top of page
O7znn6ebAwONnNr7gaSg--h29t1.jpg

COUNTRIES REGULATING AI 'ART'

Policy Watch SA explores regulatory developments in the US, EU, and other 'western'-oriented countries that are largely English-speaking, and where AI 'art' policies and laws are already either in place or works in progress

 

  • "the US has no comprehensive federal law regulating NSFW AI*, but state-level laws have begun to address deepfakes, digital likeness rights, and AI-generated adult material" (AppsUK)

  • under the EU’s AI Act, "explicit content generators may be classified as high-risk AI, requiring strict documentation and transparency" (AppsUK)

  • the EU's Digital Services Act applies to "all digital intermediary services", including hosting services, large online platforms and social networks, and large online marketplaces, pornographic platforms, app stores and search engines (Wikipedia)

  • in terms of the UK’s Online Safety Act, "platforms are required to proactively remove harmful content, potentially using NSFW AI tools" (AppsUK) (according to The Register, this Act "applies to search services and services that allow users to post content online or to interact with each other")

  • under Australia's Classification of Media Content Act, it is "reasonable" to expect platforms hosting "millions of hours of user-generated content" to have in place "processes to readily identify adult content after it has been published" and to have mechanisms that "allow users to ‘flag’ content as adult or ‘inappropriate’" (Australian Law Reform Commission)

adult/NSFW content

  • 'nudifying apps' are a growing threat to women and girls

  • will the Online Safety Act be enough? 

  • Lewis Goodall on LBC

  • 'western'-oriented countries where AI-specific adult/NSFW content policy and regulation are still works in progress include Canada (Defend Dignity) and New Zealand (Regulating AI) 

 

*NSFW/adult content is not in itself deepfake but may well be used in the process of generating it (as indicated in the section below, a new federal law in the US – the Take It Down Act  – has been on the statute books since May 2025)​

the NSFW horse has bolted (is it too late to close the gate?)

deepfakes 

images or recordings that have been "convincingly altered and manipulated to misrepresent someone as doing or saying something that was not actually done

or said" (Merriam Webster)

  • under US federal law, the Take It Down Act "criminalises knowingly publishing or threatening to publish non-consensual intimate imagery, including AI-generated deepfakes" (Gabe Regan in Reality Defender) 

  • in Italy "a comprehensive law regulating the use of artificial intelligence (AI)" imposes prison terms on anyone using the technology "to cause harm", including "generating deepfakes" (The Guardian)

bottom of page