Federal Reserve Bank of New York · May 14, 2026
Audoly, Guerin and Topa argue in Liberty Street Economics (May 14, 2026) that AI is not the main driver of slower hiring
Markdown · JSON · Original source
On May 14, 2026, Richard Audoly, Miles Guerin, and Giorgio Topa examine whether job postings show early labor-market effects of AI. They report that postings for occupations with greater declined relative to less-exposed occupations, but that divergence began before ChatGPT’s release and showed no clear additional break afterward. They also find no clear divergence between junior and senior postings within highly exposed occupations. The authors argue that these patterns make it difficult to attribute the recent slowdown in entry-level hiring to AI alone. They note that overall hiring has slowed and unemployment has increased among young workers and recent college graduates; their conclusion is that AI may be contributing to labor-market developments, but is not the main driver of the hiring slowdown. Their measure of AI exposure reflects potential exposure based on observed usage, not a prediction of reduced hiring or increased layoffs. Job postings therefore offer evidence about hiring demand, but the findings leave open how much AI affects employment through channels these postings do not capture.
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.
Original source
Federal Reserve Bank of New York · Article
Source details
Original source date: May 14, 2026 (2026-05-14T11:00:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-07T04:13:24.299Z. First observed: 2026-10-23T16:28:09.755Z. Retrieved: 2026-10-23T16:29:39.686Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
Text as retrieved on October 23, 2026; originally published May 14, 2026. The analysis uses job-posting data to study labor demand; the post does not establish the full effect of AI on employment through other channels.
Open question
How much, if at all, is AI affecting employment through channels that job postings do not capture?
Source-stated
Audoly, Guerin, and Topa report that job postings provide little indication of a distinct AI-driven decline in labor demand.
the evidence from job postings provides little indication of a distinct AI-driven decline in labor demand.
Source-stated
The authors report that the relative decline in postings for occupations with greater AI exposure began before ChatGPT’s release and showed no clear additional break afterward.
The divergence between high- and low-exposure occupations began before 2022 and does not show a clear additional break in trajectory after 2022.
Source-stated
The authors find no clear upward or downward trend in postings for junior roles relative to senior roles within highly exposed occupations.
Instead, the line fluctuates, without a clear upward or downward trend.
Source-stated
The authors conclude that AI may contribute to labor-market developments but is not the main driver of the hiring slowdown.
AI may be contributing to recent labor market developments, it is not the main driver of the slowdown in hiring.
Source-stated
The authors say their AI exposure measure describes potential exposure based on observed usage, not a guaranteed change in hiring or layoffs.
A job being exposed to AI may not translate into reduced hiring or increased layoffs for the occupation as a whole
Source-stated
The post is dated May 14, 2026.
May 14, 2026 Do Job Postings Show Early Labor‑Market Effects of AI?
Source and contributor context
Federal Reserve Bank of New York · 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.
Editorial provenance
Version 1. Publication ID: ef1a598b-54ac-71fa-8113-d7500475605c.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-07T04:08:49.452Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_screen.
Observation: a6aae9be241a0e95734d32be2e6b9ff1d407d00aa80a9a1f61e845788f728bc1. Source text hash: e76eb5acb72eb83b76cd875f4a8cdc851f9ad9d9e1c4c0b978e40914db155732.
Revision and source-check history
Version 1 · 2026-11-07T04:13:24.299Z
Audoly, Guerin and Topa argue in Liberty Street Economics (May 14, 2026) that AI is not the main driver of slower hiring
On May 14, 2026, Richard Audoly, Miles Guerin, and Giorgio Topa examine whether job postings show early labor-market effects of AI. They report that postings for occupations with greater declined relative to less-exposed occupations, but that divergence began before ChatGPT’s release and showed no clear additional break afterward. They also find no clear divergence between junior and senior postings within highly exposed occupations. The authors argue that these patterns make it difficult to attribute the recent slowdown in entry-level hiring to AI alone. They note that overall hiring has slowed and unemployment has increased among young workers and recent college graduates; their conclusion is that AI may be contributing to labor-market developments, but is not the main driver of the hiring slowdown. Their measure of AI exposure reflects potential exposure based on observed usage, not a prediction of reduced hiring or increased layoffs. Job postings therefore offer evidence about hiring demand, but the findings leave open how much AI affects employment through channels these postings do not capture.
Original source
Federal Reserve Bank of New York · Article
Source details
Original source date: May 14, 2026 (2026-05-14T11:00:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-07T04:13:24.299Z. First observed: 2026-10-23T16:28:09.755Z. Retrieved: 2026-10-23T16:29:39.686Z.
Limits
Text as retrieved on October 23, 2026; originally published May 14, 2026. The analysis uses job-posting data to study labor demand; the post does not establish the full effect of AI on employment through other channels.
Open question
How much, if at all, is AI affecting employment through channels that job postings do not capture?
Source-stated
Audoly, Guerin, and Topa report that job postings provide little indication of a distinct AI-driven decline in labor demand.
the evidence from job postings provides little indication of a distinct AI-driven decline in labor demand.
Source-stated
The authors report that the relative decline in postings for occupations with greater AI exposure began before ChatGPT’s release and showed no clear additional break afterward.
The divergence between high- and low-exposure occupations began before 2022 and does not show a clear additional break in trajectory after 2022.
Source-stated
The authors find no clear upward or downward trend in postings for junior roles relative to senior roles within highly exposed occupations.
Instead, the line fluctuates, without a clear upward or downward trend.
Source-stated
The authors conclude that AI may contribute to labor-market developments but is not the main driver of the hiring slowdown.
AI may be contributing to recent labor market developments, it is not the main driver of the slowdown in hiring.
Source-stated
The authors say their AI exposure measure describes potential exposure based on observed usage, not a guaranteed change in hiring or layoffs.
A job being exposed to AI may not translate into reduced hiring or increased layoffs for the occupation as a whole
Source-stated
The post is dated May 14, 2026.
May 14, 2026 Do Job Postings Show Early Labor‑Market Effects of AI?
Source and contributor context
Federal Reserve Bank of New York · 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.
Editorial provenance
Version 1. Publication ID: ef1a598b-54ac-71fa-8113-d7500475605c.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-07T04:08:49.452Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_screen.
Observation: a6aae9be241a0e95734d32be2e6b9ff1d407d00aa80a9a1f61e845788f728bc1. Source text hash: e76eb5acb72eb83b76cd875f4a8cdc851f9ad9d9e1c4c0b978e40914db155732.
passed · 2026-11-07T04:08:49.452Z · anthropic/claude-opus-5.5
Publication: ef1a598b-54ac-71fa-8113-d7500475605c
attribution — pass: The reading attributes the analysis to Richard Audoly, Miles Guerin and Giorgio Topa in the New York Fed's Liberty Street Economics. This matches the byline and citation line. The page is a single argued post, not a listing. The related-story links at the end are not used.
Richard Audoly, Miles Guerin, and Giorgio Topa
summary — pass: The reading conveys the core argument: the relative decline in postings for exposed occupations predates ChatGPT, and junior and senior postings show no clear divergence. It keeps the authors' hedged conclusion that AI may contribute but is not the main driver. AI exposure is defined beside the reading. Two details are omitted, the gap stabilizing after 2023 and the retraining survey, but neither changes the meaning. The title compresses 'suggests' into 'argue' but keeps the conclusion's scope.
The evidence from job postings suggests that while AI may be contributing to recent labor market developments, it is not the main driver of the slowdown in hiring.
date — pass: evidence.originalPublished is 2026-05-14T11:00Z, which matches the dateline and citation of May 14, 2026. The page shows a dateModified of June 2, 2026. The reading relies only on core body text that is plainly part of the original post. Nothing later or in hindsight is presented. Evidence kind is publisher-page, so no archive disclosure is needed.
Federal Reserve Bank of New York Liberty Street Economics, May 14, 2026
support — pass: Each claim's excerpt appears verbatim in the core body text, which is plainly original despite the later page modification. The closing point, that job postings leave other employment channels open, and the question are framed as unresolved limits, not facts. They are consistent with the authors' caveat that exposure may not translate into reduced hiring. No benefits or costs are invented.
A job being exposed to AI may not translate into reduced hiring or increased layoffs for the occupation as a whole
media — pass: No media is selected and no verified candidates were supplied, 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.
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.
Definition referenceIn labor economics, how much of a job's tasks an AI system could in principle perform; it is not a forecast of job loss.
Definition reference