Proofreader and Copy Marker
Proofreading scored 0.
Based on Microsoft's "Working with AI" study of real Copilot conversations mapped to O*NET tasks (arXiv 2507.07935): Proofreaders and Copy Markers scored 0.37 on AI applicability, more than two standard deviations above the cross-occupation mean (0.159, stdev 0.098) across all 785 SOC codes studied — banded here as High real-world AI usage relative to other occupations.
Corroborating signal from the Anthropic Economic Index (Claude.ai/API conversations, a different product and userbase than the Microsoft data above): this occupation shows observed exposure of 0.18, above the 75th percentile of the 756 occupations AEI tracks (dataset mean 0.077) but the lowest of the eight occupations in this pilot batch — a real but comparatively modest signal.
What a proofreader and copy marker actually does
Proofreaders and copy markers review typeset or drafted material for errors before it's published or printed. The job means reading text closely for grammar, spelling, and punctuation mistakes; checking that formatting, spacing, and layout match the intended style; and marking up corrections using standard proofreading marks or tracked changes so an editor or typesetter can fix them.
Proofreaders review corrected proofs a second time to confirm every flagged error was actually fixed, consult style guides and reference materials to resolve questions about usage or house style, and sometimes verify factual details like names, dates, and figures against a source document. The work is detail-intensive and repetitive by design — reading the same piece multiple times, at a slower pace than normal reading, specifically to catch what a normal read-through would miss.
It's typically the last human check before a piece of writing goes out into the world, whether that's a print publication, a legal document, or a marketing brochure.
Why it reads this way
Proofreading scored 0. 37 on Microsoft's AI applicability scale, more than two standard deviations above the cross-occupation average of 0. 159, placing it in the High usage band. That result lines up with what's already common practice: grammar and style-checking tools built on language models now catch the overwhelming majority of the mechanical errors — spelling, punctuation, subject-verb agreement, awkward phrasing — that used to require a dedicated human pass. The Anthropic Economic Index shows an observed exposure of 0.
18 for this occupation, above the 75th percentile of the 756 occupations it tracks, though the lowest of the eight occupations in this pilot batch — consistent with proofreading being a smaller, more specialized occupation with fewer people generating usage data in either direction.
BLS data shows this as one of the smallest occupations we've profiled, with a shrinking pool of dedicated proofreading roles as the function increasingly gets absorbed into editing software and folded into other jobs (editor, writer, admin) rather than staffed as a standalone position.
Skills this role draws on
Source: U.S. Bureau of Labor Statistics OEWS wage data (May 2024), accessed via O*NET OnLine wage report, SOC 43-9081.00
Safer, skill-adjacent careers
These aren't generic "consider retraining" suggestions. Each one shares real skill or task overlap with proofreader and copy marker work, and each one scores meaningfully higher on our structural exposure scale, with the reasoning shown below.
This is a near-direct skill transfer: both jobs are built around meticulously checking work against a defined standard and flagging every deviation — the difference is inspecting physical parts on a line instead of text on a page, which is why it scores 78 versus 12.
Proofreaders with strong attention to fine detail can apply that same meticulousness to dental hygiene, a hands-on clinical role that requires physical precision and direct patient care — score 92 versus 12.
Both roles reward careful, checklist-driven precision, but surgical technology applies it to physically preparing instruments and assisting in real operating-room procedures — score 89 versus 12.
Data sources & methodology
Salary data: U.S. Bureau of Labor Statistics OEWS wage data (May 2024), accessed via O*NET OnLine wage report, SOC 43-9081.00.
Task descriptions: Based on O*NET occupational analysis (43-9081.00).
Real-world AI usage band: Microsoft's "Working with AI" study of Bing Copilot conversations mapped to O*NET tasks (arXiv 2507.07935), and corroborating data from the Anthropic Economic Index.
Growth projections: -1% (2024-2034), based on BLS Occupational Outlook Handbook.