· Sep 14, 2026
Epoch AI compares weekly AI-use measures among US adults
Markdown · JSON · Original source
published this analysis on September 14, 2026. It examines how often US adults used AI services in the past week, comparing survey waves from March and August 2026. The analysis groups respondents by days of use and reports each group as a share of all US adults, not just AI users. The survey questions differed between waves: March asked about overall use, while August asked about use of each service separately. Epoch AI says it collapsed the August responses into the same groups used for March. That difference leaves an open question: how much of the measured change reflects actual changes in use, and how much may reflect the different questions? The retrieved text describes the analysis but does not include the results needed to assess the size or direction of any change.
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
· Article
Source details
Original source date: Sep 14, 2026 (2026-09-14T00:00:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-05T21:53:23.878Z. First observed: 2026-10-18T02:49:42.733Z. Retrieved: 2026-10-24T03:15:37.503Z.
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 24, 2026; originally published September 14, 2026. The retrieved text describes the analysis and methods but does not include numerical results, so the size or direction of any change cannot be assessed from this evidence. The different question formats may affect comparability between survey waves; the retrieved text does not quantify that effect.
Open question
How much of the measured change reflects actual changes in use, and how much may reflect the different questions?
Source-stated
Epoch AI says the analysis compares weekly AI service use among US adults across survey waves from March and August 2026.
This analysis examines weekly AI use among US adults between March and August 2026.
Source-stated
The analysis groups respondents by days of use and calculates the groups as shares of all US adults.
The groups are 1 day, 2–5 days, and 6–7 days. All shares are percentages of all US adults rather than of AI users.
Source-stated
The survey waves used different questions about days of AI use.
March asked a single overall question whose response options were these three groups, while August asked day counts for each service separately
Escape Meridian interpretation
Escape Meridian reads the difference in question format as a reason to ask how much of any measured change reflects survey design rather than changed use.
The two waves measured days of use differently.
Source and contributor context
Epoch AI · 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.
Epoch AI
Epoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Editorial provenance
Version 1. Publication ID: 0ad38c23-4e7d-300e-198e-4dda5b0b1369.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-05T21:48:41.835Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: f6beb7db6716101ab46babe70dd1c2b6373bf53647edba83daec9ff548658ffa. Source text hash: 963e36ef107b9f4db389c5c22332c872928e625dc7270b34a9f68c4acb5530e0.
Revision and source-check history
Version 1 · 2026-11-05T21:53:23.878Z
Epoch AI compares weekly AI-use measures among US adults
published this analysis on September 14, 2026. It examines how often US adults used AI services in the past week, comparing survey waves from March and August 2026. The analysis groups respondents by days of use and reports each group as a share of all US adults, not just AI users. The survey questions differed between waves: March asked about overall use, while August asked about use of each service separately. Epoch AI says it collapsed the August responses into the same groups used for March. That difference leaves an open question: how much of the measured change reflects actual changes in use, and how much may reflect the different questions? The retrieved text describes the analysis but does not include the results needed to assess the size or direction of any change.
Original source
· Article
Source details
Original source date: Sep 14, 2026 (2026-09-14T00:00:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-05T21:53:23.878Z. First observed: 2026-10-18T02:49:42.733Z. Retrieved: 2026-10-24T03:15:37.503Z.
Limits
Text as retrieved on October 24, 2026; originally published September 14, 2026. The retrieved text describes the analysis and methods but does not include numerical results, so the size or direction of any change cannot be assessed from this evidence. The different question formats may affect comparability between survey waves; the retrieved text does not quantify that effect.
Open question
How much of the measured change reflects actual changes in use, and how much may reflect the different questions?
Source-stated
Epoch AI says the analysis compares weekly AI service use among US adults across survey waves from March and August 2026.
This analysis examines weekly AI use among US adults between March and August 2026.
Source-stated
The analysis groups respondents by days of use and calculates the groups as shares of all US adults.
The groups are 1 day, 2–5 days, and 6–7 days. All shares are percentages of all US adults rather than of AI users.
Source-stated
The survey waves used different questions about days of AI use.
March asked a single overall question whose response options were these three groups, while August asked day counts for each service separately
Escape Meridian interpretation
Escape Meridian reads the difference in question format as a reason to ask how much of any measured change reflects survey design rather than changed use.
The two waves measured days of use differently.
Source and contributor context
Epoch AI · 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.
Epoch AI
Epoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Editorial provenance
Version 1. Publication ID: 0ad38c23-4e7d-300e-198e-4dda5b0b1369.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-05T21:48:41.835Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: f6beb7db6716101ab46babe70dd1c2b6373bf53647edba83daec9ff548658ffa. Source text hash: 963e36ef107b9f4db389c5c22332c872928e625dc7270b34a9f68c4acb5530e0.
uncertain · 2026-11-05T21:48:41.835Z · anthropic/claude-opus-5.5
Publication: 0ad38c23-4e7d-300e-198e-4dda5b0b1369
attribution — pass: The reading attributes the analysis and method description to Epoch AI, the publisher. The page is a single data insight, not a listing. Small omission: the download label names Ipsos as the survey data source, and the reading never says whose surveys were analyzed.
Weekly AI Use by Days of Use, March vs August 2026 (Ipsos)
summary — uncertain: The reading erases the source's central finding. The source title states that near-daily AI use among US adults more than doubled in six months. Yet the reading says the size and direction of any change cannot be assessed. The body text lacks numbers, but the headline claim is the source's own thesis. Smallest repair: state that Epoch AI's headline reports near-daily use more than doubled, attributed to Epoch, then keep the comparability caveat. The rest of the reading, including the share-of-all-adults detail and the change in question format, is clear to a general reader.
This analysis examines weekly AI use among US adults between March and August 2026.
date — pass: September 14, 2026 matches the UTC calendar day of originalPublished and agrees with the CSV update date. The survey waves are stated at month precision, as in the source. The related-insight dates are not used. There is no hindsight framing, and this is a publisher page, so no archive disclosure is needed.
CSV, Updated Sep. 14, 2026
support — pass: Each source-stated claim matches the text: the March–August scope, the three day-of-use groups, shares of all US adults, and the differing question formats with the August answers collapsed into the same groups. The comparability question is labeled as Escape Meridian interpretation and preserves the gap between evidence and conclusion.
March asked a single overall question whose response options were these three groups, while August asked day counts for each service separately, which we collapsed into the same groups.
media — pass: No media is selected, and no verified media candidates were supplied. This 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.
Epoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AI