Skip to content
Escape
Meridian

The future is still open.

· Jun 3, 2025

Revelio Labs tracks AI-exposed tasks disappearing from job postings

Markdown · JSON · Original source

reported its analysis on June 3, 2025, finding that tasks appeared less often in job postings between early 2022 and early 2025. The analysis defines these as day-to-day activities that AI could plausibly perform. The share fell from 29% to 25.5%, which Revelio Labs says suggests companies are redesigning processes, not merely adding AI to existing work. Most of the decline in average AI exposure came from firms hiring less for highly exposed occupations; a smaller share came from pruning tasks within jobs they continued to fill. Revelio Labs points to changes in job descriptions for clerical, managerial and technical roles, and to declines in demand for routine financial, insurance and tax work. It argues that postings offer a ground-level view of changing job content before those changes appear in employment statistics. The analysis concerns what employers list as job duties, not a direct measurement of which tasks AI actually performs or how workers’ jobs change. The extent to which the posting trends reflect changes in day-to-day work remains unknown.

Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.

Revelio LabsArticle · reveliolabs.comOpen original source

Original source

· Article

Source details

Original source date: Jun 3, 2025 (2025-06-03T16:15:28.267Z).

Date basis: Date stated on the publisher’s page.

Added to Escape Meridian: 2026-11-13T17:53:24.060Z. First observed: 2026-11-09T01:47:43.633Z. Retrieved: 2026-11-09T01:49:36.482Z.

Referenced by

Recorded citations and public file attachments. A reference alone does not establish agreement or influence.

No public references recorded here yet.

Reference data (JSON)

Limits

Text as retrieved on November 9, 2026; originally published June 3, 2025. The analysis uses job postings as evidence of changes in listed duties; it does not directly establish which tasks AI performs or how workers’ day-to-day work changes. The supplied text does not describe the method used to classify tasks as or provide a comparison showing what would have happened without AI.

Open question

Do the changes in job postings correspond to changes in the tasks workers actually perform?

Source-stated

Revelio Labs found that the share of listed activities classified as AI-exposed declined between early 2022 and early 2025.

AI-exposed tasks made up 29% of listed activities in early 2022, but only 25.5% by early 2025.

Source-stated

Revelio Labs attributes most of the decline in average AI exposure to firms hiring less for highly exposed roles, with the rest coming from duties removed within roles still being filled.

Roughly 84% of the drop in average AI exposure comes from shifts between roles—firms hiring fewer high-exposure occupations—while the remaining 16% stems from companies actively pruning AI-exposed duties within the roles they still fill.

Source-stated

Revelio Labs says Data and Document Clerks had the largest reported reduction in AI-exposed tasks among the occupations it examined.

Data and Document Clerks top the list, with the share of AI-exposed tasks in job postings dropping 8.2 percentage points between 2022 and 2024

Source-stated

Revelio Labs says routine financial, insurance and tax duties were among the tasks with the largest declines in demand within the occupations it examined.

Core back-office functions dominate the list, including processing financial transactions and client insurance needs, managing insurance and tax transactions, and financial compliance and tax advisory services.

Source-stated

Revelio Labs argues that tracking job postings offers an early view of changes in job content, before they appear in employment statistics.

tracking postings gives us a live, ground-level view of how organizations are reshaping job content and the very nature of work in the age of AI, well before those changes show up in employment statistics.

Source and contributor context

Revelio Labs · Publisher

The publisher is identified here. No individual author has been inferred.

Source media

Media retrieval status: unchecked.

Terms and references

Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.

AI exposure

In labor economics, how much of a job's tasks an AI system could in principle perform; it is not a forecast of job loss.

Revelio Labs

Revelio Labs is a workforce analytics firm publishing labor market analysis on employment, occupations and industry trends.

Editorial provenance

Version 1. Publication ID: 9607217c-c095-08ed-554e-eee8fa6ba87e.

Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.

Source check: passed by anthropic/claude-opus-5.5 on 2026-11-13T17:49:40.685Z. This checks fidelity to the source, not whether its claims are independently true.

Editorial selection: accept; reasons: bridge_admitted_by_model.

Observation: dee90d39b7fac5c4213450b358a8bb0e42f7eded47e891fd1f86aacf7827e01e. Source text hash: 9eaa2da6044744f02eaf73b288573a70d16defb946167f6699adbf9896c008b4.

Revision and source-check history

Version 1 · 2026-11-13T17:53:24.060Z

Revelio Labs tracks AI-exposed tasks disappearing from job postings

reported its analysis on June 3, 2025, finding that tasks appeared less often in job postings between early 2022 and early 2025. The analysis defines these as day-to-day activities that AI could plausibly perform. The share fell from 29% to 25.5%, which Revelio Labs says suggests companies are redesigning processes, not merely adding AI to existing work. Most of the decline in average AI exposure came from firms hiring less for highly exposed occupations; a smaller share came from pruning tasks within jobs they continued to fill. Revelio Labs points to changes in job descriptions for clerical, managerial and technical roles, and to declines in demand for routine financial, insurance and tax work. It argues that postings offer a ground-level view of changing job content before those changes appear in employment statistics. The analysis concerns what employers list as job duties, not a direct measurement of which tasks AI actually performs or how workers’ jobs change. The extent to which the posting trends reflect changes in day-to-day work remains unknown.

Original source

· Article

Source details

Original source date: Jun 3, 2025 (2025-06-03T16:15:28.267Z).

Date basis: Date stated on the publisher’s page.

Added to Escape Meridian: 2026-11-13T17:53:24.060Z. First observed: 2026-11-09T01:47:43.633Z. Retrieved: 2026-11-09T01:49:36.482Z.

Limits

Text as retrieved on November 9, 2026; originally published June 3, 2025. The analysis uses job postings as evidence of changes in listed duties; it does not directly establish which tasks AI performs or how workers’ day-to-day work changes. The supplied text does not describe the method used to classify tasks as or provide a comparison showing what would have happened without AI.

Open question

Do the changes in job postings correspond to changes in the tasks workers actually perform?

Source-stated

Revelio Labs found that the share of listed activities classified as AI-exposed declined between early 2022 and early 2025.

AI-exposed tasks made up 29% of listed activities in early 2022, but only 25.5% by early 2025.

Source-stated

Revelio Labs attributes most of the decline in average AI exposure to firms hiring less for highly exposed roles, with the rest coming from duties removed within roles still being filled.

Roughly 84% of the drop in average AI exposure comes from shifts between roles—firms hiring fewer high-exposure occupations—while the remaining 16% stems from companies actively pruning AI-exposed duties within the roles they still fill.

Source-stated

Revelio Labs says Data and Document Clerks had the largest reported reduction in AI-exposed tasks among the occupations it examined.

Data and Document Clerks top the list, with the share of AI-exposed tasks in job postings dropping 8.2 percentage points between 2022 and 2024

Source-stated

Revelio Labs says routine financial, insurance and tax duties were among the tasks with the largest declines in demand within the occupations it examined.

Core back-office functions dominate the list, including processing financial transactions and client insurance needs, managing insurance and tax transactions, and financial compliance and tax advisory services.

Source-stated

Revelio Labs argues that tracking job postings offers an early view of changes in job content, before they appear in employment statistics.

tracking postings gives us a live, ground-level view of how organizations are reshaping job content and the very nature of work in the age of AI, well before those changes show up in employment statistics.

Source and contributor context

Revelio Labs · Publisher

The publisher is identified here. No individual author has been inferred.

Source media

Media retrieval status: unchecked.

Terms and references

Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.

AI exposure

In labor economics, how much of a job's tasks an AI system could in principle perform; it is not a forecast of job loss.

Revelio Labs

Revelio Labs is a workforce analytics firm publishing labor market analysis on employment, occupations and industry trends.

Editorial provenance

Version 1. Publication ID: 9607217c-c095-08ed-554e-eee8fa6ba87e.

Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.

Source check: passed by anthropic/claude-opus-5.5 on 2026-11-13T17:49:40.685Z. This checks fidelity to the source, not whether its claims are independently true.

Editorial selection: accept; reasons: bridge_admitted_by_model.

Observation: dee90d39b7fac5c4213450b358a8bb0e42f7eded47e891fd1f86aacf7827e01e. Source text hash: 9eaa2da6044744f02eaf73b288573a70d16defb946167f6699adbf9896c008b4.

passed · 2026-11-13T17:49:40.685Z · anthropic/claude-opus-5.5

Publication: 9607217c-c095-08ed-554e-eee8fa6ba87e

attribution — pass: The reading consistently attributes the analysis, figures and interpretations to Revelio Labs, which is the publisher of this single newsletter-style analysis. It does not present the findings as established fact or as Escape Meridian's view.

we leverage longitudinal job posting data to understand the historical and current demand for “AI-exposed” tasks

summary — pass: The reading leads with the actor and the finding, and defines AI-exposed tasks in plain language. It reports the 29% to 25.5% decline and Revelio's process-redesign interpretation, the between-role versus within-role split, the occupations and tasks affected, and the argument that postings are an early signal. The question follows from a real gap. The definitions list explains AI exposure.

AI-exposed tasks made up 29% of listed activities in early 2022, but only 25.5% by early 2025.

date — pass: June 3, 2025 matches the UTC calendar day of originalPublished and the source's own citation line. There is no hindsight framing, and the limitation's retrieval note is accurate. The page was edited after publication. The cited passages are the core findings, dated by figures labelled June 2025, and are plainly part of the original piece. The kind is publisher-page, so no archive disclosure is required.

Revelio Labs, "The Tasks You Won’t See in Job Postings Anymore," June 3, 2025.

support — pass: Each claim is supported by its excerpt. The process-redesign framing is attributed as Revelio's suggestion. The limitations about posting data versus actual work, and about the undescribed classification method, are accurate gaps rather than invented claims. The cited text appears to belong to the original analysis despite the later page modification.

This trend suggests that companies are embedding AI into operations not just as an add-on, but as a core driver of process redesign.

media — pass: No media candidates were supplied, and the reading selects none, so there is no media identity or relevance issue.

Challenge or correct this reading

Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.