
11CI 5–16 · exposure 5 · augmentation 50 · importance 3.1/5 · click for rater detail
Observe and evaluate the performance of other instructors.
11| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions have been slow to adopt AI for high-stakes evaluation decisions. Adoption remains limited to supplementary tools and pilots rather than primary evaluation systems, with most organizations preferring human judgment for instructor assessment. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Adult education and ESL instruction is a low-digitization sector with minimal AI adoption for supervisory or evaluative functions; observation-based evaluation remains almost entirely manual. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist evaluators by auto-generating observation summaries from classroom video, flagging patterns of student engagement, or providing structured data on classroom activities. However, the human evaluator must remain central to judgment-making about instructional quality and effectiveness. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by transcribing lessons, analyzing speech patterns, flagging pacing or engagement metrics, or structuring feedback reports, but the core observational judgment remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Observing and evaluating instructor performance requires nuanced human judgment about pedagogical effectiveness, interpersonal dynamics, and contextual appropriateness. Current AI cannot reliably assess the quality of teaching methods, student engagement, or instructor-student interactions in ways that meet institutional evaluation standards. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires in-person or video-based observation of teaching practice, contextual judgment about pedagogy, and interpersonal feedback delivery that current AI cannot perform end-to-end reliably.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Instructor evaluation carries high stakes for employment, tenure, and compensation decisions, creating strong legal and liability barriers. Most institutions require qualified human evaluators (principals, department heads, or peer reviewers) to conduct official performance assessments, and there is significant organizational and union resistance to algorithmic replacement. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Institutional accreditation, HR policies, and often union or contractual requirements mandate qualified human supervisors or peer evaluators to conduct performance reviews, creating strong organizational and procedural barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-based classroom video analysis requires expensive infrastructure (camera systems, processing, integration with evaluation workflows) and human oversight to validate findings. This total cost approaches or exceeds the cost of having experienced instructors or administrators conduct evaluations. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this evaluative task at scale, so cost comparison favors the human evaluator by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can extract basic metrics from video (e.g., classroom activity detection, speech duration), no deployed system reliably performs comprehensive instructor evaluation as institutions require. Some pilot systems exist for classroom observation support, but they operate with significant limitations and typically require human validation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs classroom observation and instructor performance evaluation autonomously; this remains a human supervisory function in education settings. |
Meet with other professionals to discuss individual students' needs and progress.
9CI 5–13 · exposure 0 · augmentation 38 · importance 3.4/5 · click for rater detail
Meet with other professionals to discuss individual students' needs and progress.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Education sectors show slow AI adoption overall, and professional collaboration meetings are deeply embedded in institutional practice with few incentives to automate. This task is not a priority for digitization in most school systems. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors show low overall AI adoption for interpersonal, judgment-based collaborative tasks compared to faster-adopting sectors like finance or tech. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist by pre-generating student progress summaries or flagging data patterns before meetings, but such assistance is marginal compared to the core collaborative and interpersonal work of the meeting itself. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by summarizing student data, generating progress reports, or preparing talking points ahead of meetings, improving efficiency without replacing the collaborative discussion itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires human judgment about individual student needs, interpersonal understanding, and collaborative decision-making among professionals. Current AI cannot meaningfully participate in or replace these professional discussions, which depend on tacit knowledge and professional accountability. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires real-time interpersonal collaboration, shared professional judgment, and relationship-building among staff, which current AI cannot conduct autonomously in place of a human meeting participant. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Professional norms, institutional structure, and legal/ethical requirements for educator accountability create strong barriers. Education law and employment agreements typically require human instructors to participate in professional collaboration and student case discussions. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing requirement mandates a human for this specific meeting, but organizational norms, accountability for student outcomes, and need for professional judgment create moderate friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The task involves professional time spent in synchronous meetings with peers. AI cannot replace this meeting attendance or substitute for the professional judgment and dialogue required, making automation economically infeasible. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI product performing this task independently, so no meaningful cost comparison exists; the human cost remains the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product reliably performs professional peer consultations or collaborative case review meetings. While AI can summarize student data or draft notes, it cannot genuinely participate in or conduct these discussions. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for a human professional attending and contributing substantive judgment in interdisciplinary case discussions about students. |
Guide and counsel students with adjustment or academic problems or special academic interests.
8CI 0–16 · exposure 5 · augmentation 50 · importance 3.3/5 · click for rater detail
Guide and counsel students with adjustment or academic problems or special academic interests.
8| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Schools and ESL programs remain resistant to automating counseling roles; adoption is limited to supplementary resource databases rather than replacement systems. The low-digitization, human-centered nature of K–12 and adult education, combined with regulatory constraints, keeps velocity low. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education and ESL instruction sectors are generally slower AI adopters compared to fast-moving professional services, with counseling-specific AI tools seeing only pilot-stage use. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist instructors by suggesting diagnostic frameworks, recommending external resources, or helping draft communication to parents, raising efficiency on administrative aspects of counseling. However, the core empathetic and adaptive work remains human-driven, limiting augmentation to moderate gains. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help instructors draft resources, suggest interventions, or summarize student progress data, offering moderate assistance while the human retains the counseling role. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Guiding and counseling students requires nuanced emotional intelligence, understanding of individual circumstances, and adaptive relationship-building that current AI systems cannot perform end-to-end. While AI can provide information or draft suggestions, the core counseling function—establishing trust, reading nonverbal cues, and providing personalized emotional support—remains beyond current capabilities. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires building trust, reading emotional/contextual cues, and providing personalized human counsel about adjustment or academic problems—AI cannot reliably replace this relational, judgment-heavy task end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Educational institutions, accreditation bodies, and many jurisdictions legally require licensed educators or qualified counselors to provide academic and personal guidance to students. Parents and institutions typically mandate human contact for sensitive adjustment issues, creating hard legal and organizational barriers. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Counseling around personal/academic adjustment often involves duty-of-care, institutional policy, and sometimes mandated reporting obligations tied to the educator's role, creating strong organizational and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems designed to support counseling (chatbots, resource finders) still require substantial human oversight, making the integrated cost comparable to or exceeding a teacher's direct time investment. The liability and accountability for poor advice further increase true cost. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI chat costs are low, the task requires ongoing human oversight and relationship-building that adds integration and error-correction costs, making it not clearly cheaper than the instructor already performing this duty. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably performs student counseling independently; AI tools may assist with identifying resources or drafting referrals, but actual counseling sessions and meaningful adjustment guidance still require human educators. Early systems exist in limited educational contexts, but with significant gaps in reliability and scope. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed products perform holistic student counseling on adjustment or academic problems in real institutional settings; chatbots offer generic advice but are not trusted substitutes for this role. |
Attend professional meetings, conferences, and workshops to maintain and improve professional competence.
6CI 0–13 · exposure 0 · augmentation 38 · importance 3.2/5 · click for rater detail
Attend professional meetings, conferences, and workshops to maintain and improve professional competence.
6| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | There is no meaningful AI adoption for this task because the task is fundamentally about human professional presence and credentialing. Conference attendance rates have remained stable and dependent on human participation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Education sector, especially adult/ESL instruction, shows slower AI adoption generally, and this specific task (attending events) is not a target for automation initiatives. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide marginal assistance by summarizing conference sessions, identifying relevant sessions, or organizing notes afterward, but these supportive functions do not meaningfully enhance the core activity of attending and engaging at professional events. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help instructors find relevant conferences, summarize sessions, take notes, or synthesize takeaways, moderately boosting the value gained from attendance. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Attending meetings and conferences requires physical presence, networking, and real-time engagement. Current AI cannot substitute for the embodied, social, and interactive nature of professional development activities that depend on human presence and judgment. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical attendance and participation in professional development events cannot be performed by AI on a person's behalf; the task inherently requires human presence and engagement.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Professional development and conference attendance are legally and professionally required activities for maintaining teaching credentials and licensure. Regulatory bodies and accreditation standards mandate human attendance; AI cannot satisfy these requirements. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No legal requirement mandates attendance by a licensed human, but institutional/certification norms and professional development requirements create organizational friction against skipping this task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | This task involves travel, registration, and time investment that are inherently human-dependent. AI cannot reduce these costs—in fact, the instructor still pays for attendance regardless of any AI involvement. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this task, so cost comparison is moot—AI cannot replace the human activity itself. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously attend meetings, workshops, or conferences on behalf of humans. While AI can summarize conference materials or schedule events, it cannot fulfill the core requirement of human attendance and participation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for a human attending conferences or workshops; at most AI can summarize content afterward, not attend. |
Confer with leaders of government and community groups to coordinate student training or to find opportunities for students to fulfill curriculum requirements.
6CI 5–7 · exposure 0 · augmentation 38 · importance 3.1/5 · click for rater detail
Confer with leaders of government and community groups to coordinate student training or to find opportunities for students to fulfill curriculum requirements.
6| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Educational institutions remain largely analog in community coordination and stakeholder management. Adoption of AI for instructor-to-community liaison roles is minimal; institutions have not moved toward automation in this relational function. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Adult education is a modestly digitized public/nonprofit sector with slow AI adoption for external relationship management tasks specifically. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist by drafting outreach templates or flagging publicly available training opportunities, but the core task—conferring, negotiating, and judging fit—remains fundamentally human. Assistance is marginal and low-impact. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help draft communications, summarize meeting notes, track opportunities, and prepare talking points, providing moderate support to the instructor conducting these conferences. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time negotiation, relationship-building, and contextual judgment with human stakeholders to identify opportunities. Current AI cannot autonomously conduct meaningful conferences with government and community leaders or evaluate the fit between organizational opportunities and curriculum requirements. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires interpersonal relationship-building, negotiation, and situational judgment with external stakeholders that current AI cannot perform end-to-end.rn AI cannot autonomously represent an institution in these relational negotiations.rn |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: government and community partners expect to work with licensed educators; liability for misrepresenting curriculum or failing to vet opportunities rests with the instructor; and stakeholders prefer human relationship continuity and accountability. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Building trust with government and community leaders typically requires an accountable human representative with institutional authority; this is largely a relational/organizational barrier rather than a strict license requirement, but still strong. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The task's value lies in human credibility, relationship maintenance, and judgment about organizational fit—elements where AI has no cost advantage. Human instructors conducting outreach remain cheaper and more effective than any AI intermediary. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Human relationship management and trust-building with community/government leaders cannot be substituted by AI at any meaningful cost efficiency; a human must still do this work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product reliably performs independent stakeholder conferencing and opportunity coordination at scale. This requires sustained dialogue, trust-building, and knowledge of local contexts that current systems cannot handle without human direction. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product conducts autonomous stakeholder liaison and coordination with government/community leaders on behalf of instructors. |
Establish and enforce rules for behavior and procedures for maintaining order among the students for whom they are responsible.
5CI 0–10 · exposure 5 · augmentation 38 · importance 3.9/5 · click for rater detail
Establish and enforce rules for behavior and procedures for maintaining order among the students for whom they are responsible.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Schools remain low-digitization, high-human-contact environments where behavior management is viewed as a core human responsibility. There is negligible adoption of AI for enforcement of classroom rules in educational settings. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Adult education and ESL instruction remain low-digitization, human-presence-dependent settings with minimal AI adoption for classroom management tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist by analyzing patterns in behavior incidents, suggesting strategies, or helping draft clearer policies, but the core task of real-time interaction and judgment remains human-centered. Useful support tools exist but do not transform teacher productivity on enforcement itself. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help draft rule documents or suggest behavior management strategies, but it offers little real-time assistance for enforcing order among students. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Establishing and enforcing behavioral rules requires real-time judgment about context, individual student needs, emotional dynamics, and adaptive responses that current AI systems cannot perform end-to-end. This fundamentally involves human authority, presence, and social interaction that AI cannot replicate. |
| Task automatability | claude-sonnet-5 | 1/5 | Establishing classroom rules and maintaining order requires in-person authority, judgment about individual students, and real-time behavioral response that current AI cannot perform end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and organizational barriers protect this task: teachers hold formal authority delegated by institutions and parents, students have educational rights, and schools face liability for how discipline is managed. Human educators must legally hold responsibility for classroom conduct. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Maintaining classroom order is tied to instructor authority, institutional policy, and in-person supervisory responsibility, creating strong structural and role-based barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying AI to attempt this task—including required human oversight, failures, and liability—vastly exceeds the value of partial automation. A teacher's role in maintaining order is inseparable from their presence and authority, making substitution economically nonsensical. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this function, so cost comparison favors the human instructor by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can suggest behavior management frameworks or help draft classroom policies, no deployed product reliably executes the enforcement or judgment aspects of this task in practice. Classroom management requires live interaction and contextual authority that current systems lack. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product manages classroom discipline or enforces behavioral norms among adult learners in physical or live settings. |
Enforce administration policies and rules governing students.
3CI 0–5 · exposure 0 · augmentation 25 · importance 3.7/5 · click for rater detail
Enforce administration policies and rules governing students.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Educational institutions have not adopted AI for policy enforcement and disciplinary functions, reflecting both regulatory constraints and fundamental institutional resistance to removing human authority from student conduct matters. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Adult education and ESL instruction sectors have low AI adoption for disciplinary/administrative enforcement tasks, which remain manual and instructor-driven. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI might assist with minor tasks like flagging attendance patterns or organizing incident records, but the core enforcement function—deciding consequences, communicating with students, and exercising pedagogical judgment—cannot be meaningfully augmented by current AI tools. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help track attendance, flag policy violations, or draft documentation, but it provides limited direct assistance to the act of enforcing rules with students. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Enforcing policies and rules requires contextual judgment, student relationship management, and discretionary decision-making that current AI cannot perform autonomously. This task is fundamentally about behavioral management and student interaction, which demands human authority and presence. |
| Task automatability | claude-sonnet-5 | 1/5 | Enforcing policies and rules requires in-person authority, judgment about context, and real-time behavioral management that AI cannot perform end-to-end; this is inherently a human institutional role. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal, institutional, and ethical barriers exist: only licensed educators have authority to enforce school policies, student discipline requires documented human decision-making, and liability for automated enforcement is prohibitive. Human contact and judgment are legally mandated. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Enforcement of institutional rules typically requires designated staff authority, accountability, and often compliance with educational regulations, creating strong organizational and quasi-legal barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying, monitoring, and overseeing AI for policy enforcement, combined with liability concerns and required human oversight, would far exceed the cost of human instructors performing this task directly. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this enforcement function, so cost comparison favors the human by default since AI cannot deliver the output at all. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can reliably perform student discipline, policy enforcement, or rule administration in real educational settings. This requires institutional authority and human judgment that AI tools do not possess today. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product enforces classroom or administrative discipline policies; this remains a research-irrelevant, human-executed function tied to physical presence and authority. |
Attend staff meetings and serve on committees, as required.
3CI 0–5 · exposure 0 · augmentation 38 · importance 3.4/5 · click for rater detail
Attend staff meetings and serve on committees, as required.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | There is no adoption velocity to measure because the task is fundamentally non-automatable. Organizations have no path to replace human meeting attendance with AI. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Education sector adoption of AI for governance/representation functions like committee participation is essentially nonexistent and not a target of current automation efforts. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could modestly assist by preparing briefing materials, summarizing prior decisions, or drafting talking points before a meeting, but these are peripheral supports to the core task of human participation and deliberation. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help prepare meeting notes, summarize agendas, draft talking points, or take/transcribe minutes, moderately assisting the instructor's participation in these activities. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Attending meetings and serving on committees inherently require human presence, real-time interaction, and decision-making in social contexts. AI cannot meaningfully participate in these activities as a committee member or meeting attendee in any current deployed system. |
| Task automatability | claude-sonnet-5 | 1/5 | Attending staff meetings and serving on committees requires physical/virtual presence, real-time interpersonal deliberation, and institutional representation that AI cannot substitute for today.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Organizational policy, employment contracts, and governance structures explicitly require human staff to attend meetings and serve on committees. Legal and administrative frameworks mandate human participation in institutional decision-making. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Institutional governance, employment expectations, and the need for a human representative with authority and accountability create strong organizational barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The task cannot be automated, so cost comparison is not applicable. AI tools might assist with meeting preparation, but the core task of attendance and committee service requires a human and thus remains entirely human-cost-bearing. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this role, so cost comparison favors the human by default since AI cannot deliver the output at all. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No production AI system can attend a staff meeting or serve on a committee. While AI can summarize meetings or assist with administrative follow-up, it cannot participate as a required attendee or voting member. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product attends meetings or serves as a committee member on behalf of a human employee; this remains outside current product capabilities. |
Provide students with disabilities with assistive devices, supportive technology, and assistance accessing facilities, such as restrooms.
0CI 0–0 · exposure 0 · augmentation 25 · importance 3.9/5 · click for rater detail
Provide students with disabilities with assistive devices, supportive technology, and assistance accessing facilities, such as restrooms.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Educational institutions cannot and do not automate disability accommodation and physical facility assistance. This sector remains highly human-dependent for compliance and ethical reasons, with no meaningful AI adoption in this specific task. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Physical caregiving and accessibility support in educational settings show essentially no AI adoption or displacement trend. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could marginally assist by helping catalog facility accessibility or recommending assistive devices, but the core task—hands-on provision and human support—cannot be augmented meaningfully by current systems. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Assistive technology (e.g., communication devices, screen readers) can support the broader task, but the direct physical assistance component sees minimal AI-driven productivity enhancement. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical assistance, real-time judgment about individual disability needs, facility navigation, and hands-on support that current AI systems cannot perform. AI cannot physically provide devices, navigate real buildings, or assist students to restrooms. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires physical presence, hands-on assistance, and navigation of physical spaces (e.g., helping a student reach a restroom), which current AI cannot perform.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal law (ADA) explicitly requires that accommodations and assistive technology provisioning be delivered by qualified human staff; legal liability for inadequate accommodation is severe, and the requirement for direct human contact with vulnerable populations creates hard regulatory and duty-of-care barriers. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Physical assistance to students with disabilities involves duty-of-care, safety, and often legal/regulatory obligations (e.g., IEP/ADA compliance) requiring a responsible human, creating hard barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The task is not automatable, so cost comparison is moot. Even advisory AI systems would require human oversight and action, making them an additional cost rather than a replacement. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute for physical assistance, so the human cost is the only viable option; AI cannot replace this at any cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product performs the physical and interpersonal components of this task. This is fundamentally a human-presence task involving accommodation, dignity, and physical assistance that no autonomous system addresses in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product provides physical assistance or accompanies students to facilities; this remains entirely a human caregiving function. |
Related occupations — Educational Instruction & Library
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.