· May 27, 2026
Anthropic reports uneven adoption of coding agents among social scientists
Markdown · JSON · Original source
On May 27, 2026, reports findings from a survey of quantitative social scientists about AI use in research. The survey found that many respondents had tried AI chatbots, while fewer regularly used coding agents: tools that can write and run analysis code. Adoption varied by career stage, discipline, gender classification based on names, and university status. Anthropic says coding agent users reported more project starts, working papers, and grant proposals than other researchers in the same discipline and career stage. The comparison is descriptive, not evidence that the tools caused higher output: early adopters may already have been more productive. The survey found no evidence that coding agent users submitted more papers to journals or resubmitted them more quickly, and it did not assess the quality of reported work. Respondents were more optimistic about AI’s effect on writing publishable papers than its effect on social science overall. Anthropic notes that the survey recruited researchers for a study offering access to Max accounts, so respondents may have been more interested in AI than non-respondents. The authors say a randomized experiment is ongoing, with results planned for a future report. What will the ongoing experiment find about whether coding agents change research output, beyond differences between early adopters and other researchers?
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: May 27, 2026 (2026-05-27T00:00:00.000Z).
Date basis: Date stated on the publisher’s page.
Added to Escape Meridian: 2026-10-30T21:09:00.645Z. First observed: 2026-10-20T05:43:20.878Z. Retrieved: 2026-10-28T04:51:49.641Z.
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 28, 2026; originally published May 27, 2026. The survey was not representative; respondents were recruited for a study offering access to Max accounts. The reported differences in research output are descriptive, not causal, and the report does not assess the quality of researchers’ work. The source says the randomized experiment is ongoing; its results are not included in this report. The retrieved page was edited after publication; later-added passages may not reflect the original version.
Open question
What will the ongoing experiment find about whether coding agents change research output, beyond differences between early adopters and other researchers?
Source-stated
Anthropic reports that the survey found fewer respondents regularly used coding agents than had tried AI chatbots for research.
The vast majority of respondents (81%) have tried using AI chatbots in research, particularly for writing code and editing prose. But only 20% have adopted coding agents
Source-stated
Anthropic reports that coding agent users reported more early-stage research output, but the comparison does not establish that coding agents caused the difference.
Differences should not be interpreted as causal, but as a first cut comparison between researchers using coding agents and those who are not.
Source-stated
Anthropic reports no evidence that coding agent users submitted more new papers to journals or resubmitted papers more quickly.
We find no evidence that coding agent users are submitting more new papers to journals or resubmitting papers more quickly.
Source-stated
Anthropic says the survey was not representative and that respondents may have been more interested in AI than non-responders.
This was not a representative sample—respondents were recruited for a study that offered access to Claude Max accounts, so selection into the sample could tilt toward researchers curious about AI tools.
Source-stated
Anthropic says the ongoing study includes a randomized experiment providing researchers with access to Claude Code, with results planned for the future.
The survey is the baseline wave of a larger ongoing study of how coding agents affect research productivity, including a randomized experiment providing researchers with access to Claude Code. We will publish results from this experiment in the future.
Source and contributor context
Anthropic · 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.
Anthropic
Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Claude
Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Editorial provenance
Version 1. Publication ID: c740ea03-43b5-30f7-cc90-4694c4cd97c3.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-10-30T21:08:57.673Z. 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: bf2a90c4053526701fdc5a94b6ca6f8f0bbefe5e82ca27e2aca2b3a8c7037433. Source text hash: 771ce27229f09782a464205b6acde2ce45b8ed4e7847c46de188dd195bc95cc3.
Revision and source-check history
Version 1 · 2026-10-30T21:09:00.645Z
Anthropic reports uneven adoption of coding agents among social scientists
On May 27, 2026, reports findings from a survey of quantitative social scientists about AI use in research. The survey found that many respondents had tried AI chatbots, while fewer regularly used coding agents: tools that can write and run analysis code. Adoption varied by career stage, discipline, gender classification based on names, and university status. Anthropic says coding agent users reported more project starts, working papers, and grant proposals than other researchers in the same discipline and career stage. The comparison is descriptive, not evidence that the tools caused higher output: early adopters may already have been more productive. The survey found no evidence that coding agent users submitted more papers to journals or resubmitted them more quickly, and it did not assess the quality of reported work. Respondents were more optimistic about AI’s effect on writing publishable papers than its effect on social science overall. Anthropic notes that the survey recruited researchers for a study offering access to Max accounts, so respondents may have been more interested in AI than non-respondents. The authors say a randomized experiment is ongoing, with results planned for a future report. What will the ongoing experiment find about whether coding agents change research output, beyond differences between early adopters and other researchers?
Original source
· Article
Source details
Original source date: May 27, 2026 (2026-05-27T00:00:00.000Z).
Date basis: Date stated on the publisher’s page.
Added to Escape Meridian: 2026-10-30T21:09:00.645Z. First observed: 2026-10-20T05:43:20.878Z. Retrieved: 2026-10-28T04:51:49.641Z.
Limits
Text as retrieved on October 28, 2026; originally published May 27, 2026. The survey was not representative; respondents were recruited for a study offering access to Max accounts. The reported differences in research output are descriptive, not causal, and the report does not assess the quality of researchers’ work. The source says the randomized experiment is ongoing; its results are not included in this report. The retrieved page was edited after publication; later-added passages may not reflect the original version.
Open question
What will the ongoing experiment find about whether coding agents change research output, beyond differences between early adopters and other researchers?
Source-stated
Anthropic reports that the survey found fewer respondents regularly used coding agents than had tried AI chatbots for research.
The vast majority of respondents (81%) have tried using AI chatbots in research, particularly for writing code and editing prose. But only 20% have adopted coding agents
Source-stated
Anthropic reports that coding agent users reported more early-stage research output, but the comparison does not establish that coding agents caused the difference.
Differences should not be interpreted as causal, but as a first cut comparison between researchers using coding agents and those who are not.
Source-stated
Anthropic reports no evidence that coding agent users submitted more new papers to journals or resubmitted papers more quickly.
We find no evidence that coding agent users are submitting more new papers to journals or resubmitting papers more quickly.
Source-stated
Anthropic says the survey was not representative and that respondents may have been more interested in AI than non-responders.
This was not a representative sample—respondents were recruited for a study that offered access to Claude Max accounts, so selection into the sample could tilt toward researchers curious about AI tools.
Source-stated
Anthropic says the ongoing study includes a randomized experiment providing researchers with access to Claude Code, with results planned for the future.
The survey is the baseline wave of a larger ongoing study of how coding agents affect research productivity, including a randomized experiment providing researchers with access to Claude Code. We will publish results from this experiment in the future.
Source and contributor context
Anthropic · 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.
Anthropic
Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Claude
Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Editorial provenance
Version 1. Publication ID: c740ea03-43b5-30f7-cc90-4694c4cd97c3.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-10-30T21:08:57.673Z. 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: bf2a90c4053526701fdc5a94b6ca6f8f0bbefe5e82ca27e2aca2b3a8c7037433. Source text hash: 771ce27229f09782a464205b6acde2ce45b8ed4e7847c46de188dd195bc95cc3.
passed · 2026-10-30T21:08:57.673Z · anthropic/claude-opus-5.5
Publication: c740ea03-43b5-30f7-cc90-4694c4cd97c3
attribution — pass: The reading attributes the survey and its findings to Anthropic, the publisher of this single research post. Survey results and interpretations are framed as Anthropic's reports, not as established facts.
We present results from a survey of 1,260 social scientists about AI and coding agent use, fielded in February and March 2026.
summary — pass: The reading leads with the actor and the development. It explains coding agents in plain language and covers uneven adoption, early-pipeline output differences with the causal caveat, the lack of any journal-submission difference, the unassessed quality, the optimism gap, the sampling limits and the pending experiment. The question follows from the evidence. It omits gap magnitudes, such as the twofold gender gap, but that does not distort the meaning.
Twice as many researchers with typically male names use coding agents as those with female names.
date — pass: May 27, 2026 matches the UTC calendar day of originalPublished. The page was edited after publication, but the reading uses no later events and frames the experiment as planned, not as an outcome. Its limitations note the later edit. This is a publisher page, so no archive disclosure is needed.
But as of March 2026, they are not yet driving a surge in journal submissions.
support — pass: Each claim matches the source text: the 81% versus 20% gap, the descriptive (non-causal) comparison, no evidence of more journal submissions, the non-representative sample and the planned randomized experiment. The cited passages form the core body of the post, which describes early 2026 data. Nothing suggests they are later additions, though the page was edited after publication.
Finally, we only look here at the number of projects researchers report, and report nothing about their quality.
media — pass: No media was selected, and no verified media candidates were supplied, so no media identity or relevance issue arises.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore Claude