· Sep 30, 2026
Pew Research Center examines whether AI can stand in for survey-takers
Markdown · JSON · Original source
’s report, published September 30, 2026, examines whether AI-generated survey data can reproduce results from public opinion polls answered by people. The Center says its experiment used an AI model to adopt the personas of real members of its American Trends Panel and answer the same survey questions as human panelists. The Center reports that the AI-generated results often differed from human responses, with errors that varied across questions and groups. It also says the model often avoided some answer choices, portrayed subgroup attitudes as more widespread than they were, and was less likely than human respondents to select “not sure.” In a comparison of two models, the Center reports that their estimates differed in the direction of the opinions they portrayed. The report argues that synthetic surveys are not a replacement for surveying real people on matters of public importance. Its findings apply to using AI as a survey-taker; the report does not assess other uses of AI in polling or . How well would these results generalize to other models, survey designs, or groups of people?
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 30, 2026 (2026-09-30T17:55:17.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-10-30T15:39:02.051Z. First observed: 2026-10-18T03:43:22.236Z. Retrieved: 2026-10-30T04:21:45.664Z.
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 30, 2026; originally published September 30, 2026. The study examines AI as a survey-taker and does not assess other uses of synthetically generated data. The report’s findings concern its experiment and do not establish how results would generalize to other models, survey designs, or groups.
Open question
How well would these results generalize to other models, survey designs, or groups of people?
Source-stated
Pew Research Center published the report on September 30, 2026.
Report | September 30, 2026 Can AI Stand In for Human Survey-Takers? Not Really
Source-stated
Pew Research Center’s experiment compared AI responses based on real panelists’ personas with responses from human panelists.
We used a “digital twins” approach for this study, asking an AI model to adopt the personas of real humans who are members of the ATP.
Source-stated
The Center reports that AI survey results often differed from human survey results, and argues AI polling is not a replacement for surveying real people.
Our biggest takeaway from this exercise is that AI polling is not a replacement for rigorously surveying real humans.
Source-stated
The report says its study focuses on AI as a survey-taker, not other uses of synthetically generated data.
This study focuses solely on the use of AI as a survey-taker. It does not address other uses of synthetically generated data, such as estimating or imputing missing values in existing datasets.
Source and contributor context
Pew Research Center · 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.
Synthetic data
Training or test data generated by software or another model rather than collected from the world.
Pew Research Center
Pew Research Center studies public attitudes and social change through surveys and other empirical research.
Editorial provenance
Version 1. Publication ID: be6e75aa-36bd-dc14-2bea-019d66fbda98.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-10-30T15:38:59.486Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model.
Observation: 8f451dc4e00deb4fc53e09174a872cd2082418f34486af7c447575a9f2a66ff2. Source text hash: 86bf2f500964f56912a37236e09ecb2381217927eade8e88f0551782be00c893.
Revision and source-check history
Version 1 · 2026-10-30T15:39:02.051Z
Pew Research Center examines whether AI can stand in for survey-takers
’s report, published September 30, 2026, examines whether AI-generated survey data can reproduce results from public opinion polls answered by people. The Center says its experiment used an AI model to adopt the personas of real members of its American Trends Panel and answer the same survey questions as human panelists. The Center reports that the AI-generated results often differed from human responses, with errors that varied across questions and groups. It also says the model often avoided some answer choices, portrayed subgroup attitudes as more widespread than they were, and was less likely than human respondents to select “not sure.” In a comparison of two models, the Center reports that their estimates differed in the direction of the opinions they portrayed. The report argues that synthetic surveys are not a replacement for surveying real people on matters of public importance. Its findings apply to using AI as a survey-taker; the report does not assess other uses of AI in polling or . How well would these results generalize to other models, survey designs, or groups of people?
Original source
· Article
Source details
Original source date: Sep 30, 2026 (2026-09-30T17:55:17.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-10-30T15:39:02.051Z. First observed: 2026-10-18T03:43:22.236Z. Retrieved: 2026-10-30T04:21:45.664Z.
Limits
Text as retrieved on October 30, 2026; originally published September 30, 2026. The study examines AI as a survey-taker and does not assess other uses of synthetically generated data. The report’s findings concern its experiment and do not establish how results would generalize to other models, survey designs, or groups.
Open question
How well would these results generalize to other models, survey designs, or groups of people?
Source-stated
Pew Research Center published the report on September 30, 2026.
Report | September 30, 2026 Can AI Stand In for Human Survey-Takers? Not Really
Source-stated
Pew Research Center’s experiment compared AI responses based on real panelists’ personas with responses from human panelists.
We used a “digital twins” approach for this study, asking an AI model to adopt the personas of real humans who are members of the ATP.
Source-stated
The Center reports that AI survey results often differed from human survey results, and argues AI polling is not a replacement for surveying real people.
Our biggest takeaway from this exercise is that AI polling is not a replacement for rigorously surveying real humans.
Source-stated
The report says its study focuses on AI as a survey-taker, not other uses of synthetically generated data.
This study focuses solely on the use of AI as a survey-taker. It does not address other uses of synthetically generated data, such as estimating or imputing missing values in existing datasets.
Source and contributor context
Pew Research Center · 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.
Synthetic data
Training or test data generated by software or another model rather than collected from the world.
Pew Research Center
Pew Research Center studies public attitudes and social change through surveys and other empirical research.
Editorial provenance
Version 1. Publication ID: be6e75aa-36bd-dc14-2bea-019d66fbda98.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-10-30T15:38:59.486Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model.
Observation: 8f451dc4e00deb4fc53e09174a872cd2082418f34486af7c447575a9f2a66ff2. Source text hash: 86bf2f500964f56912a37236e09ecb2381217927eade8e88f0551782be00c893.
passed · 2026-10-30T15:38:59.486Z · anthropic/claude-opus-5.5
Publication: be6e75aa-36bd-dc14-2bea-019d66fbda98
attribution — pass: The reading correctly attributes the report, the experiment, its findings and its conclusion to Pew Research Center. The source is a single report, not a listing, and findings are kept as the Center's reported results rather than presented as established fact.
This Pew Research Center report examines whether AI-generated survey data can accurately duplicate the results of high-quality public opinion polls.
summary — pass: The reading explains in plain language what was tested: AI adopting real panelists' personas to answer surveys. It covers the main reported issues (errors varying by question and group, avoided answer choices, overstated subgroup attitudes, reluctance to say 'not sure', model-dependent results) and the scope limit. The magnitude (about 12 points on average) and the model names are omitted, which softens the report's force, but the 'not a replacement' conclusion is preserved. The question is a fair open issue.
Across nearly 300 individual survey questions, the estimates produced using our AI respondents differed from their human counterparts by an average of 12 percentage points.
date — pass: evidence.originalPublished is 2026-09-30T17:55:17Z, so its UTC day is September 30, 2026. This matches the page dateline and the reading. The evidence kind is publisher-page, so no archive disclosure is needed. modifiedAt equals the publication instant. The reading uses no hindsight framing and presents no later events.
Report | September 30, 2026 Can AI Stand In for Human Survey-Takers? Not Really
support — pass: The source text supports each claim. This includes the persona method, differing results, avoided answer choices, stereotyped subgroup portrayals, humans being about four times as likely to choose 'not sure', models misleading in different directions, the 'not a replacement' conclusion and the survey-taker-only scope. The generalization question and the limitation note do not assert anything unsupported.
A reader of either result would be misled, but in different directions.
media — pass: No media was selected and no verified media 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.
Pew Research Center studies public attitudes and social change through surveys and other empirical research.
Explore Pew Research CenterPew Research Center studies public attitudes and social change through surveys and other empirical research.
Explore Pew Research CenterTraining or test data generated by software or another model rather than collected from the world.
Definition referencePew Research Center studies public attitudes and social change through surveys and other empirical research.
Explore Pew Research CenterPew Research Center studies public attitudes and social change through surveys and other empirical research.
Explore Pew Research CenterTraining or test data generated by software or another model rather than collected from the world.
Definition referencePew Research Center studies public attitudes and social change through surveys and other empirical research.
Explore Pew Research Center