Poets, Lyricists and Creative Writers
27-3043.05Median wage $76,910/yr47,940 employed (US)Rank #141 of 923 scored · top 15% by substitution
Create original written works, such as scripts, essays, prose, poetry or song lyrics, for publication or performance.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution42
Exposure33
Augmentation72
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
16 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
6%
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%34
panel mean rating 2.3/5 → substitution pressure 34/100
Technical feasibility todayw 20%31
panel mean rating 2.2/5 → substitution pressure 31/100
Cost vs. human wagew 15%55
panel mean rating 3.2/5 → substitution pressure 55/100
Adoption barriersw 20%inverted — strong barriers lower the score61
panel mean rating 2.6/5 (barrier strength) → substitution pressure 61/100
Sector adoption velocityw 10%35
panel mean rating 2.4/5 → substitution pressure 35/100
Task breakdown (16 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.
Conduct research to obtain factual information and authentic detail, using sources such as newspaper accounts, diaries, and interviews.
74CI 55–92 · exposure 70 · augmentation 88 · importance 4.3/5 · click for rater detail
Conduct research to obtain factual information and authentic detail, using sources such as newspaper accounts, diaries, and interviews.
74| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Publishing, media, and content creation sectors are rapidly integrating AI research tools, fact-checking systems, and automated document retrieval in production workflows. Adoption is measurable and accelerating in competitive, digitized industries. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Creative writing and publishing sectors show growing but uneven adoption of AI research tools; use is common among some writers but not yet standard practice industry-wide. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI dramatically augments writers' research productivity by quickly surfacing candidate sources, extracting key facts, and organizing evidence, freeing the human to focus on synthesis, authentication, and creative interpretation of findings. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI substantially speeds up preliminary research, summarization, and idea generation, letting writers focus more time on verification and creative synthesis while remaining in the loop. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Current AI systems can autonomously search, retrieve, and synthesize factual information from newspapers, public records, diaries, and interview transcripts at scale with >50% time savings over manual research. Systems like web search agents, document retrieval models, and summarization tools perform this end-to-end. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can synthesize publicly available information and summarize sources quickly, but authentic primary-source research (interviews, archival diaries, verifying facts) still requires human judgment and access, limiting full end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal or regulatory barriers prevent AI-assisted or automated research; no licensing requirement mandates human researchers. Some organizational preference for human judgment on source authenticity and relevance introduces modest friction, but nothing structural blocks substitution. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There are no licensing or legal requirements restricting AI-assisted research for creative writing; adoption is purely a matter of creative preference and workflow. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven research (automated search, indexing, summarization) costs orders of magnitude less than hiring humans to manually review newspapers, archives, and conduct transcribed interviews. Overhead is minimal after initial tooling. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI can cheaply gather and summarize secondary information, but sourcing authentic primary materials (interviews, physical diaries) still requires human time, making overall cost comparable rather than dramatically cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature AI products (search engines, document databases, enterprise research platforms with AI indexing) reliably retrieve and organize factual information from standard sources. Some error in source verification or interpretation requires human oversight, preventing a perfect 5. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | AI research assistants and search-augmented LLMs are deployed and used for background research, but they have material error rates (hallucination, inability to access physical archives) and lack true verification of authenticity. |
Follow appropriate procedures to get copyrights for completed work.
62CI 43–82 · exposure 66 · augmentation 75 · importance 3.5/5 · click for rater detail
Follow appropriate procedures to get copyrights for completed work.
62| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Creative writing and publishing sectors lag significantly in AI adoption relative to finance and tech; while some platforms offer automated copyright assistance, uptake remains limited and pilots outnumber production deployments in most creative communities. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Creative writing and publishing sectors show slow, uneven AI adoption for administrative/legal tasks like copyright filing, with most creators still following manual procedures or using generic legal-tech tools sparingly. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can substantially assist writers by tracking multiple works, managing registration deadlines, prefilling forms accurately, and providing checklists, allowing humans to focus on authorship decisions rather than administrative drudgery. The human remains in control while productivity rises measurably. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI chatbots and legal assistants can meaningfully help writers understand copyright steps, draft applications, and check details, substantially easing the administrative burden even though a human must still submit and confirm. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Current AI systems can navigate copyright registration procedures end-to-end by filling out standardized forms, uploading work samples, tracking deadlines, and managing submissions to copyright offices. This process is highly procedural, rule-based, and well-documented, achieving substantial time savings over manual completion. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can help draft copyright applications and explain procedures, but filing itself requires navigating government portals, verifying ownership details, and submitting forms which still need human execution and legal accuracy checks.time savings are moderate, not full end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Copyright registration procedures are open to automated processing by the copyright office; there is no legal requirement that a human must perform the filing. The main friction is user preference for human review and minor organizational delays, not hard regulatory barriers. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Copyright registration must be filed by the rights holder or an authorized agent, and errors carry legal consequences, creating moderate procedural and liability barriers even though it isn't a licensed-professional-only task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-based form filling and workflow management costs pennies per registration, versus hiring a paralegal or attorney to manage copyright procedures, which costs hundreds of dollars per submission. The cost differential is at least an order of magnitude. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI assistance for drafting and guidance is cheap, but the overall process still requires human review and interaction with official registries, so total cost savings versus a human doing it are moderate, not order-of-magnitude. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Legal document-preparation and copyright registration guidance software exists and performs reliably in production for standard registrations. However, edge cases involving disputed authorship, derivative works, or complex corporate ownership still benefit from human review, preventing a full 5 rating. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | There is no mature deployed product that autonomously completes copyright registration end-to-end; existing tools (chatbots, legal-tech assistants) can guide but not reliably execute filings without human verification. |
Plan project arrangements or outlines, and organize material accordingly.
53CI 43–64 · exposure 38 · augmentation 88 · importance 4.0/5 · click for rater detail
Plan project arrangements or outlines, and organize material accordingly.
53| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While some writers experiment with AI for outlining and organization, adoption remains limited and cautious in creative writing sectors. Most agencies and publishing houses still rely on human project managers and editors, with AI tools treated as optional assistants rather than replacements. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Creative writing is an information-based, digitized field where AI writing tools have seen notable uptake, though many professional writers remain resistant or use AI only informally. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at suggesting outlines, reorganizing material, and offering structural alternatives that writers can evaluate and refine. This augments writer productivity significantly while keeping the human in creative control, making AI valuable as a collaborative planning tool. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI is widely used as a brainstorming and structuring aid, helping writers quickly generate and iterate on outlines while retaining full creative control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate outlines and organize text, planning project arrangements requires domain-specific judgment about creative direction, audience, and narrative structure that current systems struggle with consistently. Most creative work benefits from AI assistance rather than end-to-end automation, and the organizational choices often reflect artistic intent that humans must validate. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate outlines, chapter structures, and organizational schemes from a prompt fairly well, but adapting to a writer's unique creative vision and iterative refinement still requires substantial human input to reach equal quality. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are few legal or regulatory barriers to using AI for planning and organizing creative work; however, writers and publishing organizations often prefer human-directed project management to preserve creative control and quality standards. Market preference and organizational culture provide some friction but not hard barriers. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing, legal, or liability requirement mandating human authorship of an outline; writers are free to use any tool they choose. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-assisted outline generation and material organization costs pennies per task compared to a writer's hourly rate, making the cost ratio heavily favors automation from a pure economics standpoint. Even accounting for human oversight and revision, the per-task cost is substantially lower than human labor. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating a draft outline via AI costs pennies compared to the hours a writer might spend planning manually, though human review and revision add some cost back. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI outline and organization tools exist (GPT plugins, Notion AI), but they produce generic structures without deep understanding of the specific creative project's vision or constraints. Deployed systems lack reliability for nuanced creative planning and typically require substantial human revision and judgment. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Writing assistant products (e.g., Sudowrite, Scrivener AI integrations, ChatGPT) reliably produce outlines and structural suggestions today, but they are used as starting points rather than final organizational plans in professional creative work. |
Write fiction or nonfiction prose, such as short stories, novels, biographies, articles, descriptive or critical analyses, and essays.
52CI 45–59 · exposure 38 · augmentation 88 · importance 4.7/5 · click for rater detail
Write fiction or nonfiction prose, such as short stories, novels, biographies, articles, descriptive or critical analyses, and essays.
52| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Some publishers and content mills experiment with AI-assisted writing; adoption is accelerating in lower-stakes content (marketing copy, basic articles). Literary fiction and serious nonfiction remain human-dominant, indicating middling sectoral adoption with pockets of faster growth. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Publishing and media are adopting AI tools for brainstorming, drafting, and editing assistance, but full displacement of human authorship remains limited and adoption is uneven across fiction/nonfiction genres. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at drafting, brainstorming, outlining, and iterative refinement, meaningfully accelerating a writer's productivity. Many professional writers now use LLMs as collaborative tools for overcoming blocks and generating variations, with the human retaining creative control and final voice. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI is heavily used by writers today for brainstorming, drafting, overcoming writer's block, editing, and research, substantially boosting productivity while the human retains creative control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Current AI can generate prose rapidly, but creative writing at professional quality requires originality, emotional depth, voice, and thematic coherence that AI rarely achieves end-to-end without significant human revision. Time savings exist for drafting, but quality parity with published human work remains inconsistent. |
| Task automatability | claude-sonnet-5 | 3/5 | LLMs can draft prose quickly and can produce full short pieces, but professional-quality novels, biographies, and critical essays still require substantial human revision, research verification, and voice/originality that current AI struggles to fully replace at equal quality. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Publishing, literary attribution, and editorial standards create soft barriers: readers and markets expect human authorship, and plagiarism/originality concerns raise friction. However, no legal licensing requirement prevents AI text generation, and these are adoption friction rather than hard regulatory blocks. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing is required to write creative works, but publishers, agents, and audiences often value authentic human authorship, and copyright/authorship norms create some friction against pure AI output. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference and integration costs for generating thousands of words are negligible compared to a human writer's hourly wage, making per-word production vastly cheaper despite quality gaps that currently require human post-processing. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating draft prose via API/subscription costs cents to a few dollars versus the hours or days a human writer would charge, though oversight and editing costs reduce the gap somewhat. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | LLMs produce text at scale, but deployed products lack reliable ability to sustain narrative complexity, character development, and originality across full-length works. Most professional outputs remain hybrid (AI-assisted drafting requiring substantial human authorship) rather than AI-authored. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like ChatGPT, Claude, and specialized writing tools are widely used to draft fiction and nonfiction, but published-quality output at scale still requires heavy human editing, and no product reliably produces publishable long-form work end-to-end. |
Choose subject matter and suitable form to express personal feelings and experiences or ideas, or to narrate stories or events.
49CI 40–59 · exposure 30 · augmentation 75 · importance 4.5/5 · click for rater detail
Choose subject matter and suitable form to express personal feelings and experiences or ideas, or to narrate stories or events.
49| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Adoption of AI writing tools for creative writing is growing in some sectors (marketing, advertising, content mills) but remains contested in traditional literary and arts spaces. Pilots are common, but production adoption for serious creative writing remains limited and primarily in supplementary roles. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Creative and media sectors are adopting AI writing tools for drafting and ideation at a moderate pace, though professional creative writing/publishing still largely resists full AI authorship due to originality and voice concerns. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at augmenting creative writers by generating drafts, exploring variations, offering stylistic suggestions, and accelerating ideation and revision cycles while the human maintains creative direction and judgment. Many writers report significant productivity gains when using AI as a collaborative tool. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI is widely used to brainstorm subject matter, suggest forms, generate drafts, and overcome writer's block, substantially boosting productivity while the writer retains creative control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can generate text on specified topics and in various stylistic forms, but choosing subject matter and form that authentically expresses personal feelings, experiences, or ideas requires subjective judgment and intentionality that current systems cannot replicate at equal quality to human creative work. The task fundamentally involves personal creative vision, which AI lacks. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate text on chosen subjects and forms, but this task centers on authentic personal experience and creative intent, which current systems cannot originate meaningfully, only simulate. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While there are no hard legal barriers preventing AI-assisted writing, significant organizational and market friction exists: literary and publishing industries value human authorship, readers expect human creativity, and authorship attribution matters for reputation and sales. Contractual and professional norms create meaningful adoption friction. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or legal requirement mandates a human for creative writing; anyone can use AI tools to produce or assist with this work. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference costs for text generation are extremely low compared to the loaded wage of a professional writer, making the cost-per-output vastly favorable to automation on a pure computational basis. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Generating text output via LLM inference costs a tiny fraction of a cent compared to a human writer's time, an enormous cost differential. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI systems can produce writing in many genres and forms, deployed products do not reliably replicate the depth, authenticity, and originality expected in professional creative writing. AI-generated content is widely recognized as derivative and formulaic, making production-scale deployment for this specific task limited to assistance roles rather than independent execution. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Deployed generative AI products (ChatGPT, Claude, etc.) reliably produce poems, lyrics, and stories in various forms on demand, though quality and originality vary widely. |
Develop factors such as themes, plots, characterizations, psychological analyses, historical environments, action, and dialogue to create material.
45CI 45–45 · exposure 25 · augmentation 88 · importance 4.7/5 · click for rater detail
Develop factors such as themes, plots, characterizations, psychological analyses, historical environments, action, and dialogue to create material.
45| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | AI writing tools are increasingly adopted in professional services and content industries for drafting and brainstorming, but deep replacement in high-stakes creative contexts (literature, screenwriting) remains limited; usage is primarily augmentative rather than substitutive. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Writers and studios are experimenting with AI brainstorming and drafting tools, with growing but still contested and uneven adoption amid industry pushback (e.g., WGA disputes). |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI writing assistants demonstrably enhance human creative productivity by generating drafts, exploring alternative plot structures, suggesting dialogue, and enabling rapid iteration, allowing human writers to focus on originality and refinement. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI is widely used by writers today for brainstorming plots, generating character ideas, drafting dialogue options, and overcoming writer's block, substantially boosting creative productivity while the writer retains final creative control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate text with themes, plots, and dialogue, the creative integration and originality required to meet the ≥50% time-saving bar at equal quality is not reliably achieved. Current systems produce derivative or formulaic content that typically requires substantial human revision and creative direction. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate plausible drafts of themes, plots, and dialogue, but genuinely original, high-quality creative development that meets professional publication standards still requires substantial human judgment and revision, falling short of the 50% time-saving-at-equal-quality bar for most professional work. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are few hard legal or licensing barriers to AI-generated creative writing; however, market and quality barriers remain strong—publishers, audiences, and clients often prefer human authorship, and copyright/attribution issues create organizational friction around AI-generated content. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for creative writing, but originality, copyright/authorship concerns, and audience/publisher preference for human-created work create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference and integration costs are very low per output token, making the marginal cost substantially cheaper than a human creative writer's hourly rate, even accounting for oversight and revision overhead. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating draft text via AI is extremely cheap compared to a professional writer's time, though the output typically requires significant human revision to reach usable quality, tempering pure cost savings somewhat. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed writing assistants exist (ChatGPT, Claude) but primarily function as drafting aids with high variability in quality and originality. Production systems using AI to autonomously develop complex narratives with psychological depth and thematic coherence at professional literary standards are not yet mature at scale. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Consumer LLM products can produce story elements on demand, but no deployed product reliably generates publishable-quality literary material (novels, lyrics, screenplays) without heavy human editing and creative direction. |
Prepare works in appropriate format for publication, and send them to publishers or producers.
43CI 34–52 · exposure 38 · augmentation 75 · importance 4.5/5 · click for rater detail
Prepare works in appropriate format for publication, and send them to publishers or producers.
43| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Publishing remains a relatively traditional, human-relationship-driven sector with limited digital transformation of submission workflows; adoption of automated submission tools is present but slow and remains marginal. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Publishing and creative writing remain a cottage-industry-like sector with slow, uneven AI tool adoption for administrative/submission tasks compared to fast-moving digital sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist by formatting works to specification, generating submission checklists, identifying potential publishers by genre/imprint, and drafting cover letters, substantially raising a writer's productivity in the administrative aspects of publication. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools substantially speed up formatting, proofreading, and adapting manuscripts to submission guidelines, meaningfully augmenting writer productivity while the human retains control over final content and choice of publisher. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist with formatting and submission logistics, but the creative judgment of selecting appropriate publishers, tailoring manuscript presentation, and making final editorial decisions requires human expertise. Current systems lack the contextual understanding to fully replace this task end-to-end. |
| Task automatability | claude-sonnet-5 | 3/5 | Formatting manuscripts to publisher/producer specifications (style guides, submission templates, metadata) is largely mechanical and can be handled by AI tools, but researching appropriate publishers, tailoring pitches, and managing submission relationships still require human judgment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Submission typically involves human relationship-building with editors and publishers, plus creative judgment about market fit that organizations prefer to retain human control over; however, there are no hard legal or licensing barriers to partial automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for formatting and submitting creative work, though publishers often prefer direct human correspondence and agents, creating mild organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | Automation of formatting and submission routing could reduce labor costs, but the oversight required to ensure manuscripts reach appropriate venues and meet publisher expectations keeps total costs comparable to a human coordinator. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-assisted formatting and templating tools are cheap, but the overall task still requires human time for submission strategy and relationship management, keeping total cost roughly comparable to a human doing it manually with some AI help. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While tools exist for formatting (templates, style guides) and basic submission workflows, no deployed product reliably handles the nuanced decision-making of selecting suitable publishers, understanding submission requirements, and managing rejection/revision cycles at production scale. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Word processors, formatting plugins, and AI writing assistants can reformat and proofread text reliably, but there is no mature end-to-end product that autonomously identifies publishers and submits polished works without human oversight. |
Write words to fit musical compositions, including lyrics for operas, musical plays, and choral works.
42CI 34–51 · exposure 33 · augmentation 63 · importance 4.3/5 · click for rater detail
Write words to fit musical compositions, including lyrics for operas, musical plays, and choral works.
42| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption in theatrical and operatic contexts remains very limited; the sector values human artistry and established workflows. Use is mostly experimental or in low-budget independent projects rather than mainstream commercial musicals or opera houses. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Creative writing and performing arts sectors have been slower and more resistant to AI adoption compared to information/finance sectors, with cultural and artistic-integrity concerns slowing uptake. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by generating initial drafts, exploring rhyme and meter options, and offering variations on themes, enabling human lyricists to iterate faster. However, the human must still refine, evaluate, and ensure artistic coherence and musical fit. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI is quite useful for brainstorming rhymes, alternate phrasings, and quick drafts to fit meter, meaningfully speeding up a lyricist's iterative process while the writer retains creative control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate text that fits meter and rhyme schemes, producing lyrics that meaningfully align with specific musical compositions, match emotional intent, and meet professional quality standards requires human creativity and judgment. Current systems struggle with coherent narrative arc and thematic depth in extended works. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate lyrics that fit meter and rhyme schemes reasonably well and can iterate quickly, but matching nuanced musical phrasing, dramatic character voice, and thematic coherence for opera or musical theater still requires substantial human revision and judgment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Artistic and contractual barriers moderate adoption: producers typically require human creative authorship for legal and reputational reasons, and union rules may protect lyricist roles in theatrical productions. However, no hard legal mandate exists in most jurisdictions for human-authored lyrics. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists, but strong industry norms, collaborative creative processes with composers/directors, and audience/critical expectations for authentic authorship create moderate friction against pure AI substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Inference costs for text generation are very low, and integration oversight is minimal compared to the loaded wage of a professional lyricist or composer-lyricist, making AI substantially cheaper on a per-output basis. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI drafting is very cheap per iteration, but achieving usable quality for professional creative writing requires significant human editing time, narrowing the overall cost advantage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI lyric-generation tools exist but are primarily used for experimentation or as assistants rather than producing production-ready lyrics for operas, musicals, or major choral works. Published systems lack reliable integration with specific musical scores and client approval workflows. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Consumer AI tools produce song lyrics and can attempt fitting syllable counts to melodies, but no mature production system reliably delivers professional-grade lyrics for operas or musical theater without heavy human rewriting. |
Write humorous material for publication, or for performances such as comedy routines, gags, and comedy shows.
42CI 39–45 · exposure 25 · augmentation 75 · importance 3.4/5 · click for rater detail
Write humorous material for publication, or for performances such as comedy routines, gags, and comedy shows.
42| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | AI writing tools are gaining traction in content production sectors and entertainment, with comedy writers beginning to experiment with AI assistance. However, adoption remains in the pilot and early-stage phase rather than at the scale or depth of finance or professional services. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Entertainment and creative writing remain a laggard sector for AI adoption in production, with writers largely using AI as a brainstorming aid rather than replacing writers rooms or joke-writing staff. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at rapidly generating comedic premises, variations, and alternative punchlines, allowing human comedians and writers to iterate faster and explore more conceptual territory. Many professional comedy writers are already using AI tools to brainstorm and refine material, significantly boosting creative velocity while retaining human judgment on final selection and delivery. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI is widely used by comedy writers for brainstorming premises, generating joke variations, and overcoming writer's block, meaningfully speeding up ideation while the writer curates and refines final material. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate text-based humor quickly, producing consistently publishable or performance-ready comedic material that meets professional quality standards and achieves the intended laugh remains difficult. Current AI struggles with timing, cultural nuance, and the unpredictability required for truly effective humor, making reliable ≥50% time savings at equal quality unlikely for this creative task. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate joke-like text quickly but consistently producing genuinely funny, original, timing-sensitive material for professional publication or performance still requires heavy human curation and rewriting, falling short of the 50% equal-quality bar for most professional use cases. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are no licensing or regulatory requirements for publishing humorous content, and organizations face minimal legal barriers to deploying AI-assisted comedy. However, reputational risk (unfunny or offensive output) and audience preference for human comedic voice create some friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates human authorship, but audience/client preference for authentic human voice, comedic timing, and cultural nuance creates moderate organizational and market resistance. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | API-based language models are very inexpensive per inference, and the marginal cost of generating thousands of comedic variations is minimal compared to hiring professional comedy writers. Even accounting for human review and editing, the AI-assisted pipeline is likely 5–10× cheaper than pure human authorship. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Generating raw comedic text is cheap via LLMs, but the human review, curation, and rewriting needed to reach usable quality narrows the cost advantage versus a skilled comedy writer. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI writing tools can assist with joke generation and comedic text, but deployed products rarely produce polished, performance-ready humor without substantial human revision. The error rates and narrow scope of generated humor mean few organizations rely on AI alone for publishable comedy material. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Consumer chatbots can produce jokes and comedic drafts, but no deployed product reliably generates performance-ready comedy material at professional quality without extensive human editing. |
Revise written material to meet personal standards and to satisfy needs of clients, publishers, directors, or producers.
37CI 28–46 · exposure 30 · augmentation 75 · importance 4.6/5 · click for rater detail
Revise written material to meet personal standards and to satisfy needs of clients, publishers, directors, or producers.
37| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Publishing and creative industries are gradually adopting AI writing tools for drafting and copyediting assistance, but adoption of AI for revision remains cautious and augmentative rather than replacement-focused. Early-stage tool use is evident but not yet deeply embedded in production workflows. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Creative and publishing industries have adopted AI writing tools moderately for drafting and editing assistance, but full creative writing production remains human-led with cautious, uneven adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at augmenting revision by suggesting alternative word choices, identifying grammatical issues, checking consistency, and offering stylistic options that the writer or editor then evaluates and refines. This assistive role is already transforming productivity in revision workflows while keeping the human in creative control. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI is highly useful for suggesting alternate phrasings, catching inconsistencies, and iterating quickly on drafts, significantly speeding up the revision process while the writer retains final creative control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist with mechanical editing (grammar, style) and generate alternative phrasings, but revision requires understanding authorial intent, client vision, and creative judgment that AI cannot reliably perform end-to-end. The task involves subjective standards and often conflicting stakeholder needs that demand human decision-making. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can suggest edits and revisions quickly, but genuinely satisfying nuanced personal artistic standards or specific client/director creative intent requires judgment and taste AI cannot reliably replicate end-to-end at equal quality. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Artistic control, creative ownership, and client relationship management create strong barriers to full automation. Publishers, directors, and producers typically require human judgment on revisions, and many contracts specify human editorial review. Client relationships and trust in creative judgment are difficult to delegate to AI. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but strong client/audience preference for authentic human voice, copyright/authorship norms, and reputational stakes create moderate friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools cost a fraction of hiring a human editor, but the user (poet, writer, or client) still pays for the tool and must spend significant time reviewing, filtering, and refining AI suggestions. All-in cost remains high relative to straightforward outsourcing to a junior editor in low-cost markets. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI drafting/editing assistance is cheap per iteration, but human oversight, back-and-forth with clients, and final creative judgment keep overall costs comparable to a skilled writer's time rather than an order of magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI writing assistants exist (Grammarly, Claude, GPT-4) and can flag issues and suggest edits, but they operate as co-editing tools rather than autonomous revisers. No deployed product reliably revises material to meet a specific client's or director's creative standards without substantial human oversight and final approval. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like ChatGPT, Grammarly, and specialized writing assistants are widely used for revision suggestions, but they are not deployed as autonomous end-to-end revision systems replacing the writer's judgment in production creative contexts. |
Adapt text to accommodate musical requirements of composers and singers.
36CI 25–47 · exposure 33 · augmentation 63 · importance 4.3/5 · click for rater detail
Adapt text to accommodate musical requirements of composers and singers.
36| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption remains limited to niche experimental projects and amateur contexts; professional music production (film, theater, commercial recording) has shown slow, cautious uptake of AI-assisted lyric adaptation due to quality concerns and creator preference for human collaboration. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | placeholder |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by rapidly generating rhyming options, suggesting syllable-count adjustments, and brainstorming thematic variations, enabling a human lyricist to iterate faster; however, the augmentation is modest because the core task of judging fit between text and music remains subjective and human-centric. |
| Augmentation potential | claude-sonnet-5 | 4/5 | placeholder |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can generate rhyming alternatives and suggest structural edits to fit meter or syllable counts, but adapting text while preserving artistic intent, emotional resonance, and narrative coherence for music collaboration requires deep creative judgment that current systems cannot reliably execute end-to-end without substantial human rework. |
| Task automatability | claude-sonnet-5 | 3/5 | LLMs can generate and revise lyrics to fit meter, syllable count, and rhyme schemes reasonably well, but truly matching nuanced musical phrasing, emotional intent, and singer-specific vocal ranges still requires significant human judgment and iteration."},"feasibility":{"rating":2,"rationale":"Some AI lyric-writing tools exist and are used experimentally by musicians, but no mature product reliably handles the full task of adapting text to precise musical/vocal constraints in professional production workflows."},"cost_ratio":{"rating":3,"rationale":"AI drafting is cheap, but professional lyricists still need to refine and integrate output with composers, so overall cost savings are moderate rather than order-of-magnitude."},"barriers":{"rating":2,"rationale":"No licensing or legal barrier exists, but strong industry preference for human creative authorship and collaboration with composers creates some friction against pure AI substitution."},"adoption_velocity":{"rating":2,"rationale":"The music and creative writing industry has been slower and more resistant to AI adoption relative to other professional services, with use mostly experimental rather than embedded in production pipelines."},"augmentation":{"rating":4,"rationale":"AI can quickly generate alternative phrasings, rhyme options, and syllable-adjusted lines, meaningfully speeding up the iterative process of fitting lyrics to music while the human retains creative control."}}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong artistic and contractual barriers exist: composers and singers typically require creative control and copyright attribution over adapted lyrics, and reputation risk makes organizations reluctant to substitute human lyricists with automated text adaptation, especially for commercial releases. |
| Adoption barriers | claude-sonnet-5 | 2/5 | placeholder |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI tools are cheap to run but require significant human oversight and revision; when accounting for the human creative labor needed to fix outputs and integrate them into a composer's workflow, the all-in cost per usable adaptation remains comparable to or higher than direct human authorship. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | placeholder |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI writing tools exist and can produce text variations, no deployed product reliably performs the iterative, context-aware adaptation of existing creative text to specific musical constraints (melody, pacing, emotional arc) at production quality without extensive human correction. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | placeholder |
Write narrative, dramatic, lyric, or other types of poetry for publication.
34CI 25–43 · exposure 25 · augmentation 63 · importance 4.4/5 · click for rater detail
Write narrative, dramatic, lyric, or other types of poetry for publication.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Poetry and literary publishing are traditionally low-digital sectors with cultural resistance to algorithmic authorship. Adoption of AI poetry tools remains experimental and marginal; most adoption is in hobbyist or educational contexts rather than professional publishing pipelines. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | The literary/creative writing sector has been slow and often resistant to AI adoption, with many publications explicitly banning AI-generated submissions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist poets by suggesting rhyme schemes, generating initial stanzas to overcome block, or exploring variations on a theme, offering moderate productivity gains in brainstorming and drafting phases. However, the core creative and revision work remains primarily human-driven. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI is widely used by writers for brainstorming, overcoming blocks, generating variations, and exploring forms/meter, meaningfully assisting the creative process while the poet retains authorial control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Current AI can generate verse matching formal constraints and produce thematically coherent poetry, but struggles with originality, emotional authenticity, and the distinctive voice that distinguishes published work. Significant human curation and rewriting is required to meet publication standards, falling short of the 50% time-saving threshold for equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate poetry text quickly, but producing publishable, original creative work with authentic voice and artistic merit still falls short of the quality bar, so it doesn't cleanly meet the 50% time-savings-at-equal-quality threshold for professional-level output. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Publishing houses maintain human editorial gatekeeping; copyright and attribution norms (disclosure of AI involvement) create friction; reader and author communities strongly prefer human-authored work; and economic incentives favor original human creativity. These combine to create substantial organizational and market barriers to substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for poetry, but literary journals, publishers, and readers have strong preferences for human-authored work and disclosure norms are emerging, creating some reputational/market friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI inference is cheap, but the integration cost (prompt engineering, iterative refinement, human editorial labor to salvage output) approaches or exceeds the loaded cost of a professional poet working at reasonable speed, especially for specialized or high-quality verse. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Generating draft poetry via AI costs pennies compared to a professional poet's time, though human editing/curation is usually still needed to reach publishable quality. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While LLMs can produce grammatically correct poetry and some systems offer poetry-generation tools, no deployed product reliably produces publication-ready poems without substantial human revision. Benchmark demos exist but production use in literary publishing remains minimal and requires heavy editorial oversight. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Consumer LLM products can produce poem drafts on demand, but no deployed product reliably produces publication-quality poetry accepted by literary markets at scale; adoption remains novelty/assistive rather than production-grade authorship. |
Teach writing classes.
28CI 25–30 · exposure 25 · augmentation 75 · importance 2.8/5 · click for rater detail
Teach writing classes.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Educational institutions adopt AI gradually and cautiously, with most AI writing tools used as supplements rather than replacements for instruction. Adoption remains pilot-heavy in K–12 and higher ed, with significant organizational and cultural resistance to instructor automation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Education sector adoption of AI for full course instruction remains slow and mostly limited to supplementary tools rather than replacing instructors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI writing assistants significantly augment instruction: they provide instant feedback loops, generate diverse prompts, help instructors grade at scale, and allow teachers to focus on higher-order mentorship. Many educators report measurably improved efficiency and student engagement with AI-assisted feedback. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can generate writing prompts, give instant feedback on drafts, suggest exercises, and grade assignments, meaningfully supporting an instructor's workload. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can generate writing exercises, provide feedback on student work, and draft instructional materials, but cannot replicate the dynamic, adaptive mentorship, live critique, and motivational scaffolding that constitute effective teaching. The human instructor remains essential for responsive instruction. |
| Task automatability | claude-sonnet-5 | 2/5 | Teaching involves live interaction, adaptive feedback, classroom management, and personal mentorship that current AI cannot fully replicate end-to-end, though it can assist with materials and exercises. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Schools and universities typically require accredited human instructors by policy or regulation; student–instructor relationship and live interaction are institutional and pedagogical requirements. Liability and learning-outcome accountability strongly favor human sign-off. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement typically, but institutions and students strongly prefer human instructors for mentorship, feedback nuance, and accreditation purposes, creating moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | LLM inference costs for personalized writing instruction, combined with required human oversight and integration, approach or exceed hourly teaching wages in many contexts. Full displacement would require prohibitively cheap inference or acceptance of lower instructional quality. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | A human instructor's salary is comparable to or cheaper than building and maintaining an AI-led course with necessary human oversight, especially for small class sizes. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI tutoring systems exist for writing but are narrow in scope and produce inconsistent pedagogical results. No mature product reliably teaches writing classes end-to-end; most deployed systems offer targeted practice or asynchronous feedback, not live instruction replacement. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tutoring products exist for writing feedback but no deployed product independently runs a full writing class with the pedagogical and interpersonal quality of a human instructor. |
Confer with clients, editors, publishers, or producers to discuss changes or revisions to written material.
27CI 20–34 · exposure 20 · augmentation 63 · importance 4.2/5 · click for rater detail
Confer with clients, editors, publishers, or producers to discuss changes or revisions to written material.
27| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Publishing and creative services remain relationship-heavy, trust-dependent sectors where face-to-face or live conferencing is valued and expected. Adoption of AI for conference-replacing automation is very limited; AI is mostly used for pre-meeting summarization or post-meeting documentation, not for conducting the meeting itself. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Creative writing and publishing sectors show slow, cautious AI adoption for interpersonal creative negotiation tasks, though drafting tools are used elsewhere in the workflow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist by preparing revision summaries, tracking feedback threads across stakeholders, generating alternative phrasings for discussion, and creating redline comparisons, allowing the human writer and editor to confer more efficiently and comprehensively. This transforms the administrative overhead of conferencing while keeping the human in full control. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help writers prepare revision summaries, brainstorm alternatives, or draft responses to editorial notes, meaningfully aiding preparation even though it doesn't replace the conversation itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft revision suggestions and summarize feedback, the negotiation and nuanced discussion aspect of conferring—reading stakeholder intent, adjusting tone based on interpersonal dynamics, and reaching consensus—requires human judgment. Current AI cannot reliably conduct the full back-and-forth client interaction autonomously. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a live, relational negotiation involving judgment about creative intent, client relationships, and contractual/business considerations; AI can draft talking points but cannot conduct the actual conference or make real-time collaborative decisions.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Clients, editors, and publishers often have contractual or established relationships requiring direct human contact and personal judgment, and authors typically demand direct input on editorial decisions affecting their work. Legal and contractual obligations to involve named humans create friction against full substitution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement, but strong preference for human relationship-building, trust, and nuanced back-and-forth in creative collaboration creates real friction against replacement. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI infrastructure for supporting revision discussions (summarization, draft comparison) is cheap, but still requires human oversight and the core conferencing remains human-performed; cost is roughly comparable to a human assistant, not dramatically cheaper than the primary task performer. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply summarize notes or suggest revisions, but the core discussion still requires paid human time, so overall cost savings are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI tools exist that can generate revision notes and compare drafts, but no deployed product reliably conducts the full conference interaction with multiple stakeholders, capturing context-specific concerns and building agreement. The synchronous, relationship-dependent nature of the task remains largely unautomated in practice. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for the human conversation between a writer and editor/producer about revisions; this remains an interpersonal, research-stage-at-best application for AI. |
Collaborate with other writers on specific projects.
21CI 7–35 · exposure 13 · augmentation 63 · importance 2.7/5 · click for rater detail
Collaborate with other writers on specific projects.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption in creative writing remains slow and experimental; most professional collaborations still rely on human-to-human interaction, and AI is used primarily as an optional drafting tool rather than a core collaborator. Resistance from established creative communities and concerns about authenticity limit deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Creative writing and publishing are adopting AI tools for drafting and brainstorming, but collaborative authorship arrangements remain human-centric with limited AI integration into the collaborative process itself. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by suggesting alternative phrasings, identifying structural issues, or generating starting points for discussion, which accelerates the collaborative refinement loop. However, the core work of negotiating creative vision and reaching consensus still falls substantially to the human writers themselves. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist co-writers by generating drafts, suggesting alternatives, and facilitating brainstorming, enhancing productivity while humans retain the collaborative and creative decision-making roles. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist in generating draft content and suggesting revisions, but collaboration fundamentally requires human judgment, creative decision-making, and interpersonal negotiation about artistic direction that current systems cannot autonomously navigate. The task involves real-time creative problem-solving and consensus-building that falls short of the 50% time-saving threshold for end-to-end automation. |
| Task automatability | claude-sonnet-5 | 1/5 | Collaboration is an interpersonal, negotiated creative process involving trust, shared vision, and social dynamics that current AI cannot replace end-to-end; AI can contribute text but cannot 'collaborate' as a peer writer autonomously. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: the task requires human judgment and creative authority that clients and publishers typically demand from credentialed writers; liability and reputation risk remain with the human collaborators; and organizational norms and contractual arrangements usually require human sign-off on collaborative creative output. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing barrier exists, but the task inherently requires human relationship, trust, and shared authorial voice, which creates strong organizational and creative friction against AI substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI infrastructure and oversight costs for collaborative writing tasks remain comparable to or exceed the cost of direct human collaboration, especially when quality control and creative direction are factored in. The overhead of managing AI-assisted workflows does not yet yield significant economic advantage. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Since AI cannot perform the collaborative task itself, there's no valid cost comparison for full task substitution; hiring/maintaining a human co-writer remains necessary. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI systems can help draft content and comment on existing work, no deployed product reliably handles the full collaborative workflow—joint ideation, conflict resolution, iterative feedback loops, and creative alignment—that human writers routinely execute. Existing tools are supplements, not replacements for the collaborative process itself. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for a human co-writer in a creative partnership; AI writing tools exist but function as assistants, not autonomous collaborators fulfilling this social role. |
Attend book launches and publicity events, or conduct public readings.
3CI 0–5 · exposure 0 · augmentation 25 · importance 3.0/5 · click for rater detail
Attend book launches and publicity events, or conduct public readings.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task occurs in low-digitization, human-centric sectors (publishing, live events) with no measurable AI adoption or displacement because the task is inherently about physical, interpersonal presence. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Publishing and live events are low-digitization, relationship-driven activities showing no meaningful AI displacement trend for personal appearances. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist marginally by drafting press materials, managing event logistics, or selecting which readings to prioritize, but the core act of attending and performing cannot be meaningfully augmented by AI. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help prepare talking points, promotional materials, or rehearse readings, but offers minimal assistance during the live event itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task fundamentally requires physical presence at venues and direct human-audience interaction. AI cannot travel to events, present itself on stage, or engage authentically with live audiences in real-time conversation. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires physical or live virtual presence, personal charisma, and real-time audience interaction that AI cannot perform as a substitute for the human creator. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong barriers exist: audiences expect direct human contact with the author, contracts specify personal appearance, and literary events depend on authentic human-writer presence for marketing and audience satisfaction. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Publicity events and readings depend on audience expectation of meeting the actual author; while not legally restricted, strong social/organizational norms and personal branding make substitution highly impractical. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost comparison is moot since AI cannot perform this task. A human author's travel and appearance time is substantially cheaper than any hypothetical robotic or AI proxy infrastructure. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this task, so no meaningful cost comparison exists—the human must be present, making AI not a viable cheaper alternative. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously attend events or conduct live public readings. This is purely a human performative activity with no AI substitute in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product attends events or performs public readings on behalf of a human author; this remains entirely outside current AI product capability. |
Related occupations — Arts, Design, Entertainment, Sports & Media
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.