?asOf= parameter to see the current catalog state.Labelling, watermarking, and machine-readable provenance for AI-generated audio / video / text. Distinct from `deepfakes` (which centres on misuse harms) — this is the upstream infrastructure layer. EU AIA Art. 50, China GenAI Measures Art. 13 (mandatory tagging), NIST AI 600-1, G7 Hiroshima Code commitment 6, C2PA standard adoption.
Abstract
Synthetic-content-provenance governance — how AI-generated media is watermarked, labelled, or declared so audiences can tell it apart — is addressed directly by several catalogued instruments, including the EU AI Act (Article 50) and China's labelling rules, with others reaching it only implicitly. Policy Window records the empirical consensus as contested: the open question is where the duty should sit — with the model provider watermarking at generation, the platform labelling at distribution, or the recipient able to ask — and jurisdictions are selecting different points. This article maps each instrument's provenance obligations with primary-source citations.
Definition & scope
The cross-jurisdiction picture below shows how each of 45 tracked instruments treats this topic. The patterns vary substantially — and 28 regimes are silent, leaving gaps that future policy work could address.
Coverage across jurisdictions
Historical primacy & cross-jurisdiction tension
First addressed by Provisions on the Administration of Deep Synthesis of Internet Information Services on (governs). Subsequent regimes have either codified, diverged from, or remained silent on this baseline.
- Forum-shoppingEU AI Act↔Executive Order 14179 — Removing Barriers to American Leadership in AI
- Forum-shoppingExecutive Order 14110 on Safe, Secure, Trustworthy AI↔UK Pro-Innovation Approach to AI Regulation (White Paper)
- Forum-shoppingInterim Measures for Generative AI Service Management↔OECD AI Principles (Recommendation)
Compare jurisdictions: EU vs US · EU vs UK · EU vs CN
Enforcement & impact
Silent regimes — gap signal
Instruments that do not address Synthetic Content Provenance — candidates for future policy work.
- Executive Order 14179 — Removing Barriers to American Leadership in AIUS
- UK Pro-Innovation Approach to AI Regulation (White Paper)UK
- OECD AI Principles (Recommendation)OECD
- Council of Europe Framework Convention on AIcouncil_of_europe
- Bletchley Declaration on AI Safetyglobal
- Seoul Declaration on Safe, Innovative and Inclusive AIglobal
- California SB-1047: Safe and Secure Innovation for Frontier AI Models ActUS
- India Digital Personal Data Protection Act + AI Advisory (MEITY)IN
- ASEAN Guide on AI Governance and EthicsASEAN
- African Union Continental AI StrategyAfrican_Union
- OpenAI Preparedness FrameworkUS
- Google DeepMind Frontier Safety FrameworkUS
- Meta Frontier AI FrameworkUS
- UK-US AI Safety Institute Memorandum of Understandingglobal
- General Data Protection Regulation (GDPR)EU
- OMB Memorandum M-24-10 (Advancing Governance, Innovation, and Risk Management for Agency Use of AI)US
- GSA Generative AI and Specialized Computing Infrastructure Acquisition Resource GuideUS
- DoD Responsible AI Strategy and Implementation PathwayUS
- FedRAMP AI Cloud Procurement GuidanceUS
- DFARS Subpart 252.204 (Safeguarding Covered Defense Information and Cyber Incident Reporting)US
- California SB-53: Transparency in Frontier Artificial Intelligence Act (TFAIA)US
- California SB 243: Companion ChatbotsUS
- Revised Product Liability Directive (Directive (EU) 2024/2853)EU
- UNESCO Recommendation on the Ethics of Artificial IntelligenceUNESCO
- Directive (EU) 2024/2831 on improving working conditions in platform workEU
- New York RAISE Act: Responsible AI Safety and Education ActUS
- TAKE IT DOWN Act (Tools to Address Known Exploitation by Immobilizing Technological Deepfakes on Websites and Networks Act)US
- Japan AI Promotion Act (Act on the Promotion of Research, Development and Utilization of AI-Related Technologies)JP
See also
Further reading
10 academic & grey-literature sources bearing on this topic — catalogued metadata with a primary link; one-line findings are ✦ AI-generated summaries, labeled as such (charter §7.9). Browse the full literature index.
- Missing the Mark: Adoption of Watermarking for Generative AI Systems in Practice and Implications Under the New EU AI Act Peer-reviewed✦ AIEmpirical audit finds only 38% of AI image generators implement adequate watermarking and 18% deepfake labelling, exposing a compliance gap under EU AI Act Article 50.
- Navigating China's regulatory approach to generative artificial intelligence and large language models Peer-reviewed✦ AIAnalyses China's 2022 deep-synthesis and 2023 generative-AI rules, including mandatory labelling/watermarking of synthetic content as a provenance-governance model.
- 'Sora is incredible and scary': public perceptions and governance challenges of text-to-video generative AI models Peer-reviewed✦ AIQualitative analysis of public commentary on Sora finds blurred real/fake boundaries drive demand for law-enforced AI-content labelling and provenance.
- SoK: Watermarking for AI-Generated Content Peer-reviewed✦ AISystematizes watermarking for AI content, formalizing robustness/security goals and limits that directly ground regulatory provenance and labeling mandates.
- From Principles to Practices: Lessons Learned from Applying Partnership on AI's (PAI) Synthetic Media Framework to 11 Use Cases Preprint✦ AIApplies PAI's Synthetic Media Framework to 11 real cases, finding disclosure/provenance recommendations could have mitigated harm in several 2024-election deepfake incidents.
- Examining the Impact of Provenance-Enabled Media on Trust and Accuracy Perceptions Peer-reviewed✦ AIOnline experiment (n=595) found 'provenance information often lowered trust and caused users to doubt deceptive media,' though it could similarly reduce trust in truthful media.
- Watermarks in the Sand: Impossibility of Strong Watermarking for Generative Models Preprint✦ AIProves 'under well-specified and natural assumptions, strong watermarking is impossible to achieve,' bounding what watermark mandates for generative-AI content can guarantee.
- Can AI-Generated Text be Reliably Detected? Preprint✦ AIShows AI-text detectors including watermarking are attackable: a 'recursive paraphrasing method can significantly reduce detection rates' while only slightly degrading text quality.
- Generative AI models should include detection mechanisms as a condition for public release Peer-reviewed✦ AIArgues legislation should require foundation-model developers to 'demonstrate a reliable detection mechanism for the content it generates, as a condition of its public release.'
- The Epistemic Threat of Deepfakes Peer-reviewed✦ AIArgues deepfakes pose an epistemic threat because they "reduce the amount of information that videos carry to viewers", undermining knowledge acquired from video evidence.
References
The primary instrument sources behind the article's classifications.
- EU-AIA-2024: Art. 50(2) — provider machine-readable marking obligation; Art. 50(4) — deployer disclosure for deep fakes (distinct from the `deepfakes` topic which focuses on misuse-harms)
- US-EO-14110: §4.5(a) — content authentication + watermarking standards via NIST + Commerce
- CN-GENAI-2023: Art. 12 — mandatory marking of generative-AI output; aligns with Deep Synthesis Rules (2022) tagging requirements
- G7-HIROSHIMA: Code §6 — 'develop and deploy reliable content authentication and provenance mechanisms'
- UN-RES-2024: General call for state action on safe AI; provenance not specifically addressed
- NIST-AI-RMF: General framework applies; provenance-specific guidance lives in the GenAI Profile
- NIST-AI-RMF-GENAI: NIST AI 600-1 — Information Integrity is one of 12 named GenAI risk categories; covers synthetic-content labelling + provenance
- BR-AIBILL-2024: PL 2338 general accuracy + transparency obligations would extend to provenance via interpretation
- ANTHROPIC-RSP-2024: Deployment-stage controls would include content provenance where capability tier requires
- WH-VOLUNTARY-2023: Voluntary commitment #5 — 'develop and deploy mechanisms that enable users to understand if audio or visual content is AI-generated, including robust provenance, watermarking, or both'
- SG-MODEL-AI-2024: Framework dimension 7 — Content Provenance (one of nine framework dimensions, paired with AI Verify Foundation's technical-testing toolkit)
- JP-METI-AI-2024: Principle 5 (Transparency) + Hiroshima-alignment imply provenance obligations via reference incorporation
- EU-GPAI-COP-2025: Chapter 1 transparency commitments brush against Art. 50(2) deployer marking + Art. 53(1)(a) provider documentation
- CA-SB-942: Cal. Bus. & Prof. Code § 22757.3(b) (added by SB 942) — a covered provider must embed a machine-readable 'latent' disclosure in AI-generated image/video/audio conveying provenance metadata: provider name, GenAI system name and version, creation/alteration time, and a unique identifier; reinforced by § 22757.3.1 (AB 853, operative 2027) barring large online platforms from knowingly stripping system provenance data
- CN-DEEPSYN-2022: Art. 16 & Art. 18
- IT-AILAW-2025: No standalone watermarking/provenance-marking duty in the law itself; provenance is reached only indirectly — Art. 612-quater criminalises deceptive AI-altered media (turning on whether content is apt to deceive as to genuineness) and the general transparency principle (Art. 4). Content-marking duties are left to the EU AI Act (Art. 1(2)).
- UN-GDC-2024: GDC Objective 3, para 36(c) (A/RES/79/1, Annex I)
Cite this article 8 formats · BibTeX, RIS, APA, Chicago, … · 1-click copy
Persistent identifier: https://policywindow.org/wiki/synthetic-content-provenance — committed-stable URL with content-versioning via ?asOf= (rollout pending per methodology §7). DOIs via Zenodo are on the roadmap.
Article tools — track changes, suggest an edit
View history — every captured revision of this article · What links here
17 instruments tracked.