The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2712
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T17:13:18.626657Z
status: published
lens: p_space
sector: entertainment
headline: Synthetic News, Natural Doubts? A Meta-Analysis of Credibility Perceptions of AI-Generated News
dek: As generative artificial intelligence (AI) becomes increasingly integrated into news production, concerns about the credibility of synthetic news content are intensifying. While some studies report that audiences find AI-generated news less credible than human-written content, others reveal mixed or even reversed findings. This meta-analysis synthesizes findings across 31 studies (41 effect sizes) to assess how the source label (AI vs. human) influences news users' evaluations of source credibility and message c…
gain_title: (none)
problem_title: A meta-analysis of 31 studies (41 effect sizes) found news labeled as AI-generated got a small but statistically significant credibility penalty, on both source and message credibility, compared with news labeled as human-written.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: A meta-analysis of 31 studies (41 effect sizes) found news labeled as AI-generated got a small but statistically significant credibility penalty, on both source and message credibility, compared with news labeled as human-written.
problem_evidence: Results show a small but statistically significant penalty for AI-labeled (vs. human-labeled) news on both credibility measures.
quick_read: A meta-analysis pooled 31 studies, yielding 41 effect sizes, to test whether telling readers that a news article came from AI rather than a human changes how credible they find it. It found a small but statistically significant penalty for AI-labeled news, both for the credibility of the source and of the message.

Earlier studies disagreed: some found audiences trusted AI-generated news less, while others found mixed or even reversed results. Pooling them gives newsrooms using generative AI a clearer picture of how disclosure labels may affect audience trust.

The researchers tested three explanations for the variation between studies: how value-laden the topic was, participants' source orientation, and who actually wrote the text. Only the last was statistically significant, with the penalty larger when the articles were in fact written by humans only rather than by AI.
limitation: Two of the three moderators, topic value-ladenness and participants' source orientation, showed only descriptive trends and did not reach statistical significance, so the sources of variation between studies remain only partly explained.
tag: Evidence-backed problem
key_points: The analysis covers 31 studies and 41 effect sizes on how an AI versus human source label affects news credibility. | AI-labeled news received a small but statistically significant credibility penalty on both source credibility and message credibility. | Three moderators were examined: topic value-ladenness, participants' source orientation, and actual authorship of the text. | Topic value-ladenness and source orientation showed descriptive trends in the expected direction but were not statistically significant. | Actual authorship was the only significant moderator: the penalty was larger when the articles were written by humans only rather than by AI.
rundown: The paper, published in Cyberpsychology, Behavior, and Social Networking in April 2026, synthesizes experimental evidence on audience reactions to AI-generated news. It compares the source label, AI or human, and measures effects on two outcomes: how credible readers find the source and how credible they find the message.

Individual studies had pointed in different directions, with some showing lower credibility for AI-generated news and others showing mixed or reversed results. The meta-analysis reports an overall small but statistically significant penalty for AI labels.

To explain the variation, the authors tested topic value-ladenness, participants' source orientation, and actual authorship. Only authorship reached significance: the label penalty was more pronounced for articles actually written by humans only, compared with AI-written articles. The authors say this clarifies when and why audiences mind the source.
sources:
- peer_reviewed | Cyberpsychology, Behavior, and Social Networking | https://doi.org/10.1177/21522715261439452 | 2026-04-07
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

8c88abd77e7689a42c562bd3bb1a2f3c9203b42a9528b598a508c6c8ce9a6417

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000