Cross-corpus research synthesis
Deepfakes / Synthetic Content
AI-generated content disclosure, watermarking, election integrity protections.
Synthesised deterministically from 18 articles that engage this theme. Empirical consensus: contested · contested: Is robust watermarking durable under adversarial removal at deployment scale? Field is split on technical feasibility despite policy convergence on the requirement.. Full theme article: /wiki/deepfakes. Machine-readable: /wiki/synthesis.json.
Cross-jurisdiction stances (11 govern, 14 engage)
| Instrument | Verdict | Provision excerpt / citation |
|---|---|---|
| EU AI Act | governs | “Deployers of an AI system that generates or manipulates image, audio or video content constituting a deep fake, shall disclose that the content has been artificially generated or manipulated.” Art. 50(4) (disclosure obligation for deep fakes) |
| Executive Order 14110 on Safe, Secure, Trustworthy AI | governs | §4.5 (content authentication, watermarking) — rescinded 20 Jan 2025 by EO 14148; successor EO 14179 is silent on deepfakes, leaving only NIST provenance artifacts |
| Interim Measures for Generative AI Service Management | governs | Art. 12 (labelling) + Deep Synthesis Rules |
| G7 Hiroshima AI Process Code of Conduct | governs | Code §5 (content provenance + watermarking) |
| UN GA Resolution on Safe, Secure, Trustworthy AI | implicit | References disinformation broadly |
| NIST AI Risk Management Framework | implicit | Risks associated with transparency and accountability — as identified in the MAP function — are examined and documented. (paraphrase) GenAI Profile addresses synthetic content |
| NIST AI RMF Generative AI Profile | governs | NIST AI 600-1 §3.11 Confabulation + §3.10 Information Integrity (synthetic content) |
| India Digital Personal Data Protection Act + AI Advisory (MEITY) | governs | MEITY Mar-2024 Advisory + IT Rules 2021 §3(1)(b)(v) deepfake takedown obligations |
| White House Voluntary AI Commitments | governs | Commitments §5 (watermarking + content provenance for AI-generated content) |
| Singapore Model AI Governance Framework for Generative AI | governs | Framework Dimension 7 — content provenance + synthetic-content disclosure |
| California SB 942: AI Transparency Act | implicit | 'Deepfake' appears only in the SB 942 Legislative Counsel's Digest (a recital about a separate law), never in operative §§ 22757.1–22757.4; a deepfake produced by a covered provider's GenAI system is nonetheless a subset of the AI-generated image/video/audio reached by the § 22757.3(b) latent-disclosure and § 22757.2 detection duties |
| Provisions on the Administration of Deep Synthesis of Internet Information Services | governs | “深度合成服务提供者提供以下深度合成服务……应当在生成或者编辑的信息内容的合理位置、区域进行显著标识，向公众提示深度合成情况：……（三）人脸生成、人脸替换、人脸操控、姿态操控等人物图像、视频生成或者显著改变个人身份特征的编辑服务” Art. 17 |
| TAKE IT DOWN Act (Tools to Address Known Exploitation by Immobilizing Technological Deepfakes on Websites and Networks Act) | governs | 'Digital forgery' [is an intimate visual depiction] created through the use of software, machine learning, artificial intelligence, or any other computer-generated or technological means … indistinguishable from an authentic visual depiction. (paraphrase) Pub. L. 119-12 — criminalizes nonconsensual intimate 'digital forgeries' (AI deepfakes) of adults and minors and requires covered platforms to remove them within 48 hours; the statute names 'artificial intelligence' in its operative digital-forgery definition |
| Italy Law No. 132/2025 on Artificial Intelligence (Legge 23 settembre 2025, n. 132) | governs | “«Art. 612-quater … Chiunque cagiona un danno ingiusto … diffondendo, senza il suo consenso, immagini, video o voci falsificati o alterati mediante l'impiego di sistemi di intelligenza artificiale … è punito con la reclusione da uno a cinque anni.»” Art. 26(1)(c) inserts new Criminal Code Art. 612-quater: illicit dissemination of AI-generated or altered images/video/voices, without consent, apt to deceive and causing unjust harm — 1 to 5 years' imprisonment (querela-based; ex officio in aggravated cases). |
Evidence convergence
Sources the corpus cites for this theme across multiple articles — a scientometric consensus signal computed from inline prose citations (the more articles independently cite a source, the more load-bearing it is for this theme). 29 sources are cited by ≥2 articles.
- 10×Audio deepfakes and the regulation of the landlords of creativity — cited by 10 articles
- 10×Missing the Mark: Adoption of Watermarking for Generative AI Systems in Practice and Implications Under the New EU AI Act — cited by 10 articles
- 10×An interdisciplinary account of the terminological choices by EU policymakers ahead of the final agreement on the AI Act: AI system, general purpose AI system, foundation model, and generative AI — cited by 10 articles
- 10×A Teleological Interpretation of the Definition of DeepFakes in the EU Artificial Intelligence Act—A Purpose-Based Approach to Potential Problems With the Word 'Existing' — cited by 10 articles
- 8×The Current Landscape of Deepfake Legislation in the United States — cited by 8 articles
- 7×The EU model of AI governance: regulating artificial intelligence through law and policy — cited by 7 articles
- 6×Generative AI and data protection — cited by 6 articles
- 6×Artificial intelligence and synthetic biology: biosecurity risks, dual-use concerns, and governance pathways — cited by 6 articles
- 6×Navigating China's regulatory approach to generative artificial intelligence and large language models — cited by 6 articles
- 4×Human detection of political speech deepfakes across transcripts, audio, and video — cited by 4 articles
- 4×Reimagining U.S. Tort Law for Deepfake Harms: Comparative Insights from China and Singapore — cited by 4 articles
- 4×Defending Compute Thresholds Against Legal Loopholes — cited by 4 articles
- 3×Infrastructure for AI Agents — cited by 3 articles
- 3×AI, Climate, and Regulation: From Data Centers to the AI Act — cited by 3 articles
- 3×European ambitions captured by American clouds: digital sovereignty through Gaia-X? — cited by 3 articles
- 3×Open Foundation Models and TDM Exceptions to Copyright – Building Blocks for an AI Ecosystem — cited by 3 articles
- 3×When non-consensual intimate deepfakes go viral: The insufficiency of the UK Online Safety Act — cited by 3 articles
- 3×GPTs are GPTs: Labor market impact potential of LLMs — cited by 3 articles
- 3×AI or Your Lying Eyes: Some Shortcomings of Artificially Intelligent Deepfake Detectors — cited by 3 articles
- 3×'Sora is incredible and scary': public perceptions and governance challenges of text-to-video generative AI models — cited by 3 articles