Library Science Teachers, Postsecondary

25-1082.00
Median wage $80,340/yr3,630 employed (US)Rank #372 of 923 scored · top 40% by substitution

Teach courses in library science. Includes both teachers primarily engaged in teaching and those who do a combination of teaching and research.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution31
Exposure28
Augmentation61

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

25 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

8%

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%28

panel mean rating 2.1/5 → substitution pressure 28/100

Technical feasibility todayw 20%28

panel mean rating 2.1/5 → substitution pressure 28/100

Cost vs. human wagew 15%33

panel mean rating 2.3/5 → substitution pressure 33/100

Adoption barriersw 20%inverted — strong barriers lower the score38

panel mean rating 3.5/5 (barrier strength) → substitution pressure 38/100

Sector adoption velocityw 10%28

panel mean rating 2.1/5 → substitution pressure 28/100

Task breakdown (25 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.

Maintain student attendance records, grades, and other required records.

78

CI 70–86 · exposure 80 · augmentation 75 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5Postsecondary institutions have nearly universal adoption of LMS and student information systems; automation of attendance and grade recording is standard practice across higher education.
Sector adoption velocityclaude-sonnet-54/5Higher education has broadly adopted digital LMS platforms for attendance and grade tracking for years, representing deep, mature adoption in this specific administrative function.
Augmentation potentialclaude-haiku-4-5-202510014/5Beyond full automation, AI can assist instructors by flagging attendance anomalies, predicting at-risk students from grade patterns, and generating summary reports, meaningfully reducing administrative burden while maintaining human oversight.
Augmentation potentialclaude-sonnet-54/5AI-enabled LMS tools significantly reduce time spent on manual recordkeeping, letting instructors focus on review and final approval rather than data entry.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI and automation tools can handle record maintenance, data entry, and grade calculation end-to-end with minimal human intervention. Learning management systems and student information systems already automate most of these functions, easily achieving 50% time savings or more.
Task automatabilityclaude-sonnet-54/5Recordkeeping of attendance and grades is a structured, rule-based data entry and calculation task well suited to existing LMS/SIS automation and AI-assisted tools, though setup and integration with institutional systems is needed.
Adoption barriersclaude-haiku-4-5-202510013/5While no law requires human administrators for record-keeping per se, institutional policies, FERPA compliance oversight, and data governance frameworks create moderate friction; many institutions still require human verification or sign-off on records.
Adoption barriersclaude-sonnet-52/5Some institutional policy requires instructor sign-off on final grades, but no legal requirement mandates human-only recordkeeping, so barriers are modest.
Cost vs. human wageclaude-haiku-4-5-202510015/5Institutional LMS and SIS software costs per student-record are orders of magnitude cheaper than hiring administrative staff to manually maintain these records, and integration costs are minimal for established systems.
Cost vs. human wageclaude-sonnet-54/5Automated gradebook/attendance software costs a small fraction of the instructor time it replaces, though there is nonzero licensing and administrative overhead.
Technical feasibility todayclaude-haiku-4-5-202510015/5Mature, production-deployed systems (Canvas, Blackboard, Banner, Workday) reliably perform attendance tracking, grade recording, and record management at scale across thousands of institutions daily.
Technical feasibility todayclaude-sonnet-54/5Learning management systems (Canvas, Blackboard, etc.) already automate gradebook calculations and attendance tracking in production at most universities, though some manual input and oversight remain.

Prepare course materials, such as syllabi, homework assignments, and handouts.

76

CI 76–76 · exposure 75 · augmentation 100 · importance 4.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Adoption is visible in pilot and early-production phases across universities, particularly in large systems and well-resourced institutions, but remains uneven. Many departments still rely on manual preparation; widespread routine use is emerging but not yet dominant in the sector.
Sector adoption velocityclaude-sonnet-53/5Higher education is adopting generative AI for course prep at a moderate pace, with many pilots and informal use but no institution-wide production deployment as standard practice yet.
Augmentation potentialclaude-haiku-4-5-202510015/5AI strongly augments faculty productivity by drafting initial syllabi, generating diverse problem sets, and creating accessibly formatted handouts, freeing instructors to focus on pedagogical refinement and customization. Most faculty using these tools retain full control and judgment, making this a clear augmentation scenario.
Augmentation potentialclaude-sonnet-55/5AI tools substantially speed up drafting of syllabi, assignments, and handouts while the instructor retains control over final content, pedagogy, and accuracy.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI systems can generate syllabi, homework assignments, and handouts with minimal human input using prompts or templates. While final review and customization for course-specific pedagogy are typically required, AI can produce 60–80% of the content in production-ready form, achieving well over 50% time savings at acceptable quality.
Task automatabilityclaude-sonnet-54/5Drafting syllabi, homework assignments, and handouts is largely a text-generation task that current LLMs handle well when given course objectives and topics, with instructor review needed mainly for accuracy and alignment.
Adoption barriersclaude-haiku-4-5-202510012/5No legal license or regulatory requirement mandates that a human author course materials; institutional policies may require faculty sign-off, but this is light organizational friction rather than a hard barrier. Academic tradition favors faculty ownership, but technical/legal substitution barriers are minimal.
Adoption barriersclaude-sonnet-52/5No licensing requirement blocks AI-assisted drafting, though institutional policies, accreditation standards, and academic freedom norms create some friction around fully automated course design.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI inference cost (marginal, cents per assignment) plus lightweight integration is orders of magnitude cheaper than faculty preparation time, which carries full loaded wages. A single course material generation session costs pennies in AI resources versus hours of faculty labor.
Cost vs. human wageclaude-sonnet-55/5Generating a draft syllabus or assignment set via an LLM costs a few cents in compute versus hours of faculty time, an order of magnitude cost difference.
Technical feasibility todayclaude-haiku-4-5-202510014/5Large language models and educational AI tools are deployed in real academic settings to draft course materials. Products like ChatGPT, specialized educational platforms, and LMS integrations routinely generate syllabi and assignments; while human review is standard practice, the systems perform the core drafting task reliably at scale.
Technical feasibility todayclaude-sonnet-54/5Widely deployed tools like ChatGPT, Copilot, and course-design AI assistants are already used by instructors to draft syllabi and assignments, though customization to specific pedagogy and institutional requirements still requires human editing.

Write grant proposals to procure external research funding.

69

CI 55–84 · exposure 70 · augmentation 88 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Academic and research institutions, information-sector employers of postsecondary teachers, are among the fastest adopters of AI for document composition and grant support. Universities and research offices are rapidly integrating AI drafting tools into grant workflows, with widespread pilot and early production adoption.
Sector adoption velocityclaude-sonnet-53/5Higher education is a moderate-adoption sector; AI writing assistants are increasingly used for grant drafting and literature reviews, but institutional caution and grant-specific compliance rules slow full-scale adoption.
Augmentation potentialclaude-haiku-4-5-202510015/5AI strongly augments grant writing: LLMs draft sections, suggest framing, integrate citations, and accelerate iteration, allowing human researchers to focus on strategic vision, novelty, and persuasive narrative rather than mechanical composition. The human researcher stays in the loop and is substantially more productive.
Augmentation potentialclaude-sonnet-54/5AI substantially speeds up drafting, editing, and formatting of proposal sections, letting faculty focus on research design and strategic framing, making it a strong augmentation tool even though full automation is limited.
Task automatabilityclaude-haiku-4-5-202510015/5Grant proposal writing involves structured document composition (literature review, methodology outline, budget justification, narrative framing) that LLMs and AI writing systems can generate at high quality with 50%+ time savings. Current AI can draft full proposals, handle formatting, generate boilerplate sections, and integrate research findings with minimal human oversight for equal or superior output.
Task automatabilityclaude-sonnet-53/5AI can draft substantial portions of grant proposal text (background, literature review, boilerplate sections) but cannot independently generate the novel research vision, budget justification tailored to institutional specifics, or ensure strategic alignment with funder priorities without heavy human revision.
Adoption barriersclaude-haiku-4-5-202510012/5No licensing or legal requirement mandates a human write proposals; grants are evaluated on merit. Modest barriers include institutional preference for human authorship, funder skepticism of AI-authored content, and institutional policies still under development. These are soft, not hard barriers to AI adoption.
Adoption barriersclaude-sonnet-52/5No licensing requirement, but funders and universities expect original scholarly voice and accountability from the named PI, and academic integrity/originality norms create some friction against full AI authorship.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI writing tools cost $20–100/month or per-use inference fees, generating a full proposal in minutes. A grant writer's loaded cost is $60–100/hour, and proposal drafting takes 40–80 hours. AI cost per proposal is orders of magnitude lower, delivering comparable or superior output at 1–2% of human labor cost.
Cost vs. human wageclaude-sonnet-53/5AI drafting tools are cheap relative to faculty time, but the human oversight, subject-matter expertise, and iterative revision required keep total costs roughly comparable to traditional writing time savings being partial rather than total.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed products (ChatGPT, Claude, specialized grant-writing platforms like GrantWriterLab, Proposal.ai) demonstrably assist and substantially automate grant drafting in production use by academic institutions and research offices. Some material variability exists in domain specificity and compliance with funder requirements, but mature offerings handle this task reliably at scale.
Technical feasibility todayclaude-sonnet-53/5Products like ChatGPT, Grantable, and Writefull are used by academics to draft and edit grant sections, but no deployed system reliably produces submission-ready proposals without extensive faculty review and editing.

Compile bibliographies of specialized materials for outside reading assignments.

60

CI 39–81 · exposure 62 · augmentation 88 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Higher education and library science have lagged in AI adoption compared to finance or software; while some institutions experiment with AI-assisted resource discovery, production deployment for bibliography compilation remains limited, and cultural/professional resistance to automation of core academic functions remains high.
Sector adoption velocityclaude-sonnet-53/5Higher education is moderate in AI adoption—individual faculty increasingly use AI search tools, but institutional-level integration into curriculum tasks remains uneven.
Augmentation potentialclaude-haiku-4-5-202510014/5AI assistants meaningfully augment this task by rapidly generating formatted lists, suggesting related materials, and reducing clerical work around citation formatting, allowing instructors and librarians to focus on critical evaluation and pedagogical alignment—a strong use case for assistive AI even if end-to-end automation is limited.
Augmentation potentialclaude-sonnet-55/5AI greatly speeds up locating, filtering, and formatting specialized sources, letting instructors focus on curating quality and pedagogical relevance.
Task automatabilityclaude-haiku-4-5-202510013/5AI can assist in identifying relevant materials, generating initial bibliographies, and formatting references with high consistency. However, subject expertise and curation of 'specialized materials' appropriate to specific curricula and reading levels typically require human judgment, preventing fully automated end-to-end performance that meets the 50% time-saving threshold without substantial oversight.
Task automatabilityclaude-sonnet-54/5AI can search, identify, and format specialized bibliographies quickly using literature databases and citation tools, though verifying accuracy and relevance still requires some human review.4
Adoption barriersclaude-haiku-4-5-202510014/5Academic library practices are governed by professional standards, institutional curricula approval, and faculty authority over reading lists; libraries face institutional inertia and professional norms favoring human expert curation, and there is often implicit requirement that assignment design and resource selection remain human-vetted for accreditation and pedagogical integrity.
Adoption barriersclaude-sonnet-51/5No licensing or regulatory requirement exists for bibliography compilation; it's a routine academic administrative task with no liability constraints.
Cost vs. human wageclaude-haiku-4-5-202510012/5While AI inference is cheap, the task requires domain expertise integration and human verification of recommendations and accuracy, making the total cost (AI + human oversight) approach or exceed the loaded wage of a library science instructor or support staff performing this task.
Cost vs. human wageclaude-sonnet-55/5AI-based literature search and citation compilation is vastly cheaper than a faculty member manually curating sources, often near-instant versus hours of work.
Technical feasibility todayclaude-haiku-4-5-202510013/5Deployed systems (AI writing assistants, citation managers with AI features) can generate bibliographies and identify sources reliably in many domains, but they struggle with nuance in 'specialized materials' selection, can miss domain-specific collections, and sometimes hallucinate citations—requiring material human review in production academic settings.
Technical feasibility todayclaude-sonnet-54/5Tools like reference managers, AI search assistants, and citation generators (e.g., Perplexity, Elicit, Zotero integrations) are already deployed and widely used for compiling reading lists and bibliographies.

Compile, administer, and grade examinations, or assign this work to others.

52

CI 51–54 · exposure 50 · augmentation 75 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Higher education has begun integrating automated grading tools and LMS-based exam systems, but adoption remains uneven and often piloted rather than deeply embedded. Many institutions remain hesitant to delegate high-stakes assessment to AI without human faculty review, limiting velocity.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI grading and exam tools unevenly and cautiously, with pilots more common than widespread production use in postsecondary teaching in general, and library science being a niche field with slow tech diffusion.
Augmentation potentialclaude-haiku-4-5-202510014/5AI substantially augments exam compilation (question generation, item analysis, difficulty calibration) and administration (test delivery, automated scoring), allowing instructors to focus on feedback and pedagogical refinement. Faculty oversight remains central, but productivity gains on routine assessment tasks are significant.
Augmentation potentialclaude-sonnet-54/5AI tools substantially speed up exam question generation, rubric creation, and first-pass grading feedback, letting instructors focus on final review and edge cases.
Task automatabilityclaude-haiku-4-5-202510013/5AI can automate significant portions: generating exam questions, administering computer-based tests, and basic grading of objective items. However, constructing pedagogically sound exams, setting appropriate difficulty, and grading subjective responses (essays, conceptual understanding) still require human judgment, preventing full end-to-end automation at the required quality threshold.
Task automatabilityclaude-sonnet-53/5AI can help compile and grade many exam formats (especially objective and short-answer) with strong quality, but grading nuanced written work and ensuring exam validity for a specific curriculum still needs faculty oversight, so only partial time savings are realized end-to-end.
Adoption barriersclaude-haiku-4-5-202510013/5Academic institutions typically expect faculty to maintain pedagogical authority and accountability for assessment; there is organizational friction and professional norms around faculty ownership of evaluation. However, no strict legal licensing barrier prevents use of AI for examination administration and grading, only institutional policy.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement for grading itself, but academic integrity policies, grade appeals, and institutional accreditation norms create moderate friction requiring faculty accountability for final grades.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-driven exam administration and grading of objective content is substantially cheaper than human labor per task equivalent, with minimal marginal cost per assessment. Full replacement is limited by subjective grading, but the assisted pathway keeps costs well below human-equivalent wages for the automatable portions.
Cost vs. human wageclaude-sonnet-54/5AI-based exam generation and automated grading tools are dramatically cheaper per assessment than faculty time, especially for large courses, though setup and review add some cost.
Technical feasibility todayclaude-haiku-4-5-202510013/5Deployed products (learning management systems with automated quiz grading, AI content generators) exist and function reliably for objective testing and some assignment grading. However, reliable performance on complex subjective assessment and exam design tailored to specific learning outcomes remains limited and typically requires human oversight.
Technical feasibility todayclaude-sonnet-53/5Deployed products (LMS AI graders, quiz generators, plagiarism/rubric-based scoring tools) reliably handle objective exams and offer draft feedback on essays, but are not fully trusted for high-stakes grading without human review.

Edit manuscripts for professional journals.

47

CI 37–56 · exposure 42 · augmentation 75 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While some journals use AI tools for copyediting screening, the adoption of AI for substantive manuscript evaluation remains slow in academic publishing due to conservative norms, reputation risk, and the centrality of human expert judgment to peer review.
Sector adoption velocityclaude-sonnet-52/5Academic publishing and higher education are historically slower adopters of AI tools for substantive editorial work, though grammar/formatting tools have modest uptake.
Augmentation potentialclaude-haiku-4-5-202510014/5AI excels at augmenting human editors by flagging grammar, style issues, reference formatting, and consistency; editors using these tools can work significantly faster while maintaining full control over acceptance and substantive feedback decisions.
Augmentation potentialclaude-sonnet-54/5AI substantially speeds up grammar, clarity, formatting, and citation-checking work, letting the human editor focus on substantive content judgments.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with copyediting, grammar checking, and formatting, but manuscript evaluation requires human judgment on academic merit, novelty, and fit—core aspects that current systems cannot reliably perform end-to-end with 50% time savings at equal quality.
Task automatabilityclaude-sonnet-53/5AI can draft copyedits, check grammar, consistency, and citation formatting effectively, but substantive scholarly editing (assessing argument quality, disciplinary relevance, methodological soundness) still requires human judgment, so only part of the task meets the 50% threshold.
Adoption barriersclaude-haiku-4-5-202510014/5Journal publishers and institutions typically require human editors and peer reviewers to take legal/professional responsibility for editorial decisions; liability for accepting flawed work and professional standards around academic gatekeeping create strong friction against full automation.
Adoption barriersclaude-sonnet-52/5No formal licensing requirement bars AI-assisted editing, but journal reputational standards and editorial board accountability create some institutional friction against fully automated editing decisions.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-powered copyediting and proofreading tools are inexpensive per use (often SaaS subscriptions), making the cost per task substantially lower than paying a professional editor's hourly rate, though integration oversight adds some cost.
Cost vs. human wageclaude-sonnet-54/5AI copyediting and formatting assistance costs a small fraction of a human editor's or professor's time for equivalent surface-level editing work, though substantive review still requires paid expert time.
Technical feasibility todayclaude-haiku-4-5-202510013/5AI writing assistants and grammar tools are deployed in academic workflows, but they require significant human oversight and cannot make substantive editorial decisions; no production system fully handles peer-review-level manuscript editing autonomously.
Technical feasibility todayclaude-sonnet-53/5Tools like Grammarly, Writefull, and LLM-based editors are used in production for copyediting and language polishing, but no deployed product reliably handles full scholarly manuscript review including substantive peer-editing decisions.

Develop and teach online courses.

41

CI 28–54 · exposure 45 · augmentation 88 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Higher-education institutions have widely adopted learning management systems and AI content tools, but deployment remains concentrated in administrative and asynchronous components rather than full course autonomy; faculty adoption of AI co-development is nascent and variable across disciplines.
Sector adoption velocityclaude-sonnet-53/5Higher education has moderate AI adoption for course design and content creation tools, but actual teaching delivery by AI remains rare and cautious due to institutional and accreditation constraints.
Augmentation potentialclaude-haiku-4-5-202510015/5AI dramatically augments faculty productivity in course development—generating discussion prompts, auto-grading assignments, personalizing learning pathways, and drafting lecture notes—allowing instructors to focus on mentorship and adaptive pedagogy while maintaining full pedagogical authority and student relationship.
Augmentation potentialclaude-sonnet-54/5AI substantially helps instructors draft syllabi, generate quizzes, create multimedia content, and produce course materials, meaningfully boosting productivity while the instructor remains central.
Task automatabilityclaude-haiku-4-5-202510014/5AI can automate much of the course development pipeline (content structuring, assessment generation, video transcription, lecture slides) and deliver asynchronous instruction with near-total autonomy, saving >50% of development and delivery time. However, real-time student interaction, live discussion facilitation, and adaptive feedback still benefit from human pedagogical judgment.
Task automatabilityclaude-sonnet-52/5Course design requires pedagogical judgment, curriculum alignment, and live teaching/interaction that current AI cannot fully replicate end-to-end despite being able to draft materials.
Adoption barriersclaude-haiku-4-5-202510014/5Accreditation bodies, institutional policies, and faculty unions often mandate human faculty design and oversight of courses; tenure protections and intellectual property concerns around course materials create friction; student outcomes expectations and institutional liability for academic integrity remain tied to credentialed instructors.
Adoption barriersclaude-sonnet-54/5Postsecondary teaching typically requires credentialed faculty, accreditation standards, and institutional oversight, creating strong structural barriers to full AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510013/5AI infrastructure and oversight for online course delivery approaches parity with adjunct/contract instructor costs when amortized across enrollment, but full replacement still requires human course ownership and student support, narrowing the cost advantage.
Cost vs. human wageclaude-sonnet-52/5While AI can cut content-drafting time, the human instructor's ongoing teaching, grading, and interaction still dominate cost, keeping AI only modestly cheaper for the full task.
Technical feasibility todayclaude-haiku-4-5-202510013/5Products exist (LMS platforms with AI-powered content suggestion, generative AI tools for lecture prep, automated grading systems) and are deployed in production educational settings, but material gaps remain in maintaining academic rigor, detecting plagiarism reliably, and handling nuanced student support that institutions expect from faculty.
Technical feasibility todayclaude-sonnet-52/5AI tools (course builders, content generators, chatbots for TA support) exist but no deployed product independently develops and teaches a full accredited postsecondary course reliably.

Evaluate and grade students' class work, assignments, and papers.

38

CI 29–48 · exposure 38 · augmentation 63 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI grading assistance in postsecondary institutions remains sparse and experimental; most universities have not integrated automated grading into production workflows, and faculty resistance to algorithmic evaluation of student work remains high in most disciplines.
Sector adoption velocityclaude-sonnet-52/5Higher education adoption of AI grading tools is still in early pilot stages with cautious, uneven uptake across institutions and disciplines, slower than fast-moving corporate sectors.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by generating preliminary feedback comments, suggesting rubric scores to review, or flagging potential plagiarism, which can accelerate faculty grading; however, the augmentation is partial because the instructor must always verify and take responsibility for final grades.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully speed up feedback generation, flag issues, and suggest grades, letting instructors review and finalize faster while retaining final judgment.
Task automatabilityclaude-haiku-4-5-202510012/5Grading routine assignments with clear rubrics could be partially automated, but evaluating papers and class work at the postsecondary level typically requires nuanced judgment about argument quality, originality, and pedagogical progress that current AI systems struggle with reliably. Meaningful automation would cover only lower-order components like format checking or simple completion verification.
Task automatabilityclaude-sonnet-53/5AI can draft feedback and score structured assignments against rubrics, but nuanced grading of papers requiring judgment of originality, argument quality, and context still needs faculty oversight, so only partial time savings are realized at equal quality.
Adoption barriersclaude-haiku-4-5-202510014/5Significant barriers exist: institutional policies typically require faculty sign-off on grades, institutional liability concerns about algorithmic grading, accreditation and grade integrity requirements, and strong cultural/legal norms that a human instructor must be responsible for evaluation in higher education. These create real friction against full automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement mandates human grading, but academic integrity concerns, institutional policy, and accreditation expectations create moderate friction against full automation of grading responsibility.
Cost vs. human wageclaude-haiku-4-5-202510013/5AI inference and prompt engineering for grading assistants is relatively cheap, but meaningful oversight and human spot-checking of grades remain necessary, bringing all-in costs closer to paying a teaching assistant for partial grading work rather than achieving major cost reduction.
Cost vs. human wageclaude-sonnet-53/5AI grading assistance is cheap per assignment, but required human review and calibration to avoid errors narrows the net cost advantage over a professor's grading time, which is already a partial-time task.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI tools can assist with scoring and provide feedback suggestions, no deployed product reliably grades complex postsecondary papers and assignments at the quality and consistency expected in academic evaluation. Existing tools work at a narrow scope (multiple choice, simple rubrics) and are rarely used as the sole grading authority in production courses.
Technical feasibility todayclaude-sonnet-53/5AI writing assessment and grading-assistant tools (e.g., automated essay scoring, LLM-based feedback generators) are deployed in some ed-tech products, but postsecondary faculty rarely rely on them exclusively due to accuracy and fairness concerns.

Select and obtain materials and supplies, such as textbooks.

37

CI 30–44 · exposure 33 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Higher education institutions adopt procurement automation slowly; most libraries still rely on manual review processes and librarian expertise. Limited digital transformation in library operations relative to other sectors.
Sector adoption velocityclaude-sonnet-52/5Higher education administrative and curricular processes adopt AI slowly, with textbook selection remaining a largely manual, instructor-driven decision even where AI tools are available for other tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by searching vendor catalogs, filtering by price and specifications, and comparing reviews, helping librarians focus on pedagogical evaluation and institutional fit decisions.
Augmentation potentialclaude-sonnet-54/5AI can effectively assist by summarizing textbook options, comparing content coverage, checking pricing, and drafting justifications, meaningfully speeding up the human-led selection process.
Task automatabilityclaude-haiku-4-5-202510012/5Selecting materials requires domain expertise, budget constraints, and institutional knowledge that current AI systems struggle with reliably. While AI could assist in cataloging or filtering options, the human judgment needed to match materials to curriculum and students keeps automatability low.
Task automatabilityclaude-sonnet-53/5AI can help identify, compare, and recommend textbooks and materials based on syllabi and course goals, but final selection involves judgment about pedagogy, budget, and institutional fit that still requires human decision-making and procurement steps.
Adoption barriersclaude-haiku-4-5-202510013/5Library budgets, vendor relationships, and accreditation standards create moderate friction; institutional procurement policies and the need for human judgment on educational fit create adoption friction but no hard legal barriers.
Adoption barriersclaude-sonnet-52/5No licensing requirement dictates who selects materials, but institutional purchasing policies, budget approval chains, and instructor autonomy create moderate organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI assistance in procurement search and filtering costs less than hiring additional staff, but the task still requires significant human oversight to ensure quality selection, making the all-in cost comparable to having a part-time procurement specialist.
Cost vs. human wageclaude-sonnet-52/5While AI-assisted research is cheap, the procurement, vendor negotiation, and approval workflows still require human effort, so overall cost savings are modest rather than order-of-magnitude.
Technical feasibility todayclaude-haiku-4-5-202510012/5No production systems reliably perform full material selection for academic libraries; procurement tools exist but require substantial human curation and decision-making. Chatbots can help search vendors but cannot independently evaluate pedagogical fit or institutional needs.
Technical feasibility todayclaude-sonnet-52/5There is no widely deployed product specifically for automating textbook/material selection and procurement in academic settings; general AI search/recommendation tools can assist but aren't purpose-built or reliably used in production for this task.

Plan, evaluate, and revise curricula, course content, course materials, and methods of instruction.

28

CI 25–30 · exposure 25 · augmentation 75 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Higher education adoption of AI for curriculum design remains tentative and primarily experimental, with many institutions still establishing governance. Faculty resistance and accreditation uncertainty slow deployment compared to faster-moving sectors like software or finance.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI tools slowly for core curricular decisions, with pilots for content generation more common than production-level curriculum redesign.
Augmentation potentialclaude-haiku-4-5-202510014/5AI can meaningfully assist faculty by generating initial drafts, suggesting content improvements, and helping organize materials, significantly reducing preparation time while faculty retain final judgment on pedagogy and learning objectives. This is already in use in some institutions for course-design assistance.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by generating draft materials, suggesting readings, summarizing trends, and helping evaluate course content, substantially speeding up preparatory work while the instructor retains final judgment.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can generate course outlines, materials, and draft syllabi quickly, the task requires pedagogical judgment about learning outcomes, institutional fit, and student needs that currently demand substantial human oversight. AI cannot reliably evaluate and revise curricula based on classroom assessment data and student feedback without significant human guidance.
Task automatabilityclaude-sonnet-52/5AI can draft syllabi or suggest readings, but genuine curriculum planning requires institutional judgment, accreditation alignment, and pedagogical evaluation that current systems cannot autonomously perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510014/5Faculty curriculum oversight is often governed by accreditation standards, department policies, and institutional review processes that legally or professionally require human faculty sign-off. Many institutions have formal curriculum committees and accreditation bodies that mandate human expertise in course design.
Adoption barriersclaude-sonnet-53/5No licensing requirement mandates a human specifically, but tenure structures, faculty governance, and accreditation processes create organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for curriculum generation have low inference costs, but the integration, customization, and necessary human review for educational institutions mean total cost remains substantial relative to a faculty member's time savings on initial drafting only, not full curriculum management.
Cost vs. human wageclaude-sonnet-52/5Faculty time is expensive, but the substantial human review and institutional approval needed to validate AI-suggested curriculum changes limits net savings today.
Technical feasibility todayclaude-haiku-4-5-202510012/5Products exist for generating course content and materials, but no deployed system reliably performs the full cycle of planning, evaluating, and revising curricula for postsecondary education with the quality and contextual judgment required. Most implementations remain in pilot or assistance mode rather than autonomous operation.
Technical feasibility todayclaude-sonnet-52/5No deployed product independently plans and revises postsecondary curricula in production; existing tools are narrow drafting aids requiring heavy faculty oversight.

Advise students on academic and vocational curricula and on career issues.

28

CI 25–30 · exposure 25 · augmentation 63 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While some institutions pilot AI-assisted advising (chatbots for FAQ, course lookup), production deployment of AI as the primary advisor remains rare; adoption is slow and concentrated in administrative triage, not core decision-making, reflecting institutional caution around liability.
Sector adoption velocityclaude-sonnet-52/5Higher education is a slower-adopting sector for interpersonal advising functions, though AI chatbots for basic student services are being piloted in some institutions.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist advisors by pulling curriculum data, summarizing career trend information, and drafting preliminary recommendation frameworks, meaningfully raising advisor productivity for routine case preparation without replacing judgment-heavy advising conversations.
Augmentation potentialclaude-sonnet-54/5AI can help advisors quickly retrieve curriculum requirements, career data, and draft communications, meaningfully speeding up the informational component of advising while humans handle the relational and judgment-based aspects.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can draft generic guidance on career paths and curricula, advising students requires contextual judgment about individual circumstances, learning styles, and nuanced career-fit assessment that current AI cannot reliably perform end-to-end at the quality threshold. The task involves substantive personalization and real-time adaptation that exceeds what off-the-shelf systems deploy today.
Task automatabilityclaude-sonnet-52/5Advising requires nuanced understanding of an individual student's history, goals, institutional policies, and relationship-building that current AI cannot fully replicate end-to-end.dreams reliably.
Adoption barriersclaude-haiku-4-5-202510014/5Academic advising is a fiduciary responsibility in higher education institutions; advisors must document decisions and bear responsibility for student outcomes, creating strong institutional and legal barriers to full automation. Accreditation standards and student expectations typically require human accountability in advising.
Adoption barriersclaude-sonnet-53/5No formal licensure required for most academic advisors, but institutional policies, accreditation expectations, and student trust in human mentors create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Integrating AI advising systems requires significant domain customization, instructor oversight, and liability management, making the all-in cost comparable to or exceeding the cost of a faculty advisor's time when quality and oversight demands are accounted for.
Cost vs. human wageclaude-sonnet-52/5While AI chat tools are cheap per interaction, the need for human review and the liability/reputational cost of bad advice keeps effective cost comparable to or higher than a trained advisor for substantive guidance.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI chatbots and career-matching tools exist as proof-of-concept, but no mature production system reliably advises students on complex academic and vocational curricula decisions with the judgment and accountability expected in higher education. These systems remain narrow and high-error on edge cases.
Technical feasibility todayclaude-sonnet-52/5Chatbot advising tools exist for basic FAQs and scheduling but no production system reliably handles nuanced academic/career advising for postsecondary students without human oversight.

Participate in student recruitment, registration, and placement activities.

28

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Postsecondary education has been slow to adopt AI for recruitment and placement; most institutions still rely on human recruitment officers, advisors, and placement coordinators. Pilots with AI-assisted tools exist, but production displacement remains minimal and adoption is lagging relative to digital-native sectors.
Sector adoption velocityclaude-sonnet-52/5Higher education administrative functions have seen slow, uneven AI adoption, with pilots in chatbots for admissions but limited integration into faculty-led recruitment and placement duties.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist recruitment staff by automating email triage, organizing applicant data, and flagging placement matches, raising staff productivity on routine tasks. However, the strategic aspects of recruitment strategy and personalized student-employer matching remain human-driven, limiting transformative potential.
Augmentation potentialclaude-sonnet-53/5AI can assist with drafting recruitment materials, managing registration data, and analyzing student placement patterns, meaningfully aiding faculty without replacing their judgment.
Task automatabilityclaude-haiku-4-5-202510012/5Student recruitment, registration, and placement involve substantial human interaction, relationship-building, and contextual judgment that current AI cannot fully automate. While AI can assist with email outreach and data entry, the core task of persuasion, personalized guidance, and placement matching requires human discretion and cannot achieve 50% time savings at equal quality end-to-end.
Task automatabilityclaude-sonnet-52/5This task involves interpersonal outreach, advising, and judgment-based placement decisions that require human relationship-building; AI can support but not replace the end-to-end process today.
Adoption barriersclaude-haiku-4-5-202510014/5Educational institutions typically require human staff involvement in recruitment (compliance, relationship-building with students and employers) and placement decisions (advisement, legal accountability). Professional judgment and institutional liability create strong adoption barriers to full automation.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement mandates a human, but institutional norms, personalized advising expectations, and faculty governance roles create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI services for outreach and registration (chatbots, email automation) cost less per interaction than staff time, but end-to-end recruitment and placement require human judgment that dominates the cost equation. The blended cost remains comparable to or exceeds hiring dedicated recruitment staff.
Cost vs. human wageclaude-sonnet-52/5AI tools can lower costs for routine communications but human faculty involvement in recruitment events, interviews, and placement decisions remains necessary, keeping overall costs comparable to human-driven processes.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably performs the full recruitment-to-placement workflow. CRM and registration systems exist, but they are tools rather than autonomous agents; human staff must conduct recruitment conversations, evaluate student fit, and negotiate placements. AI chatbots handle narrow parts (scheduling, FAQ) with material limitations.
Technical feasibility todayclaude-sonnet-52/5Some CRM and chatbot tools assist with recruitment inquiries and registration logistics, but no deployed product handles the full recruitment-registration-placement cycle for postsecondary faculty tasks reliably.

Prepare and deliver lectures to undergraduate or graduate students on topics such as collection development, archival methods, and indexing and abstracting.

25

CI 25–25 · exposure 25 · augmentation 63 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI in postsecondary teaching remains pilot-stage and often resistance-heavy; institutions are experimenting with AI for administrative tasks and student support, but autonomous lecture delivery by AI faces slow adoption due to accreditation, quality concerns, and faculty governance.
Sector adoption velocityclaude-sonnet-52/5Higher education is generally slow to formally adopt AI as a replacement for instructors, though piecemeal use of AI for lecture prep and slide generation is growing among individual faculty.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist instructors by drafting lecture content, generating quiz questions, summarizing readings, and curating examples on library science topics, thus raising preparation efficiency. However, the augmentation is partial because the instructor must still deliver, engage students, and make real-time pedagogical decisions.
Augmentation potentialclaude-sonnet-54/5AI tools can significantly speed up drafting lecture notes, generating examples, summarizing readings, and creating supplementary materials for topics like indexing and archival methods, while the instructor retains delivery and judgment.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can generate lecture outlines and draft content on library science topics, delivering live or synchronous lectures requires real-time student interaction, adaptive pacing, and pedagogical judgment that current systems cannot reliably replicate. The task remains heavily dependent on human presence and responsiveness.
Task automatabilityclaude-sonnet-52/5AI can help draft lecture content and slides, but live delivery, adaptive teaching, and interactive classroom engagement with students still requires substantial human involvement, so full end-to-end automation with equal quality is not yet achievable.
Adoption barriersclaude-haiku-4-5-202510014/5Significant barriers exist: accreditation bodies and institutions legally require qualified faculty to design and deliver curricula, student learning outcomes accountability rests with the institution, and there is strong institutional and regulatory expectation that instruction be led by a credentialed educator.
Adoption barriersclaude-sonnet-54/5Accredited postsecondary teaching typically requires a credentialed, subject-expert instructor of record, and institutions have strong organizational and accreditation-driven preferences for human faculty delivering instruction.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI lecture generation and delivery systems are still nascent and expensive per course deployment, while a faculty member's cost is already sunk within institutional budgets; the economic case for substitution is weak for a single course, though economies of scale might shift this over time.
Cost vs. human wageclaude-sonnet-52/5While AI can reduce prep time cheaply, the delivery portion still requires a paid instructor or at minimum recorded/produced content with human oversight, keeping costs comparable to or only modestly less than a human lecturer.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably substitutes for postsecondary instruction delivery; lecture capture and content generation exist, but educational institutions have not adopted AI to teach accredited courses autonomously at scale. Student feedback mechanisms and curriculum accreditation remain human-centered.
Technical feasibility todayclaude-sonnet-52/5AI content-generation tools are used to help draft lecture materials, but no deployed product autonomously delivers university-level lectures with pedagogical judgment and live student interaction at scale.

Initiate, facilitate, and moderate classroom discussions.

25

CI 20–30 · exposure 20 · augmentation 50 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Postsecondary education remains a laggard sector for instructor replacement; while some institutions experiment with online discussion forums and AI-augmented tools, there is minimal production displacement of classroom facilitation roles. Adoption is exploratory rather than deployive.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI unevenly and cautiously for live teaching functions, with pilots more common in course design/grading than real-time discussion facilitation.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can provide useful assistance by generating discussion questions, summarizing student contributions, or identifying gaps in dialogue, helping instructors prepare and reflect. However, the augmentation is supplementary rather than transformative, as the human instructor remains fully in the loop during live moderation.
Augmentation potentialclaude-sonnet-53/5AI can help instructors prepare discussion prompts, summarize contributions, or suggest follow-up questions, meaningfully aiding preparation and facilitation without replacing the live moderation itself.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can generate discussion prompts and summarize points, facilitating live classroom discussions requires real-time responsiveness, emotional intelligence, and adaptive pedagogical judgment that current systems cannot reliably deliver end-to-end at 50% time savings. The core value—moderating interpersonal dynamics and guiding intellectual exchange—remains dependent on human presence.
Task automatabilityclaude-sonnet-52/5Leading discussions requires real-time reading of student engagement, adaptive follow-up questioning, and classroom management that current AI cannot reliably replicate end-to-end.'
Adoption barriersclaude-haiku-4-5-202510014/5Educational institutions have strong institutional and regulatory barriers: faculty governance, accreditation requirements that mandate instructor-led instruction, and contractual/union protections. Moreover, there is substantial organizational and student preference for human-led classroom experience, particularly in discussion-based pedagogy.
Adoption barriersclaude-sonnet-53/5No licensing bars AI from assisting, but institutional norms, accreditation expectations, and student preference for human interaction create moderate friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5Even if partial automation were feasible (e.g., generating discussion scaffolds), the infrastructure, oversight, and integration costs would likely exceed the loaded salary cost of a faculty instructor whose primary value is presence and real-time judgment.
Cost vs. human wageclaude-sonnet-52/5Even where AI chat facilitation exists, the need for human oversight and the specialized nature of live moderation keeps costs comparable to or above instructor time for this specific task.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably moderates live classroom discussions in production educational settings. Chatbots can participate in asynchronous forums or generate prompts, but they cannot meaningfully facilitate synchronous group discourse with the contextual awareness and authority required in a classroom.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously runs live classroom discussions for postsecondary courses; existing tools support discussion boards or prep but don't moderate live sessions.

Keep abreast of developments in the field by reading current literature, talking with colleagues, giving presentations at conferences, and serving on committees in professional associations.

24

CI 13–35 · exposure 17 · augmentation 63 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Academia has low adoption velocity for professional engagement tasks because faculty autonomy, reputation, and committee service are core to academic identity and governance. Even information-sector organizations (universities) resist automating what is seen as essential human professional participation.
Sector adoption velocityclaude-sonnet-52/5Academia adopts AI tools slowly and unevenly; literature review aids are used but professional engagement activities remain largely unchanged by AI adoption trends.
Augmentation potentialclaude-haiku-4-5-202510013/5AI tools can usefully assist with literature review (summarization, trend detection) and presentation preparation, moderately reducing preparation time. However, augmentation is limited to support tasks; the core activities—genuine colleague dialogue, conference attendance, and committee deliberation—remain fundamentally human-driven.
Augmentation potentialclaude-sonnet-54/5AI tools can significantly assist in tracking literature, drafting presentation materials, and summarizing developments, meaningfully boosting efficiency even though the human must remain engaged in professional community activities.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires human judgment, relationship-building, and genuine engagement with a professional community. While AI can summarize literature or draft presentations, it cannot authentically participate in colleague conversations, serve on committees, or develop the tacit professional networks that constitute the core of staying abreast in a field.
Task automatabilityclaude-sonnet-52/5AI can help surface and summarize literature, but the core activity (networking, presenting, committee service, professional judgment) requires human presence and cannot be end-to-end automated.
Adoption barriersclaude-haiku-4-5-202510014/5Professional association committees require a licensed expert's signature, judgment, and accountability. Academic institutions and professional bodies expect human experts in leadership roles, not AI. Credibility in a field depends on human presence and reputation, creating strong organizational and institutional friction against automation.
Adoption barriersclaude-sonnet-52/5No formal licensing requirement, but professional norms and social/institutional expectations around participation and reputation create moderate friction to full substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5This task involves professional development and networking that creates value through human presence and credibility. AI cannot substitute for a faculty member's actual attendance, participation, and relationship-building, so the all-in cost of AI assistance (literature tools + human time) does not meaningfully reduce the human burden or wage requirement.
Cost vs. human wageclaude-sonnet-52/5AI can cheaply summarize papers, but the bulk of the task (attending conferences, networking, serving on committees) still requires paid human time, so overall cost savings are limited.
Technical feasibility todayclaude-haiku-4-5-202510012/5Current AI systems can assist with reading literature (summarization, search) or drafting presentation materials, but no deployed product can autonomously maintain professional engagement, network relationships, or serve committee roles. The human must remain the active agent in professional community participation.
Technical feasibility todayclaude-sonnet-52/5Tools exist for literature alerts and summarization, but no product substitutes for conference presentation or committee participation at scale in production.

Select and invite guest speakers to speak to classes.

23

CI 11–35 · exposure 13 · augmentation 50 · importance 2.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Higher education is moving slowly on this; while some institutions experiment with speaker databases and matching tools, actual AI-driven speaker selection remains pilot-stage. Most departments still rely on faculty networks and manual curation.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI slowly for interpersonal and administrative relationship tasks like this, with most AI use confined to content creation, not outreach coordination.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by generating candidate lists from academic networks, drafting initial outreach emails, and tracking speaker availability, allowing faculty to focus on strategic selection and relationship-building rather than administrative legwork.
Augmentation potentialclaude-sonnet-53/5AI can help brainstorm potential speakers, draft invitation emails, and research candidates' backgrounds, meaningfully assisting but not replacing the human-driven selection process.
Task automatabilityclaude-haiku-4-5-202510012/5Identifying potential speakers and sending invitations could be partially automated, but the task requires significant human judgment in assessing speaker fit, credibility, and student interest. Email generation and basic outreach could save time, but the core relationship-building, negotiation, and final selection remain fundamentally human-dependent.
Task automatabilityclaude-sonnet-51/5Selecting and inviting guest speakers requires relationship-building, professional judgment about fit, and personal outreach that current AI cannot meaningfully substitute for.
Adoption barriersclaude-haiku-4-5-202510014/5Educational institutions typically maintain policies requiring faculty authority over guest speaker selection due to academic freedom concerns, institutional liability, and pedagogical alignment. Human judgment on appropriateness and relevance is often a formal institutional requirement, creating adoption friction.
Adoption barriersclaude-sonnet-52/5No licensing or legal barrier exists, but institutional norms favor faculty personally curating and inviting speakers based on academic judgment and networks.
Cost vs. human wageclaude-haiku-4-5-202510012/5The cost of AI-assisted outreach (drafting, database queries) is modest, but the human overhead for vetting, relationship management, and final selection remains substantial. The all-in cost is comparable to or exceeds having a faculty member or assistant do it directly.
Cost vs. human wageclaude-sonnet-52/5While AI could cheaply draft outreach emails, the actual task—vetting, relationship management, negotiation—still requires human time, so overall cost savings are minimal.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can help draft invitation emails and identify candidate speakers from academic databases, no deployed product reliably performs the full task end-to-end. The judgment calls about relevance, speaker suitability, and institutional fit remain manual and context-dependent.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs speaker selection and invitation as an end-to-end professional task; at best AI can draft emails or suggest names as a research aid.

Conduct research in a particular field of knowledge and present findings in professional journals, books, electronic media, or at professional conferences.

16

CI 0–32 · exposure 13 · augmentation 63 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Adoption of AI for autonomous research generation is negligible; academia actively resists and scrutinizes AI authorship. The task remains firmly in human hands across all sectors of postsecondary education.
Sector adoption velocityclaude-sonnet-53/5Higher education and academia show growing but uneven AI adoption for writing and research assistance, with many institutions still cautious about AI-authored scholarship.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist with literature searches, data analysis, manuscript drafting, and citation management, raising researcher productivity on components of the task. However, the human researcher must remain fully in control of research questions, methodology, and interpretation.
Augmentation potentialclaude-sonnet-54/5AI substantially aids literature searches, summarization, drafting, editing, and even brainstorming research questions, meaningfully boosting researcher productivity while the scholar retains intellectual ownership.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires original research design, domain-specific expertise, critical judgment, and creative synthesis of findings—capabilities far beyond current AI systems. While AI can assist with literature review or data analysis components, it cannot independently conceive, execute, and present novel research findings at the required scholarly level.
Task automatabilityclaude-sonnet-52/5AI can assist with literature review, drafting, and data analysis, but original research design, novel contribution, and scholarly judgment in library science remain human-led activities that current AI cannot fully replace end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Legal and institutional barriers are very high: only credentialed humans can author research, sign research ethics approvals, and take accountability for findings. Peer review, institutional affiliation, and professional reputation are inherently human gatekeepers.
Adoption barriersclaude-sonnet-53/5Academic norms, peer review, authorship attribution, and institutional tenure/promotion requirements create moderate friction against full automation, though no strict licensing barrier exists as in some professions.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI cannot perform this task end-to-end, making cost comparison moot. The value of original research derives from human expertise and credibility, which AI cannot replicate or replace at any cost.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce some costs of drafting and lit review, but the core research process still requires substantial paid faculty time, so overall cost savings versus the human researcher are modest.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product can autonomously conduct original research, formulate novel findings, or produce publication-ready scholarly work. Current AI systems lack the embodied expertise, institutional credibility, and accountability required for peer-reviewed publication.
Technical feasibility todayclaude-sonnet-52/5Products like AI writing assistants and literature-review tools exist and are used by academics, but no deployed system independently conducts original research and publishes findings reliably at scale.

Provide professional consulting services to government or industry.

13

CI 0–25 · exposure 13 · augmentation 50 · importance 2.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Library science consulting is a specialist professional service sector with slow digital adoption; organizations continue to rely on human consultants for credibility, customization, and accountability rather than exploring AI substitution.
Sector adoption velocityclaude-sonnet-52/5Higher education and specialized academic consulting sectors adopt AI slowly relative to fast-moving information/finance sectors, with pilots more common than production use.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with literature searches, data synthesis, or report drafting during consulting work, but the core consulting value—professional judgment, client engagement, and sign-off—remains human-driven with limited AI augmentation potential.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist with literature reviews, data analysis, report drafting, and research synthesis, enhancing a consultant's productivity while the human retains final judgment.
Task automatabilityclaude-haiku-4-5-202510011/5Consulting services in library science require deep subject expertise, client relationship-building, understanding of organizational contexts, and nuanced judgment about information systems strategy—capabilities that current AI cannot reliably deliver end-to-end for professional consulting outcomes.
Task automatabilityclaude-sonnet-52/5Consulting requires contextualized judgment, relationship management, and synthesis of nuanced institutional knowledge that current AI cannot autonomously deliver end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Consulting services implicitly require a licensed or credentialed professional to sign off on recommendations, assume liability for advice, and maintain professional ethics standards; regulatory and liability frameworks strongly protect this task.
Adoption barriersclaude-sonnet-54/5Professional consulting often requires named expert credibility, accountability, and trust from clients (government/industry), creating strong barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5A library science consultant commands significant fees ($150–300+/hour) for their subject expertise and accountability; AI cannot replicate that value proposition, and the oversight required to make AI output usable by clients would exceed the cost savings.
Cost vs. human wageclaude-sonnet-52/5Human consultants combine credibility, expertise, and accountability that AI cannot yet replace, so AI substitution still requires significant human oversight, limiting cost savings.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product can independently provide credible professional consulting services to government or industry; this requires accountable expertise, legal liability, and client trust that current AI systems cannot establish or maintain.
Technical feasibility todayclaude-sonnet-52/5AI tools can support research and drafting for consulting engagements, but no deployed product independently delivers professional consulting advice in this specialized domain reliably.

Supervise undergraduate or graduate teaching, internship, and research work.

12

CI 7–16 · exposure 5 · augmentation 50 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Higher education remains cautious about automating faculty supervision roles; adoption is slow and limited to administrative aids. Institutional culture and accreditation constraints discourage deep displacement.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI tools for administrative and instructional support but supervisory/mentorship roles remain largely untouched by automation initiatives.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist with grading support, progress tracking, scheduling coordination, and literature review for research supervision, meaningfully easing administrative burden while the faculty member retains all judgment and accountability.
Augmentation potentialclaude-sonnet-53/5AI can help by tracking student progress, suggesting feedback on drafts, or organizing research materials, but the core supervisory judgment and mentorship remain human-driven.
Task automatabilityclaude-haiku-4-5-202510011/5Supervision of teaching, internship, and research work requires ongoing human judgment about student performance, mentoring, feedback, and adaptive guidance. Current AI cannot autonomously manage the relational and evaluative dimensions of academic supervision.
Task automatabilityclaude-sonnet-51/5Supervising students' teaching, internships, and research requires ongoing relational mentorship, judgment calls, and evaluation of developing professionals that current AI cannot perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510014/5Academic institutions have strong legal and accreditation requirements that supervisory faculty must personally oversee students, provide direct feedback, and take responsibility for educational outcomes. Human sign-off is mandated.
Adoption barriersclaude-sonnet-54/5Academic accreditation, mentorship norms, and institutional policies require a qualified faculty member to supervise students; this is not merely regulatory but deeply embedded in academic culture and evaluation authority.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI assistance with administrative tasks may offset some supervisory overhead, but the irreducible human labor cost of actual mentoring and evaluation remains high. AI cost-effectiveness is limited here.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this supervisory role, so any AI cost comparison is moot—human faculty remains the only functional option.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools can assist with scheduling, documentation, and basic progress tracking, but no deployed product reliably performs end-to-end supervision of academic work. The task requires real-time human oversight and accountability.
Technical feasibility todayclaude-sonnet-51/5No deployed product exists that autonomously supervises students in these capacities; this remains firmly a human academic responsibility.

Collaborate with colleagues to address teaching and research issues.

6

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Higher education and library science remain relatively slow adopters of process automation, with strong institutional resistance to replacing human collegial interaction with AI-mediated or AI-led collaboration.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI tools for specific tasks like drafting or research support, but collaborative decision-making among faculty remains untouched by AI systems in practice.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could minimally assist by drafting agendas, summarizing prior discussions, or organizing meeting notes, but the collaborative core—exchanging ideas, negotiating approaches, building consensus—remains inherently human and offers limited augmentation surface.
Augmentation potentialclaude-sonnet-53/5AI can help by summarizing research, drafting shared documents, or organizing meeting notes, providing moderate support to the underlying human collaborative process.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires human interpersonal dialogue, negotiation, and judgment about pedagogical and research strategy. Current AI cannot meaningfully participate in collaborative problem-solving with colleagues in ways that would achieve 50% time savings at equal quality.
Task automatabilityclaude-sonnet-51/5This task is inherently interpersonal and collaborative, requiring relationship-building, negotiation, and shared decision-making among faculty that AI cannot perform end-to-end.AI cannot substitute for the human collegial relationship itself.
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: academic environments typically require genuine human collegial engagement; institutional culture and professional norms expect authentic peer collaboration; liability concerns around delegating research decisions; and academic freedom principles embed human judgment.
Adoption barriersclaude-sonnet-54/5Faculty governance, tenure structures, and academic collegiality norms mean this task is deeply embedded in human institutional roles, creating strong organizational barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The overhead of integrating AI into genuine collaboration, combined with human oversight to ensure accuracy and appropriateness, exceeds the cost of direct human collaboration among colleagues.
Cost vs. human wageclaude-sonnet-51/5There is no AI product performing this function, so no meaningful cost comparison exists; the human activity is irreplaceable at any AI cost.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs collaborative problem-solving with human colleagues in professional academic settings. While AI can draft emails or summarize discussions, it cannot authentically participate in or drive collaborative resolution of teaching and research issues.
Technical feasibility todayclaude-sonnet-51/5No deployed product exists that autonomously collaborates with colleagues on institutional teaching or research matters; this remains a purely human social process.

Maintain regularly scheduled office hours to advise and assist students.

6

CI 0–11 · exposure 0 · augmentation 38 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Educational institutions operate under strong norms and policy frameworks that require faculty presence and direct student advising; adoption of automation for this task is negligible because it conflicts with core institutional structures.
Sector adoption velocityclaude-sonnet-52/5Higher education adopts AI tools for content and admin support but has been slow to formally replace direct faculty-student advising interactions.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could provide minor assistance (e.g., scheduling tools, preliminary research summaries for students to bring to hours), but the core activity—personalized advising during office hours—offers limited scope for augmentation since the human must be present and engaged.
Augmentation potentialclaude-sonnet-53/5AI can help prep materials, draft answers to common questions, or triage inquiries before office hours, but the core advising interaction remains human-led.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires synchronous human presence, real-time judgment about individual student needs, and personalized advising that depends on understanding each student's unique academic and personal context. AI cannot meaningfully replace the essential human interaction component.
Task automatabilityclaude-sonnet-51/5Holding office hours is an interpersonal, scheduled human presence task requiring relational trust and institutional accountability; AI cannot substitute for the human availability itself.
Adoption barriersclaude-haiku-4-5-202510015/5Institutional and contractual requirements typically mandate that faculty maintain office hours as part of employment; there is a legal/employment expectation that the instructor themselves provide this service, not a delegated system.
Adoption barriersclaude-sonnet-54/5Institutional policy, accreditation expectations, and student/faculty relationship norms require actual faculty availability, creating strong organizational and role-based barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task inherently requires a human instructor's time at scale (maintaining scheduled hours), so the cost of an AI system plus oversight would exceed the loaded wage of direct human provision.
Cost vs. human wageclaude-sonnet-52/5While AI chat tools are cheap, they don't fulfill the actual task of a professor holding office hours, so cost comparison for equivalent output favors the human since substitution is incomplete.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs scheduled office hours with genuine advising capacity; this task requires presence, empathy, and contextual judgment that current AI systems cannot deliver at production quality.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs 'being available as a professor during office hours'; chatbots can supplement but do not replace this scheduled human interaction role.

Act as advisers to student organizations.

4

CI 0–9 · exposure 0 · augmentation 25 · importance 3.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Educational institutions have shown minimal interest in automating faculty mentoring roles. Higher education remains human-centric, and the advising function is explicitly tied to faculty responsibility and campus culture.
Sector adoption velocityclaude-sonnet-51/5Higher education advising roles show minimal AI adoption for this specific relational, mentorship-based task.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with routine tasks like scheduling meeting spaces or compiling policy information, but the core advising work—building relationships, making judgment calls about student welfare, and providing mentorship—remains fundamentally human and resists augmentation by current systems.
Augmentation potentialclaude-sonnet-52/5AI can help with scheduling, drafting communications, or organizing event logistics, but offers limited assistance to the core advisory/mentorship function.
Task automatabilityclaude-haiku-4-5-202510011/5Advising student organizations requires nuanced judgment, interpersonal trust, and real-time responsiveness to evolving group dynamics. Current AI systems cannot build the sustained mentoring relationships or navigate the social and organizational complexity that this task demands.
Task automatabilityclaude-sonnet-51/5Advising student organizations requires ongoing relationship-building, mentorship, judgment calls, and institutional representation that current AI cannot perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and institutional barriers exist: faculty advisers are accountable for student safety and organizational governance, institutions require a responsible human to sign off on student group activities, and student welfare is a protected domain requiring licensed professionals.
Adoption barriersclaude-sonnet-54/5Institutions typically require a designated faculty/staff adviser for liability, oversight, and accreditation purposes, creating strong organizational and policy barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Even if partial automation were feasible (e.g., scheduling or procedural guidance), the overhead of integration and the irreplaceable value of faculty presence means AI would not be meaningfully cheaper than employing a qualified librarian-educator.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute providing this output, so cost comparison favors the human by default.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system reliably performs the role of faculty adviser to student organizations. This task requires institutional authority, accountability, and human presence that AI cannot supply in any production environment.
Technical feasibility todayclaude-sonnet-51/5No deployed product substitutes for a human faculty adviser role in student organizations; this remains entirely a human relational function.

Perform administrative duties, such as serving as department head.

4

CI 0–7 · exposure 0 · augmentation 38 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5No sector is automating the department head role itself; AI has not penetrated institutional leadership positions, and organizational inertia combined with legal requirements makes this highly resistant to displacement.
Sector adoption velocityclaude-sonnet-52/5Higher education administration adopts AI slowly for governance and leadership functions, though it uses AI tools for scheduling, reporting, and communications support.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with routine data analysis, scheduling, or document drafting for reports and correspondence, but the core decision-making and leadership judgment remains fundamentally human-dependent.
Augmentation potentialclaude-sonnet-53/5AI can assist with scheduling, drafting reports, summarizing meetings, and handling correspondence, meaningfully easing administrative burden even though the core leadership function remains human.
Task automatabilityclaude-haiku-4-5-202510011/5Serving as department head requires strategic decision-making, personnel management, budgeting judgment, and institutional policy-setting that depend critically on human authority, accountability, and contextual wisdom. Current AI cannot autonomously assume these fiduciary and leadership responsibilities.
Task automatabilityclaude-sonnet-51/5Department headship involves relational leadership, personnel decisions, budget negotiation, and institutional politics that current AI cannot execute end-to-end.There is no meaningful automatable core beyond drafting support.
Adoption barriersclaude-haiku-4-5-202510015/5Institutional governance, employment law, fiduciary duties, and faculty oversight requirements legally mandate human leadership with signing authority and personal accountability. Automation is formally prohibited by educational governance structures.
Adoption barriersclaude-sonnet-54/5Institutional governance structures typically require a tenured faculty member with formal authority and accountability to serve as department head, a role tied to organizational trust and legal signing authority.
Cost vs. human wageclaude-haiku-4-5-202510011/5Department head roles command substantial salaries (often $80k–$120k+ plus benefits) reflecting institutional decision-making authority and accountability; AI cannot substitute for this cost-of-labor ratio.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the role itself, so any 'cost' comparison is moot—the human must be paid regardless, making AI a supplement rather than a cheaper alternative.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product exists that can autonomously serve as a department head in an educational institution; this role requires legal authority, contract signature authority, and human accountability that cannot be delegated to AI systems.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs departmental administrative leadership; this remains squarely a human role requiring judgment, authority, and interpersonal accountability.

Participate in campus and community events.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5No adoption of AI for this task exists in any sector because the task is fundamentally interpersonal and presence-dependent, making it incompatible with automation.
Sector adoption velocityclaude-sonnet-51/5Higher education community engagement activities are low-digitization, relationship-driven functions with essentially no AI adoption trend.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could offer minimal assistance, such as helping draft talking points or creating promotional materials for events, but it cannot assist with the core task of actual participation and real-time engagement.
Augmentation potentialclaude-sonnet-52/5AI could help with scheduling, event planning, or drafting talking points, but offers minimal assistance for the actual act of participating and engaging in person.
Task automatabilityclaude-haiku-4-5-202510011/5Participating in campus and community events fundamentally requires human presence, interpersonal engagement, and real-time responsiveness to social contexts that AI cannot authentically replicate. This task has no meaningful automation pathway.
Task automatabilityclaude-sonnet-51/5Physical presence, networking, and interpersonal engagement at events cannot be performed by AI systems; this is inherently a human social/physical activity.
Adoption barriersclaude-haiku-4-5-202510015/5High barriers exist because the task requires authentic human presence and engagement; institutional reputation, community trust, and the social nature of event participation legally and practically mandate human participation.
Adoption barriersclaude-sonnet-54/5Participation requires physical presence, personal reputation, and social relationship-building tied to the individual faculty member's role, creating strong organizational and social barriers to substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5An AI system cannot perform this task at all, making cost comparison inapplicable; a human must attend and participate.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this task, so cost comparison favors the human entirely; AI cannot replace the value of physical/social participation.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can substitute for human participation in live events, which depends on physical attendance, genuine social interaction, and representing the institution or community—all inherently human functions.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product attends or participates in campus/community events on behalf of a person; this is outside current product capabilities.

Serve on academic or administrative committees that deal with institutional policies, departmental matters, and academic issues.

0

CI 0–0 · exposure 0 · augmentation 38 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Academic institutions have shown no adoption of AI substitutes for committee participation, as institutional policy and governance require human deliberation and accountability.
Sector adoption velocityclaude-sonnet-51/5Higher education governance and committee work show minimal AI adoption; this is a slow-moving, tradition-bound institutional process.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could marginally assist with preparation (summarizing policy documents, drafting agenda items) but provides limited augmentation to the core deliberative and decision-making function of committee service.
Augmentation potentialclaude-sonnet-53/5AI can help summarize meeting materials, draft policy language, or prepare briefing documents, providing moderate assistance to committee members.
Task automatabilityclaude-haiku-4-5-202510011/5Committee service requires human judgment on institutional policies, consensus-building, and nuanced deliberation on academic matters—capabilities far beyond current AI. No AI system can autonomously participate in committee decision-making or represent institutional interests.
Task automatabilityclaude-sonnet-51/5Committee service requires representing human interests, deliberation, negotiation, and institutional judgment that AI cannot perform end-to-end today.
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and organizational barriers: committee service requires human institutional representation, legal accountability, and formal authority. Governance structures mandate human participants with fiduciary responsibility.
Adoption barriersclaude-sonnet-55/5Institutional governance requires human faculty membership, voting rights, and accountability, creating hard structural and policy barriers to AI substitution.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI systems cannot perform this task at all, making cost comparison inapplicable. The task requires human institutional authority and accountability.
Cost vs. human wageclaude-sonnet-51/5There is no AI product performing this function, so no cost comparison favors AI; the human is the only viable option.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs committee service as a substitute for human participation. AI cannot legally or functionally represent an institution or make binding institutional decisions.
Technical feasibility todayclaude-sonnet-51/5No deployed product substitutes a human as a committee member; this remains entirely a human institutional role.

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.