Truace
Subscribe

Quality perceptions and intended engagement in response to AI-generated and AI-assisted news

The increasing use of artificial intelligence (AI) in news production raises important questions about how audiences perceive and respond to AI-generated journalism. This preregistered survey experiment (N = 599, German-speaking Switzerland) examines (i) perceptions of article quality (measured as credibility, readability, and expertise) across news excerpts that were human-written, AI-assisted, or fully AI-generated, and (ii) self-reported intentions to engage following disclosure of AI involvement. Participant…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-0696· cite & verify
Quality perceptions and intended engagement in response to AI-generated and AI-assisted news
"World War II Short Sunderland flying boats - mystery location" by whatsthatpicture is marked with Public Domain Mark 1.0. To view the terms, visit https://creativecommons.org/publicdomain/mark/1.0/.

In brief

A preregistered survey experiment in German-speaking Switzerland with 599 participants tested how audiences perceive news excerpts described as human-written, AI-assisted, or fully AI-generated. Participants first rated quality without knowing the production method, then learned how their excerpts were produced and reported engagement intentions.

The comparable quality ratings suggest AI-generated news can meet audience quality expectations in this context, while the brief lift in continued reading after disclosure without a change in longer-term intentions indicates curiosity does not yet translate into sustained preference. Uncertainty remains about actual reading behavior, longer articles, and audiences outside this regional sample.

Main points

  1. Preregistered survey experiment with N = 599 participants in German-speaking Switzerland.
  2. Compared perceptions of credibility, readability, and expertise for human-written, AI-assisted, and fully AI-generated news excerpts.
  3. Participants rated two short excerpts before learning production method, then reported engagement intentions after disclosure.
  4. Future willingness to read AI-generated news did not differ across conditions after disclosure.

The gain

In a Swiss survey experiment, readers judged AI-assisted and fully AI-generated news excerpts as comparable to human-written excerpts on credibility, readability and expertise, and reported higher immediate willingness to keep reading after AI involvement was disclosed.

The rundown

The study randomly assigned participants to human-written, AI-assisted, or fully AI-generated conditions and had them rate two short news excerpts on credibility, readability, and expertise before any disclosure of production method.

After disclosure, the AI conditions showed higher self-reported intention to continue reading the assigned articles versus control, while self-reported future willingness to read AI-generated news showed no difference across conditions.

Sources

  1. Peer-reviewedScientific Reports2026-06-29

The debate