Organizations that fail to comply could face penalties of up to €15 million or 3% of their worldwide annual turnover. The European Commission has published a set of black-and-white disclosure symbols that companies can use, although businesses are also permitted to create their own labels.
The rules are intended to address growing concerns about deceptive synthetic media, particularly deepfakes that imitate real people or events. Recent examples have included fabricated audio appearing to show Slovak opposition leader Michal Šimečka discussing election manipulation, altered videos depicting French President Emmanuel Macron dancing in a 1980s nightclub, and fake audio of former UK Labour leader Keir Starmer verbally abusing staff.
"It is a matter not only of customer protection, it’s also a matter of democracy protection," Green Member of the European Parliament Sergey Lagodinsky, who helped negotiate the AI Act, told The Guardian. "Making transparent this information is something which we need to preserve our democracy and the authenticity of facts online."
The transparency requirements exempt AI content created solely for personal use, while "evidently artistic", satirical, and fictional works are also excluded. Companies are encouraged, but not required, to label AI-generated material created before the new rules took effect.
The obligations extend beyond standalone AI products and cover a broad range of digital services that generate or modify content using AI. That wider scope has fueled concerns among technology companies that users could become overwhelmed by constant disclosures as AI becomes embedded across everyday software.
Industry groups argue that guidance published by the European Union has broadened the practical reach of the rules beyond what many companies expected. Boniface de Champris, AI policy lead at the Computer and Communications Industry Association Europe, said the requirements risk treating significantly different uses of AI in the same way.
"The label was meant to flag deceptive content," de Champris said. "That distinction has been removed, so now almost everything gets labelled – a landscape in an advert is the same as a manipulated political speech. Like cookie banners, once labels are everywhere, users stop noticing them. That doesn’t protect anyone, it just burdens commercial activity across the board."
He said consumers are likely to notice the biggest changes in industries where AI already operates behind the scenes, including advertising, film, and publishing, rather than on social media platforms that already disclose much AI-generated content.
Several major technology companies have already introduced AI labeling tools ahead of the regulation. TikTok requires creators to disclose realistic AI-generated images, audio, and video, and says more than three billion pieces of content have been labeled through its systems. Google introduced SynthID in 2023 and says the technology has embedded invisible digital watermarks into more than 100 billion images and 60,000 years' worth of audio. Meta has also rolled out policies to identify AI-generated photorealistic content and has worked on technical standards for AI audio and video.
Google and Meta are among the companies that have signed the European Union's voluntary AI code of practice, which was developed under the supervision of the EU AI Office. According to the European Commission, more than 180 organizations have joined the code, while companies that do not participate must demonstrate compliance through other means.
Even companies supporting AI transparency have warned that excessive disclosure requirements could reduce their effectiveness. Announcing Meta's decision to sign the code of practice, the company's vice-president for public policy in Europe, Markus Reinisch, said: "We must avoid a situation where online content is subject to a growing array of different labels and disclosures that end up overwhelming people, while at the same time adding additional regulatory complexity for providers."
Lagodinsky argued that implementation challenges can be overcome. "I have a deep respect for industry, but in all too many cases, they complain, and … then after things are implemented, it doesn’t look that burdensome to them. And there is always a solution."
The transparency obligations mark another stage in the broader rollout of the EU AI Act, which introduces requirements in phases. European Commission Executive Vice-President Henna Virkkunen described artificial intelligence as a "transformative technology" capable of delivering "extraordinary benefits," while warning that the most advanced AI models "create risks on an entirely new scale."
"Europe anticipated this development. With the AI Act, we established a clear, risk-based and durable framework for trustworthy AI —one that gives innovators legal certainty while protecting the public interest."
This analysis is based on reporting from The Guardian.
Image courtesy of The Decoder.
This article was generated with AI assistance and reviewed for accuracy and quality.
