Federal Reserve Bank of New York · Apr 14, 2026
Ali Hashim, Gizem Kosar, and Wilbert van der Klaauw argue in Liberty Street Economics that workplace AI training access is uneven
Markdown · JSON · Original source
On April 14, 2026, Ali Hashim, Gizem Kosar, and Wilbert van der Klaauw report findings from supplemental questions in the November 2025 Survey of Consumer Expectations, fielded to a representative sample of the U.S. population. They find that workplace AI use is concentrated among higher-income, higher-educated, and full- workers, while some workers report that their employers do not offer AI tools or prohibit their use. Among employed respondents, 38 percent say AI training is important to them, but only 15.9 percent report that their employer offers it. Workers without employer-provided training report an average willingness to give up 11.4 percent of their salary for a job with extensive training; workers who already have access report that, on average, they would require a 24.2 percent salary increase to accept a similar job without it. The authors say this gap may reflect loss aversion or selection into jobs where training is more valuable. They emphasize that some groups with higher reported value for training, including workers without a college degree, also have lower AI use and access to employer-provided training. They argue that closing this gap may be essential to reaching workplace productivity gains. The survey records respondents’ reports and expectations; it does not establish how training changes productivity or what effects broader access would have.
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: Apr 14, 2026 (2026-04-14T14:30:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-14T11:08:24.784Z. First observed: 2026-10-23T17:27:46.499Z. Retrieved: 2026-10-23T17:29:41.098Z.
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 April 14, 2026. The article reports survey respondents’ answers; it does not establish that training causes productivity gains or explain how training affects employment outcomes. The authors note that employer-provided access to training does not necessarily mean workers have received training or feel confident using AI tools.
Open question
How would access to AI training change workers’ productivity and employment outcomes, particularly for workers who currently report lower access?
Source-stated
The authors report that workplace AI use varies across worker groups and is concentrated among higher-income, higher-educated, and full-time workers.
Taken together, these patterns suggest that AI adoption at work currently favors higher-income, higher-educated, and full-time workers.
Source-stated
The authors report that 38 percent of employed respondents consider AI training important, while 15.9 percent say their employer offers it.
Around 38 percent of employed respondents said that having training in how to use AI tools is important to them, yet only 15.9 percent report that their employer currently offers any AI training.
Source-stated
Workers without employer-provided AI training report an average willingness to give up 11.4 percent of their current salary for a job with extensive training.
Among workers who currently lack access to training, the average willingness to pay (WTP) for gaining this access is 11.4 percent of current salary.
Source-stated
Workers who already have access to training report that they would require an average salary increase of 24.2 percent to accept an otherwise identical job without it.
On average, these workers report that they would require a 24.2 percent salary increase to accept an otherwise identical job that doesn’t offer access to any AI training
Source-stated
The authors say that some workers who place high value on AI training, including workers without a college degree, also have lower AI use and access to employer-provided training.
some of the workers who place the highest value on AI training, such as those without a college degree, are also those with the lowest rates of AI usage and the lowest share of access to employer-provided training in how to use AI tools.
Source-stated
The authors argue that closing the training-access gap may be essential to reaching workplace productivity gains.
Closing this gap may be essential to reaching the productivity gains from having generative AI tools in the workplace.
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.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 99b311c5-228e-d7d7-96e4-713748bbb5d1.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-14T11:03:48.563Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model.
Observation: ec35390637f4e97d57ce10d6b66b83d77bd330e82db988fdb7b5772e8458b3dd. Source text hash: debdaa0a83b471d660b98444a129a869332e835df8766f8058ad30734b8e19ee.
Revision and source-check history
Version 1 · 2026-11-14T11:08:24.784Z
Ali Hashim, Gizem Kosar, and Wilbert van der Klaauw argue in Liberty Street Economics that workplace AI training access is uneven
On April 14, 2026, Ali Hashim, Gizem Kosar, and Wilbert van der Klaauw report findings from supplemental questions in the November 2025 Survey of Consumer Expectations, fielded to a representative sample of the U.S. population. They find that workplace AI use is concentrated among higher-income, higher-educated, and full- workers, while some workers report that their employers do not offer AI tools or prohibit their use. Among employed respondents, 38 percent say AI training is important to them, but only 15.9 percent report that their employer offers it. Workers without employer-provided training report an average willingness to give up 11.4 percent of their salary for a job with extensive training; workers who already have access report that, on average, they would require a 24.2 percent salary increase to accept a similar job without it. The authors say this gap may reflect loss aversion or selection into jobs where training is more valuable. They emphasize that some groups with higher reported value for training, including workers without a college degree, also have lower AI use and access to employer-provided training. They argue that closing this gap may be essential to reaching workplace productivity gains. The survey records respondents’ reports and expectations; it does not establish how training changes productivity or what effects broader access would have.
Original source
Federal Reserve Bank of New York · Article
Source details
Original source date: Apr 14, 2026 (2026-04-14T14:30:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-14T11:08:24.784Z. First observed: 2026-10-23T17:27:46.499Z. Retrieved: 2026-10-23T17:29:41.098Z.
Limits
Text as retrieved on October 23, 2026; originally published April 14, 2026. The article reports survey respondents’ answers; it does not establish that training causes productivity gains or explain how training affects employment outcomes. The authors note that employer-provided access to training does not necessarily mean workers have received training or feel confident using AI tools.
Open question
How would access to AI training change workers’ productivity and employment outcomes, particularly for workers who currently report lower access?
Source-stated
The authors report that workplace AI use varies across worker groups and is concentrated among higher-income, higher-educated, and full-time workers.
Taken together, these patterns suggest that AI adoption at work currently favors higher-income, higher-educated, and full-time workers.
Source-stated
The authors report that 38 percent of employed respondents consider AI training important, while 15.9 percent say their employer offers it.
Around 38 percent of employed respondents said that having training in how to use AI tools is important to them, yet only 15.9 percent report that their employer currently offers any AI training.
Source-stated
Workers without employer-provided AI training report an average willingness to give up 11.4 percent of their current salary for a job with extensive training.
Among workers who currently lack access to training, the average willingness to pay (WTP) for gaining this access is 11.4 percent of current salary.
Source-stated
Workers who already have access to training report that they would require an average salary increase of 24.2 percent to accept an otherwise identical job without it.
On average, these workers report that they would require a 24.2 percent salary increase to accept an otherwise identical job that doesn’t offer access to any AI training
Source-stated
The authors say that some workers who place high value on AI training, including workers without a college degree, also have lower AI use and access to employer-provided training.
some of the workers who place the highest value on AI training, such as those without a college degree, are also those with the lowest rates of AI usage and the lowest share of access to employer-provided training in how to use AI tools.
Source-stated
The authors argue that closing the training-access gap may be essential to reaching workplace productivity gains.
Closing this gap may be essential to reaching the productivity gains from having generative AI tools in the workplace.
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.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 99b311c5-228e-d7d7-96e4-713748bbb5d1.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-14T11:03:48.563Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model.
Observation: ec35390637f4e97d57ce10d6b66b83d77bd330e82db988fdb7b5772e8458b3dd. Source text hash: debdaa0a83b471d660b98444a129a869332e835df8766f8058ad30734b8e19ee.
passed · 2026-11-14T11:03:48.563Z · anthropic/claude-opus-5.5
Publication: 99b311c5-228e-d7d7-96e4-713748bbb5d1
attribution — pass: The reading correctly attributes the post to Ali Hashim, Gizem Kosar, and Wilbert van der Klaauw in Liberty Street Economics. It also keeps the survey findings and the interpretation that closing the gap 'may be essential' attributed to the authors.
Ali Hashim, Gizem Kosar, and Wilbert van der Klaauw
summary — pass: The reading names the actors, the survey and its key findings in plain language: uneven AI use, the gap between valuing training and employers offering it, and the willingness-to-pay asymmetry. It keeps the authors' hedged conclusion and its caveats. 'Loss aversion' is not explained, but it is a secondary interpretive term and does not block understanding. The question follows from the evidence.
a significant share of employers are nonetheless not yet providing access to AI tools or training on how to use them
date — pass: The stated date, April 14, 2026, matches both the source dateline and evidence.originalPublished. The page was edited after publication (dateModified 2026-06-02). However, the cited passages are the article's core body under its April 14 dateline and citation, not sidebar or update material. The reading does not use the later related-story links or present any hindsight. The retrieval-date note is accurate.
Federal Reserve Bank of New York Liberty Street Economics, April 14, 2026
support — pass: The source supports every figure and claim: 38 percent versus 15.9 percent, willingness to pay of 11.4 percent versus a required 24.2 percent, the loss aversion and selection explanations, the non-college group's low access, and the closing argument. The limitations correctly note that the data are self-reports, and the source includes the authors' caveat that access to training does not mean workers have received it.
may also reflect patterns of loss aversion
media — pass: No media was selected, and no verified candidate media was supplied, so the absence is appropriate.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIMETIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIME