Word Processors and Typists
43-9022.00Median wage $49,280/yr35,010 employed (US)Rank #1 of 923 scored · top 1% by substitution
Use word processor, computer, or typewriter to type letters, reports, forms, or other material from rough draft, corrected copy, or voice recording. May perform other clerical duties as assigned.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution89
Exposure89
Augmentation67
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
19 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
89%
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%89
panel mean rating 4.6/5 → substitution pressure 89/100
Technical feasibility todayw 20%88
panel mean rating 4.5/5 → substitution pressure 88/100
Cost vs. human wagew 15%91
panel mean rating 4.6/5 → substitution pressure 91/100
Adoption barriersw 20%inverted — strong barriers lower the score94
panel mean rating 1.2/5 (barrier strength) → substitution pressure 94/100
Sector adoption velocityw 10%79
panel mean rating 4.2/5 → substitution pressure 79/100
Task breakdown (19 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
File and store completed documents on computer hard drive or disk, or maintain a computer filing system to store, retrieve, update, and delete documents.
100CI 100–100 · exposure 100 · augmentation 63 · importance 4.2/5 · click for rater detail
File and store completed documents on computer hard drive or disk, or maintain a computer filing system to store, retrieve, update, and delete documents.
100| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | File management automation is nearly ubiquitous in digital-first organizations; enterprise adoption of DMS and cloud storage is deep and widespread across finance, professional services, government, and information sectors. This task has been substantially displaced already in most white-collar settings. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Office and administrative environments have near-universal adoption of automated file systems, cloud storage, and document management software already in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | While human typists are largely removed from this task in modern workflows, AI and automated systems still augment human knowledge workers by providing search, tagging, version control, and intelligent organization that improves productivity for those who interact with document systems. |
| Augmentation potential | claude-sonnet-5 | 3/5 | While largely automated already, AI can further assist by suggesting file organization, tagging, and search improvements for any remaining manual filing tasks. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | File management and document storage/retrieval is entirely automatable; modern systems already perform classification, organization, versioning, and deletion at scale with >50% time savings. Current file management, cloud synchronization, and document management systems handle this end-to-end with minimal human intervention. |
| Task automatability | claude-sonnet-5 | 5/5 | File naming, storage, and organization of digital documents is fully automatable today via scripts, cloud storage rules, and AI-assisted file management tools with equal or better consistency than manual filing. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No licensing, authorization, or regulatory requirement mandates human file management; legal liability is minimal for automated filing systems. Organizational adoption is purely voluntary and driven by cost/efficiency gains with no hard barriers to substitution. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or human-contact requirements apply to digital file organization; it's a purely administrative function with no regulatory protection. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Cloud-based document storage costs pennies per document per month; oversight is minimal and largely automated. This is orders of magnitude cheaper than hiring a human to manually file and manage documents, especially at organizational scale. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated file management costs are negligible (storage and compute) compared to paying a human hourly wage to manually file and retrieve documents. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products like SharePoint, Google Drive, Dropbox, and enterprise DMS platforms reliably perform file storage, retrieval, updating, and deletion in production at massive scale across organizations worldwide. These are mature, production-grade systems with robust capabilities. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature, widely deployed products (OS-level file systems, cloud storage with auto-organization, document management systems) already handle storage, retrieval, versioning, and deletion reliably at scale. |
Address envelopes or prepare envelope labels, using typewriter or computer.
100CI 100–100 · exposure 100 · augmentation 50 · importance 4.0/5 · click for rater detail
Address envelopes or prepare envelope labels, using typewriter or computer.
100| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Adoption is already complete across office, mail, and information-work sectors; mail merge and label automation have been standard for >20 years, and manual envelope typing is now rare in digitized organizations. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Office/administrative functions across virtually all sectors have long since adopted automated mail-merge and label printing as standard practice. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | While the task is nearly fully automated, AI can still assist humans by suggesting address corrections, flagging invalid entries, or previewing label layouts before printing, providing useful quality-control support. |
| Augmentation potential | claude-sonnet-5 | 3/5 | For any remaining manual cases, software still assists with formatting and address validation, though the task is largely already automated rather than augmented. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | This task is fully automatable end-to-end with >50% time savings: mail merge tools, label-printing software, and document automation systems can extract addresses from databases and produce labels/envelope text with minimal human input, eliminating manual typing and formatting work. |
| Task automatability | claude-sonnet-5 | 5/5 | Addressing envelopes or generating labels from data is a fully structured, repetitive task easily handled by mail-merge and label-printing software with no quality loss. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No legal, regulatory, or licensing barriers exist; envelope addressing is a purely clerical function with no human-signature or human-contact requirement, and organizations face no organizational friction to adopting automation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or human-contact requirements attach to this clerical task; nothing legally or organizationally blocks automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI/automation cost is negligible—a few cents per envelope via templates and batch printing—compared to human labor cost ($15–30/hour for typing and addressing), making automation orders of magnitude cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Software-based label/envelope generation costs a tiny fraction of a cent per item versus manual typing labor, an order-of-magnitude or greater saving. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature, widely deployed products (mail merge in Office, dedicated label software like Avery Design Pro, enterprise postal systems) reliably perform this task at scale in production across businesses, nonprofits, and government agencies today. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature, widely deployed tools (Word mail merge, label printing software, shipping platforms) have performed this reliably in production for decades. |
Search for specific sets of stored, typed characters to make changes.
100CI 100–100 · exposure 100 · augmentation 88 · importance 3.8/5 · click for rater detail
Search for specific sets of stored, typed characters to make changes.
100| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Search and replace functionality is ubiquitous in information work and has been automated for decades across all digital document contexts. Adoption is essentially complete wherever digital text exists. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | This capability has been universally adopted for decades across all sectors using digital text editing, representing complete and mature deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI-powered search tools (semantic search, regex patterns, contextual suggestions) substantially augment human typists and document managers by enabling faster, more accurate, and more intelligent text manipulation while the human reviews and approves changes. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enhanced search (semantic, fuzzy matching) further improves on basic find-replace, helping users locate and edit content faster, though the core function was already highly optimized pre-AI. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Finding and replacing specific character strings is a core text-processing function that AI and standard automation tools (find-and-replace) can perform reliably and at massive time savings. This task meets the ≥50% time-saving bar trivially with off-the-shelf systems. |
| Task automatability | claude-sonnet-5 | 5/5 | Find-and-replace and text search across documents is a fully solved, deterministic text-processing operation handled natively by word processors and AI tools with perfect reliability and massive time savings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No legal, regulatory, or organizational barriers prevent automation of text search. This is purely a technical task with no licensing, liability, or human-contact requirements. It is already widely automated. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or human-contact requirements apply to searching and editing text strings within a document. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The computational cost of string searching is negligible compared to the loaded wage of a human word processor. AI-based or algorithmic search is orders of magnitude cheaper than paying a human to manually locate and modify text. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | This is a near-zero marginal cost operation built into standard software; there is no meaningful per-use cost compared to manual searching by a human typist. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products including all modern word processors, text editors, and AI agents reliably perform search and find-replace operations at scale in production environments every day. This is among the most mature and stable automated capabilities. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Search/find-replace functionality has been a mature, production-grade feature in word processing software for decades, used reliably at scale by billions of users. |
Reformat documents, moving paragraphs or columns.
100CI 100–100 · exposure 100 · augmentation 88 · importance 3.7/5 · click for rater detail
Reformat documents, moving paragraphs or columns.
100| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Document automation and reformatting are already deeply embedded in information work. Organizations routinely use templates, mail merge, document management systems, and workflow automation; manual reformatting is increasingly rare in digitized workplaces. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Office and administrative work has seen fast, deep adoption of automated formatting tools and AI writing assistants, consistent with high-digitization white-collar sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI assists by offering real-time suggestions for document structure, automatic column alignment, and intelligent paragraph reordering; these augmentations substantially raise human productivity even when the human retains final control over formatting decisions. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI and software tools significantly speed up reformatting tasks for remaining human operators, though the task is largely fully automatable rather than merely augmented. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Reformatting documents by moving paragraphs or columns is a straightforward structural manipulation task that current AI and automation tools handle reliably end-to-end. Document processing systems, word processors with macro/API support, and LLM-based agents can parse, reorder, and output reformatted documents with >50% time savings compared to manual rearrangement. |
| Task automatability | claude-sonnet-5 | 5/5 | Reformatting text, moving paragraphs or columns is a well-defined text manipulation task that current word processing software and AI tools can fully automate with equal or better quality and drastic time savings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | There are no legal, licensing, or regulatory barriers to automating document reformatting. No human sign-off or contact is required, and organizational adoption is straightforward with no liability concerns unique to this task. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirements exist for document reformatting; it's a purely administrative task with no legal barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of automation (API calls to document processing services, word processor macros, or simple scripts) is orders of magnitude cheaper than the loaded wage of a typist or word processor operator performing manual reformatting. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated reformatting via software/AI costs fractions of a cent per document versus paying a human typist by the hour, an order-of-magnitude cost difference. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | This capability is mature and widely deployed in production. Word processors (Word, Google Docs) have built-in formatting and rearrangement tools, and document automation platforms reliably perform paragraph/column reordering at scale with minimal errors. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Modern word processors (Word, Google Docs) with built-in formatting tools and AI copilots reliably perform reformatting tasks in production today for millions of users. |
Transmit work electronically to other locations.
99CI 97–100 · exposure 100 · augmentation 50 · importance 4.1/5 · click for rater detail
Transmit work electronically to other locations.
99| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Information-sector organizations (corporate offices, professional services, publishing) have already broadly adopted automated document workflows, email integration, and file-sharing systems that eliminate manual transmission tasks. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Electronic transmission of documents has been fully adopted across virtually all office and administrative settings for decades. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems already assist by auto-organizing, routing, and batch-transmitting documents; a human supervising these workflows experiences substantial productivity gains. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Since transmission is already automated via basic software, there is little additional productivity AI adds specifically to this sub-task beyond existing automation. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | This task is entirely routine and deterministic: receiving completed work and sending it via email, file-sharing platforms, or network systems. Current AI agents can fully automate this workflow with significant time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 5/5 | Electronic transmission of files (email, file sharing, cloud upload) is already fully automatable and typically automated via standard software with no AI needed, easily exceeding the 50% time-saving bar. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No licensing, regulatory, or legal requirement mandates that a human must transmit work electronically; the task involves no legal sign-off or human contact requirement. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or human-contact requirements restrict electronic file transmission; it is routine automated infrastructure. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated transmission via APIs or scheduled workflows costs pennies per execution, orders of magnitude cheaper than paying human labor to manually send files. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Electronic transmission costs fractions of a cent per transfer versus any human labor cost, making it an order of magnitude or more cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Email clients, file transfer systems, and document management platforms with API integration are mature, widely deployed, and reliably perform this task at scale in organizations worldwide. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature, ubiquitous products (email clients, cloud storage, network file transfer) perform this reliably at scale in production today. |
Use data entry devices, such as optical scanners, to input data into computers for revision or editing.
99CI 97–100 · exposure 100 · augmentation 63 · importance 3.6/5 · click for rater detail
Use data entry devices, such as optical scanners, to input data into computers for revision or editing.
99| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Optical scanning automation has been widely adopted across digitized sectors (finance, healthcare, insurance, government) for years; this is not an emerging technology but an established practice. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Office/administrative and clerical work has seen fast, broad adoption of digitization and OCR technology over the past two decades. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | While the core task is fully automatable, AI can assist humans in reviewing and correcting scanned output, flagging confidence scores, and handling edge cases requiring human judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-driven scanning and OCR significantly boosts a typist's throughput and accuracy when combined with human review for edge cases. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Optical scanning and data input is a fully automatable workflow with current OCR and document processing systems; tools like Tesseract, AWS Textract, and similar platforms achieve >50% time savings at equal or superior quality compared to manual entry. |
| Task automatability | claude-sonnet-5 | 5/5 | OCR/scanning-to-digital-text pipelines are mature and fully automatable, and modern AI-enhanced OCR combined with automated formatting exceeds the 50% time-saving threshold easily. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | There are no licensing, legal, or regulatory barriers preventing automation of optical scanning and data input; it is purely a technical substitution with no human sign-off requirement. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory barriers restrict use of optical scanning and automated data entry for this task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated optical scanning costs (hardware + software licenses + infrastructure) are orders of magnitude cheaper than paying loaded wages for manual data entry, especially at volume. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated scanning and OCR software costs pennies per page compared to the loaded wage of a human typist manually keying in data. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature OCR and document scanning products are deployed at scale in production across finance, healthcare, and logistics; systems like Tesseract, Adobe Acrobat, and specialized scanning platforms reliably perform this task with high accuracy. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Production-grade OCR and document digitization tools (e.g., Adobe, ABBYY, Google Document AI) are widely deployed and reliably used at scale in enterprises today. |
Adjust settings for format, page layout, line spacing, and other style requirements.
99CI 97–100 · exposure 100 · augmentation 88 · importance 3.6/5 · click for rater detail
Adjust settings for format, page layout, line spacing, and other style requirements.
99| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Digital document tools with built-in formatting automation are ubiquitous in offices, with templates and macros widely deployed; adoption is fast and deep in white-collar sectors. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Office/administrative software with built-in formatting automation and AI features is already ubiquitously adopted across virtually all sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI-assisted formatting suggestions, style templates, and one-click layout adjustments significantly boost human productivity and reduce manual adjustment time. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI and software tools substantially speed up formatting tasks for remaining human typists/editors, though some manual oversight or customization is still common. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Current AI and automation tools can fully configure document formatting, layout, and styling via APIs or UI automation with negligible human oversight, easily exceeding 50% time savings at equivalent quality. |
| Task automatability | claude-sonnet-5 | 5/5 | Modern word processing software and AI-assisted tools can automatically apply formatting, page layout, line spacing, and style templates with minimal human input, fully meeting the time-saving threshold.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No licensing, regulatory, or legal barriers exist; formatting is a pure technical task with no human authorization requirement or liability asymmetry. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirements attach to document formatting; it's a purely administrative task with no legal barrier to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated formatting via templates, macros, or agent systems costs pennies per document versus minutes of human labor, yielding at least an order-of-magnitude cost advantage. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated formatting via templates/macros/AI costs a fraction of a cent versus paying a human typist's time for repetitive formatting adjustments. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature products (MS Word, Google Docs, LaTeX automation, RPA tools) reliably handle formatting adjustments in production environments at scale with standard APIs and macro systems. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature, widely deployed products (Word, Google Docs, styles/templates, macros, AI formatting assistants) reliably perform this in production at massive scale today. |
Check completed work for spelling, grammar, punctuation, and format.
97CI 95–100 · exposure 100 · augmentation 100 · importance 4.3/5 · click for rater detail
Check completed work for spelling, grammar, punctuation, and format.
97| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Grammar and spell-checking AI is already ubiquitous in mainstream office software, email platforms, and publishing workflows; adoption is nearly universal across digital-first sectors with little remaining friction. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Spelling/grammar checking software is deeply embedded in nearly all office software and word processing workflows already, representing near-universal adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI proofreading tools dramatically augment human editors and word processors by flagging issues in real-time, suggesting corrections, and enforcing style consistency, enabling humans to focus on higher-level quality control and judgment. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-based proofreading tools substantially boost human editing speed and catch rate while the human retains final judgment over style and meaning. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | AI language models and specialized grammar-checking tools (Grammarly, LanguageTool) routinely identify and correct spelling, grammar, punctuation, and format errors at scale with high accuracy, easily meeting the 50% time-saving threshold by automating the entire task end-to-end. |
| Task automatability | claude-sonnet-5 | 5/5 | Grammar/spelling/format checking is a well-solved NLP task; tools like Grammarly, Word's editor, and LLMs can proofread text end-to-end faster than a human with equal or better quality on most documents. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some organizations may prefer human review for high-stakes documents and style guides can vary, there are no legal or licensing requirements mandating human word processors perform this task, creating minimal structural barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or human-contact requirement blocks automated proofreading; it's already ubiquitous in consumer and enterprise software. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated grammar and spell-checking via APIs or off-the-shelf software costs pennies per document or is bundled into office suites, making it an order of magnitude cheaper than paying a human word processor's loaded wage for the same output. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated grammar/spellcheck costs pennies per document versus a typist's hourly wage, an order-of-magnitude or greater cost advantage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature, deployed products (Grammarly, MS Word's built-in editor, professional proofreading APIs) reliably perform spell-check, grammar validation, and format analysis in production at enterprise scale with minimal error rates. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Deployed products (Grammarly, MS Editor, Google Docs suggestions, ChatGPT) are used in production at massive scale for exactly this proofreading function today. |
Electronically sort and compile text and numerical data, retrieving, updating, and merging documents as required.
97CI 97–97 · exposure 100 · augmentation 75 · importance 3.8/5 · click for rater detail
Electronically sort and compile text and numerical data, retrieving, updating, and merging documents as required.
97| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Information and administrative sectors have rapidly adopted RPA and document management automation over the past 5–10 years; this is a classic early-automation target with strong production deployment momentum in digitized organizations. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Office/administrative software with built-in automation and AI features has been broadly adopted for years, though smaller and legacy-reliant offices still lag. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI augments this task by accelerating retrieval and offering intelligent merge suggestions or duplicate detection, allowing a human to oversee and refine outputs faster than manual assembly; productivity gains are substantial even with human oversight. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Even where full automation isn't deployed, AI tools substantially speed up sorting, retrieving, and merging documents for human operators. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | This task involves routine data manipulation and document assembly—core competencies of current automation tools. AI agents with access to file systems and document processing APIs can retrieve, sort, merge, and update text/numerical data end-to-end with >50% time savings and equal or superior quality. |
| Task automatability | claude-sonnet-5 | 5/5 | Sorting, compiling, retrieving, updating and merging documents is a structured data-manipulation task well within the capability of current software and AI tools operating end-to-end with major time savings.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No licensing requirement, liability, or regulatory mandate requires a human to perform or sign off on document compilation and sorting. Organizational adoption is purely voluntary and cost-driven. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, legal, or human-judgment requirement blocks automating this clerical data-handling task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Inference and integration costs for document sorting, merging, and retrieval are minimal compared to human wage costs; cloud-based or on-premise automation handles bulk operations at near-zero marginal cost per instance. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated document processing and merging costs a tiny fraction of a cent per operation compared to a human typist's hourly wage for the same volume of work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Multiple mature products (RPA platforms, document automation tools, Python-based agents with file I/O) perform these operations reliably in production at scale across organizations. Standard APIs and libraries make this task straightforwardly deployable. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature products (mail-merge, database query tools, RPA, AI copilots in Office/Google Suite) already perform this reliably in production across countless organizations. |
Collate pages of reports and other documents.
97CI 97–97 · exposure 100 · augmentation 38 · importance 3.7/5 · click for rater detail
Collate pages of reports and other documents.
97| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Document management and workflow automation have been widely deployed in information and professional services sectors for over a decade; collation is among the earliest and most common targets for such automation. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Office and administrative functions in most industries have already widely adopted digital document tools that automate collation, though some physical paper workflows lag. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While collation can be partially assisted by sorting previews or batch operations, the task itself offers limited scope for meaningful human-AI collaboration once automation is feasible; the augmentation value is modest. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Where physical collation still occurs, software can assist by pre-sorting or generating digital equivalents, but the assistance is narrow since the task itself is nearly fully automatable already. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Collating pages—ordering and stacking physical or digital documents—is a routine, sequential task that modern automation tools (document management systems, RPA, even simple scripts) can perform end-to-end with negligible error and dramatic time savings, easily exceeding the 50% threshold. |
| Task automatability | claude-sonnet-5 | 5/5 | Collating pages is a mechanical, rule-based document assembly task that can be fully handled by software (PDF tools, printer/scanner collation, mail-merge, document assembly scripts) with no quality loss and massive time savings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | There are no licensing, regulatory, liability, or human-contact requirements for automated collation; technical and organizational setup is the only friction, presenting no legal or compliance barriers to substitution. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing, liability, or human-contact requirement tied to physically or digitally collating pages; it's purely administrative and unregulated. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated collation via software infrastructure costs pennies per task after setup, vastly cheaper than paying a worker to manually sort and stack pages, yielding an order-of-magnitude cost advantage. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Software-based collation costs fractions of a cent per document versus paying a human typist's time, an order-of-magnitude or greater cost advantage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Multiple mature products (document management platforms, workflow automation tools, print management software) reliably collate documents in production across organizations; this is a standard feature of enterprise systems with no material reliability concerns. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature deployed products (Adobe Acrobat, printer collation features, office suites, document management systems) reliably perform automated collation at scale today. |
Type correspondence, reports, text and other written material from rough drafts, corrected copies, voice recordings, dictation, or previous versions, using a computer, word processor, or typewriter.
96CI 92–100 · exposure 100 · augmentation 88 · importance 4.0/5 · click for rater detail
Type correspondence, reports, text and other written material from rough drafts, corrected copies, voice recordings, dictation, or previous versions, using a computer, word processor, or typewriter.
96| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Adoption is already fairly rapid in information-intensive sectors (legal, medical, business services). Dictation and speech-to-text are embedded in mainstream platforms (Microsoft 365, Google Workspace), and companies are actively deploying automated transcription for internal documentation and customer support. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Administrative/clerical text-processing functions are among the fastest and most deeply automated tasks in office/information sectors, with widespread production use of dictation and AI drafting tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI provides transformative assistance to typists and knowledge workers: real-time transcription, auto-formatting, grammar/style suggestions, and automated draft cleaning all meaningfully accelerate output while the human remains in control of content quality and corrections. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Even where humans remain involved (e.g., reviewing dictation output or polishing AI-drafted text), AI substantially speeds up the transcription and drafting process. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Speech-to-text (transcription) and automated typing from voice recordings are now highly accurate with off-the-shelf systems (Whisper, professional transcription services). Format correction from drafts and previous versions can be handled by AI with minimal human oversight, easily clearing the 50% time-saving threshold for the full task. |
| Task automatability | claude-sonnet-5 | 5/5 | Transcription, dictation-to-text, and drafting-from-rough-copy are core capabilities of current speech-to-text and LLM tools, meeting or exceeding the 50% time-saving threshold with equal or better quality. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are minimal legal or regulatory barriers to automating typing tasks; no professional license is required. The main friction is organizational inertia and user comfort with dictation or AI transcription, but no hard institutional requirement for a human to perform the task. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human typist; organizations can freely substitute software with minimal friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Cloud transcription costs $0.01–0.10 per minute; a word processor/typist loaded wage is $25–40/hour. For dictated material, AI transcription is orders of magnitude cheaper per output unit, even accounting for quality review and correction overhead. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated transcription and text generation cost fractions of a cent to a few cents per page versus a loaded typist wage, an order-of-magnitude or greater cost advantage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Multiple mature products perform this at scale: cloud-based transcription APIs (Google Cloud Speech-to-Text, Azure Speech Services), specialized transcription platforms (Rev, Otter.ai), and integrated word processors with dictation features all reliably handle voice-to-text and draft formatting in production environments. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature, widely deployed products (e.g., dictation software, transcription services, AI writing assistants integrated into word processors) already perform this reliably at scale in production. |
Compute and verify totals on report forms, requisitions, or bills, using adding machine or calculator.
96CI 92–100 · exposure 100 · augmentation 75 · importance 4.0/5 · click for rater detail
Compute and verify totals on report forms, requisitions, or bills, using adding machine or calculator.
96| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Accounting and finance sectors have adopted invoice automation and bill verification systems rapidly over the past 5–10 years. Financial automation is a high-priority category with documented deep deployment in medium to large firms. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Office and clerical functions have long since adopted calculators, spreadsheets, and automated accounting software as standard practice, representing mature, deep adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered calculators and verification tools significantly assist human operators by flagging discrepancies, auto-populating forms, and catching errors in real time, substantially raising productivity even when a human reviews the final output. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Even where humans still verify totals, calculators and spreadsheet tools substantially speed up and reduce errors in this task. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | This task is entirely computational and rule-based: extracting numbers, summing them, and verifying totals. Current OCR + calculator APIs can perform end-to-end automation with near-perfect accuracy and near-zero human time, far exceeding the 50% threshold. |
| Task automatability | claude-sonnet-5 | 5/5 | Simple arithmetic verification on structured forms is fully automatable with spreadsheet formulas, OCR-plus-calculation pipelines, or basic scripts, meeting the 50% time-saving bar easily. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some organizations may require human sign-off for financial accuracy or audit compliance, the computational task itself has no legal licensing requirement. Organizational inertia and audit preferences create modest friction but no hard legal barrier. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates human computation of totals; this is a purely clerical arithmetic function. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of OCR, arithmetic verification, and automated reporting via cloud APIs is negligible per invoice or form compared to human wages for manual computation and verification. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated calculation via software costs fractions of a cent per transaction versus paying a human wage to manually re-total figures. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Multiple deployed products (accounting software, invoice automation platforms, RPA tools) perform this task reliably at scale in production environments. Document-to-calculation pipelines are mature and widely used. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Spreadsheet software, accounting systems, and OCR-based data extraction tools already perform this reliably in production across virtually all office settings. |
Print and make copies of work.
96CI 91–100 · exposure 92 · augmentation 38 · importance 4.2/5 · click for rater detail
Print and make copies of work.
96| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Adoption is ubiquitous and complete: automated printing and copying have been standard in office environments for decades, with near-total penetration in digitized sectors; human manual printing and copying is now exceptional rather than routine. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Automated printing and copying via office equipment and software has been fully adopted for decades across nearly all sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While humans can still be assisted by print-workflow software (job scheduling, status tracking), the core task of printing and copying offers minimal augmentation value since the task itself is largely mechanistic and already highly automated. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-driven document management tools can streamline formatting and print job batching, but the core mechanical action offers limited room for further augmentation beyond existing automation. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Printing and copying are fully automatable end-to-end tasks; modern office systems and cloud services handle document distribution, batch printing, and copy workflows with zero human intervention, achieving >50% time savings and equal quality. |
| Task automatability | claude-sonnet-5 | 4/5 | Printing and copying is a simple mechanical/software operation already handled by automated print drivers, network print queues, and office equipment with minimal human involvement beyond initiating the job.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No regulatory, licensing, or liability barriers exist; printing and copying require no human authorization or sign-off, and organizations face no friction substituting automated systems for manual print/copy tasks. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, regulatory, or liability barriers exist for automating printing and copying; it is a purely administrative mechanical task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The per-copy cost via automated systems (hardware amortization + toner/paper) is orders of magnitude cheaper than the loaded hourly wage of a human clerk managing print jobs and copies manually. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated print/copy hardware and software cost pennies per job compared to paying a human wage to manually manage printing and copying tasks. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products reliably perform this at scale today: multifunction printers, document management systems, and cloud platforms (Google Workspace, Microsoft 365) integrate printing and copying as core features in production environments across millions of organizations. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Automated printing, copying, and document management systems are mature, ubiquitous, and reliably deployed in virtually all office environments today. |
Keep records of work performed.
95CI 92–97 · exposure 100 · augmentation 63 · importance 3.9/5 · click for rater detail
Keep records of work performed.
95| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Record automation is deeply embedded in digitized work environments (offices, professional services, tech firms), with widespread adoption of automated time-tracking and logging systems already commonplace. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Administrative and clerical support functions have seen fast adoption of automated tracking and productivity software, though smaller offices may lag. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Even where humans review records, AI-assisted systems that suggest categorization, detect anomalies, or auto-populate logs significantly enhance human productivity in maintaining and validating work records. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can auto-populate or summarize records, saving time, though the underlying task is largely already automatable rather than needing human augmentation. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Record-keeping of work performed is highly automatable through time-tracking software, logging systems, and event capture that can document tasks, hours, and outputs with minimal human intervention, easily meeting the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 5/5 | Logging work performed (time, document counts, job metadata) is a structured data-entry task that off-the-shelf automation, macros, or AI-integrated software can fully handle with equal or better accuracy. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some organizations prefer human oversight of records for compliance and accuracy verification, there are no licensing requirements or legal mandates that a human must perform the record-keeping itself, allowing straightforward automation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing, liability, or regulatory requirement mandating a human perform routine work-record logging. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated record-keeping via software costs orders of magnitude less than paying a human to manually log and maintain work records, especially when amortized across users. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated logging via software integrations costs a fraction of a cent per record versus the labor cost of manual record-keeping by a typist. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products like time-tracking software (Toggl, Harvest), automated logging systems, and workflow management platforms reliably perform this task at scale in production environments across many organizations. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Time-tracking and work-logging software (often with AI features) is already deployed at scale across office environments, reliably capturing and recording work activity. |
Manage schedules and set dates, times, and locations for meetings and appointments.
86CI 75–97 · exposure 87 · augmentation 88 · importance 3.9/5 · click for rater detail
Manage schedules and set dates, times, and locations for meetings and appointments.
86| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Adoption of AI scheduling tools is rapid and widespread in professional services, corporate offices, and information-sector organizations. Smart calendar assistants are now standard features in mainstream productivity suites. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Administrative and office support functions in professional services and information sectors have seen fast adoption of AI scheduling tools embedded in widely used calendar and email platforms. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI significantly augments human productivity in this domain: it drafts proposals, surfaces conflicts, automates routine bookings, and lets humans focus on complex or sensitive scheduling decisions, directly multiplying output. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI scheduling assistants substantially reduce back-and-forth communication and cognitive load for humans managing calendars, while humans retain oversight for judgment calls and exceptions. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Calendar management is highly automatable: AI can parse meeting requests, check availability, propose times, and update shared calendars with minimal human intervention. Current systems (calendar assistants, AI scheduling tools) routinely save >50% of the time spent on this coordination task. |
| Task automatability | claude-sonnet-5 | 4/5 | Scheduling meetings involves finding mutual availability and coordinating logistics, which AI calendar assistants can handle end-to-end for most straightforward cases with significant time savings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While there are minor organizational friction points (user preference, integration setup, oversight of unusual requests), there are no legal or licensing barriers preventing automation. A human does not legally need to sign off on scheduling decisions. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There are no licensing, regulatory, or liability barriers to using software for scheduling meetings; it is a purely administrative function with no legal requirement for human performance. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated scheduling costs pennies per meeting (API calls + inference) versus the loaded hourly wage of a word processor/typist managing schedules manually, making AI at least 10–100× cheaper per task. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated scheduling tools cost a small fraction of a human assistant's hourly wage for the equivalent volume of scheduling tasks, though some oversight and exception handling costs remain. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products like Google Calendar's smart scheduling, Calendly, and AI meeting assistants already perform this task reliably in production across thousands of organizations. Integration with email and meeting platforms is mature and widely in use. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Products like Microsoft Copilot, Google Calendar AI features, and dedicated scheduling assistants (e.g., Clara, x.ai successors) are deployed in production and handle routine scheduling reliably, though edge cases (VIP preferences, complex multi-party negotiations) still need human intervention. |
Gather, register, and arrange the material to be typed, following instructions.
78CI 72–84 · exposure 75 · augmentation 75 · importance 4.0/5 · click for rater detail
Gather, register, and arrange the material to be typed, following instructions.
78| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Document management and RPA adoption is widespread in information-intensive sectors (finance, legal, professional services) where word processors work, reflecting fast, deep automation already in production. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Office/clerical functions are adopting AI-based document processing tools steadily, but full deployment across all word processing environments remains uneven, particularly in smaller organizations still using manual methods. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by suggesting organization schemes, flagging ambiguous instructions, and auto-populating metadata fields, meaningfully raising human productivity even when a human remains responsible for final arrangement decisions. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI tools significantly speed up gathering, tagging, and organizing material for a human typist, who can then focus on formatting, editing, and quality checks. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can reliably gather documents, extract metadata, and arrange materials for typing with minimal human intervention, though some edge cases (ambiguous instructions, unclear document hierarchy) may require review. This task is highly procedural and structured, meeting the 50% time-saving threshold in most common scenarios. |
| Task automatability | claude-sonnet-5 | 4/5 | AI systems can ingest, sort, and organize source material according to instructions with high reliability, especially with document-processing agents and OCR/NLP tools integrated into workflows.deviation lies in messy physical materials requiring manual handling. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal or regulatory barriers exist; the task is straightforward data handling with minimal liability risk. Organizational friction and preference for human verification provide modest friction, but nothing prevents automation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human perform this organizational/preparatory task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of automated document gathering, registration, and arrangement via cloud APIs and RPA tools is orders of magnitude cheaper than human labor for this repetitive, rule-based task. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated document intake and organization software costs far less per unit of throughput than a human typist performing the same organizational task, especially at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Document management systems, optical character recognition, and workflow automation tools already exist in production and can automatically collect, register, and organize materials with high accuracy. However, edge cases with non-standard instructions or ambiguous hierarchies still require occasional human oversight. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Deployed products (document management systems, AI-based intake and sorting tools, e-discovery software) already perform gathering and arranging of digital material reliably, though physical document handling still needs human intervention. |
Work with technical material, preparing statistical reports, planning and typing statistical tables, and combining and rearranging material from different sources.
76CI 67–84 · exposure 70 · augmentation 88 · importance 3.7/5 · click for rater detail
Work with technical material, preparing statistical reports, planning and typing statistical tables, and combining and rearranging material from different sources.
76| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Information-sector organizations (publishing, finance, analytics firms) have rapidly adopted document automation and LLM-based report generation; adoption is measurable and accelerating in digitized office environments. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Clerical/administrative functions in offices are adopting AI drafting and data tools at a moderate pace, with pilots and partial deployment common but not yet universal replacement. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI augmentation is strong: humans can prompt systems to draft tables and reorganize sources, then review and refine, significantly accelerating turnaround on statistical reporting tasks while maintaining quality control. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI substantially speeds up combining, reformatting, and structuring statistical material from multiple sources, letting the human focus on verification and final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems can handle substantial portions of this task: extracting data from sources, generating statistical tables, formatting reports, and reorganizing material. LLMs and document-generation tools can produce statistical reports with minimal human input, though complex statistical interpretation and quality assurance typically require oversight, preventing a full 5. |
| Task automatability | claude-sonnet-5 | 4/5 | AI tools can extract, combine, and format data from multiple sources into statistical tables and reports with substantial time savings, though complex technical formatting may need human review.LLMs and spreadsheet automation handle most of this workflow today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Minimal legal or regulatory barriers exist; the task does not require human licensure or sign-off, and organizations face only modest friction from quality-control preferences and integration into existing workflows. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or regulatory requirement mandates a human typist for this clerical task; organizational adoption friction is minimal. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven report generation and table automation cost pennies per document after setup, compared to loaded human wages of $25–40/hour for word processors. The cost advantage is several orders of magnitude. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | AI-assisted drafting and table generation is dramatically cheaper per unit output than manual typing and formatting, though some oversight cost remains for accuracy checks on technical content. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature products (e.g., document automation platforms, Python/R scripting tools, and LLM-based report generators) reliably perform statistical table creation and material reorganization in production. Some error rates exist in complex formatting or cross-source integration, but the core task is demonstrably deployed at scale. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like Excel Copilot, ChatGPT with code interpreter, and document automation tools perform data aggregation and table generation, but accuracy on technical/statistical material still requires verification, limiting fully autonomous production use. |
Perform other clerical duties, such as answering telephone, sorting and distributing mail, running errands or sending faxes.
56CI 47–64 · exposure 45 · augmentation 50 · importance 4.4/5 · click for rater detail
Perform other clerical duties, such as answering telephone, sorting and distributing mail, running errands or sending faxes.
56| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Word processor and typist roles are in declining sectors with slower digital maturity; while call forwarding and basic mail automation exist, widespread agent-based automation of the full task bundle remains sparse and slow. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Office/administrative sectors are adopting AI voice assistants and automation tools at a moderate pace, though many small offices still rely on humans for the full multi-task clerical bundle. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by drafting responses to common inquiries, flagging priority mail, and logging call details, meaningfully boosting human productivity on routine clerical work while the person remains in control of phone interactions and judgment calls. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist with message triage, call transcription, and reminders for errands, improving efficiency, but doesn't replace the physical components of the job. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Portions of this task—sorting/distributing mail, sending faxes, logging calls—can be automated with current systems, but answering phones meaningfully and handling ad-hoc errands require human judgment and context. Overall time savings likely reach 40–50%, placing it near the threshold but not solidly above 50%. |
| Task automatability | claude-sonnet-5 | 3/5 | Phone answering and message handling can be automated with AI voice agents, but mail sorting/distribution and running errands are physical tasks requiring embodiment that AI cannot perform.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal barriers exist for automating mail, faxes, or basic phone routing, but customer expectations, privacy/liability concerns (especially around call recording and message handling), and organizational habit create moderate friction to adoption. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory barriers prevent using AI for call answering or clerical coordination; adoption is a matter of convenience and workflow, not policy. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Sending faxes, sorting mail, and basic call logging via AI agents cost far less than human labor; even with oversight, the all-in cost per unit is a small fraction of a typist's loaded wage for equivalent output. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | For the phone-answering portion, AI call handling is much cheaper than a human, but since the task bundles physical duties requiring a human anyway, overall cost savings are muted. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While email automation and basic call routing exist in production, reliable end-to-end automation of the full task suite (especially intelligent phone handling and contextual errands) remains limited to narrow use cases and demos rather than mature, widely deployed systems. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | AI phone/receptionist products and virtual assistants exist and work reasonably well for call handling, but physical tasks (sorting mail, errands, faxing) have no viable AI product substitute. |
Operate and resupply printers and computers, changing print wheels or fluid cartridges, adding paper, and loading blank tapes, cards, or disks into equipment.
22CI 15–29 · exposure 8 · augmentation 0 · importance 3.4/5 · click for rater detail
Operate and resupply printers and computers, changing print wheels or fluid cartridges, adding paper, and loading blank tapes, cards, or disks into equipment.
22| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Hardware maintenance tasks like printer resupply occur in traditional office settings with low automation investment and minimal digitization of the maintenance process itself. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Physical equipment servicing tasks show essentially no AI adoption trend since robotics for this narrow task is not commercially deployed. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers no meaningful assistance for physical resupply and maintenance tasks; there is no knowledge work component that could be augmented by language models or decision-support systems. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance to physically resupplying printers or loading tapes/disks into equipment. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task involves physical manipulation of hardware components (changing print wheels, loading paper, handling cartridges) that requires dexterity, spatial reasoning, and real-world interaction. Current AI systems cannot perform physical operations in unstructured office environments. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a physical maintenance task (swapping cartridges, loading paper/media) that requires manual dexterity and physical presence; AI software cannot perform this end-to-end without robotic embodiment.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | There are no legal, regulatory, or licensing barriers to automation of printer and equipment maintenance; the barrier is purely technical (lack of viable robotic systems for this niche application). |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or regulatory barrier exists, but the physical nature of the task itself is the barrier to AI substitution rather than organizational or legal friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of a robotic system capable of reliable hardware manipulation and resupply far exceeds the cost of a human performing these routine maintenance tasks, which typically take minutes to complete. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based alternative to compare cost against for this purely physical task; a human doing it directly remains the only practical option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can autonomously perform the physical operations described—changing cartridges, loading paper, or handling hardware components. This remains a task requiring physical embodiment and manipulation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product replaces the physical act of resupplying printers or loading media; this remains a manual human task in virtually all offices. |
Related occupations — Office & Administrative Support
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.