Denmark – Landmark Bill: Proposal to Regulate AI-Deepfakes Under Copyright Law

Background 

Denmark’s Government, on the initiative of the Minister of Culture, proposed a groundbreaking amendment to its Copyright Act, with the primary focus on protecting its nation from AI-generated deepfakes that replicate the personal characteristics of its citizens.

AI-generated deepfakes, digital imitations, that replice persons physical characteristics, voice, and other characteristics, when used without a consent should be banned, and the authors of this replicas or platforms that miss to develop systems that prevent creation, distribution, or deletion of the digital material that is labelled that belongs to this category should also face fines, according to the approach of the supporters of this amendments. 

In recent years, the production of deepfakes, realistic AI-generated images, videos, or audio files that accurately impersonate individuals or events, has bloomed. The AI technology and related AI tools used to create deepfakes are everywhere, and their ease of use makes deepfakes popular and widely used, without the need for “special training or education”. 

The continuous enrichment of OpenAI, Google, and the like in their AI tool-supported portfolio undoubtedly helps their users raise the quality of deepfakes and, inevitably, their later distribution. Given that the use of AI-generated deepfakes will only grow, without any indication that it will slow, the Danish government decided it is time to act. 

In a practical sense, it is time to regulate and to establish additional specific regulatory requirements regarding the use of AI tools that produce and distribute deepfakes. The Government hopes the submitted bill to amend the Copyright Act will be adopted in 2026. 

If the bill passes into law, it will, in these times, give Danish citizens a much-needed means to demand the removal of unauthorized content from online platforms. At the moment, Denmark is one of the few countries in the world working this hard to pass this kind of legislation. 

The present regulatory framework in Denmark, as well as in other countries, provides little or no protection against AI-generated deepfakes. The scarcity of legal instruments to combat the production and distribution of deepfakes is something that the Danish government is looking to change, at least in its own backyard.

Denmark – Deepfakes and Push for an AI Regulatory Framework

Legal background 

Denmark belongs to the group of countries that are very “progressive” when it comes to “driving through regulation” of the effects of the AI-bubble. In June 2024, Danish political parties reached a cross-party agreement on the overall terms for the use of AI in political activities.

On June 26, 2025, Denmark’s Minister of Culture presented before the Parliament the “Draft Proposal for an Act to Amend the Danish Copyright Act”.

The bill, if adopted by Parliament, will introduce legal instruments to combat the unauthorised use of a person’s face, body, or voice for the creation of AI-generated deepfakes and their later distribution without the person’s consent. 

The bill specifically defines these attributes on the personal level as “protected expressions” and, as such, clearly diverges from conventional copyright frameworks. The copyright regulatory framework is mainly established to safeguard only original creative works.

If the Parliament gives the proposed bill the green light, the law could initiate a regulatory shift in how governments safeguard their citizens’ digital identities and personal rights in the digital age. 

Denmark AI regulatory framework – Proposed Amendments 

The proposed amendments submitted by the Minister of Culture to the Copyright Act have been referred to the EU Commission. The proposed amendments are organized in two parts: general protection against realistic digitally generated imitations of personal characteristics, and protection against realistic digitally generated imitations of performers. 

One of the core enforcement procedures under the proposed amendment is a 50-year protection against the use of a person’s likeness, voice, or image in creating digital imitations, with the protection extending to the person’s death. 

Using a person’s personal characteristics without prior consent is strongly forbidden. The enforcement of the amendment to the Copyright Act shall be supported through the enforcement procedures established by the Digital Services Act (DSA). 

The “modernized” Danish Copyright Act should serve as the foundation and legal basis for activating the DSA procedures against AI-deepfakes. The proposed amendments should come into force in Q3 2026. 

If the Parliament accepts the submitted amendments to the Copyright Act, the new regulatory mechanism against unauthorized use of a person’s personal traits shall apply only within Denmark’s borders. 

The draft Bill, in its current form, doesn’t include fines or prison sentences for individual social media users who use AI tools to generate deepfakes. On the other hand, under the amended Copyright Act, the responsibilities for digital platforms are quite different. 

If platform operators, such as Meta (which owns Facebook, Instagram, and others), Google, and TikTok, to name a few, fail to act expeditiously on deepfakes flagged by their users, they could face severe financial penalties. 

As Danish media outlets report, the bill has strong support among Parliament members, who are “well informed” about the “danger” that AI-generated deepfakes pose to society as a whole, not only to public figures. 

Bill and Global Impact 

Denmark’s government initiative to regulate the production and distribution of AI-generated deepfakes under the Copyright Act has attracted attention from other European countries. The France and Ireland representatives have already shared that their countries are very interested in Denmark’s initiative to establish new mechanisms for better control of social media platforms on which deepfakes circulate; AI tools that enable the development of this material and its users; as well as a pecuniary framework for enablers. 

Multinational tech enterprises are closely following these developments. 

Denmark AI Framework Under Copyright Law 

Enforcement of the Digital Services Act 

The proposed bill seeks to enhance the enforceability framework of the EU’s Digital Services Act. Under DSA, users of social media platforms have the right to demand that the platform operator remove a deepfake, only after submitting a notice accompanied by substantive evidence. 

The bill should provide harmed users with fewer administrative barriers when requesting the removal of deepfakes and, on the other hand, be preventive in nature, with a penalty system, among other measures. 

Outcome if the bill passes 

The proposed amendments to the country’s Copyright Act by the Minister of Culture represent, without exaggeration, a major legal shift in how personal identity is viewed from a legal perspective in the AI digital era. 

The bill aligns with the broader EU regulatory framework for closely connected subjects, including the Digital Services Act and the EU Artificial Intelligence Act. 

The provisions of this amendment, if adopted in the submitted format, shall grant Danish citizens exclusive rights to the digital replication of their voices, images, and personal attributes. 

Takeaway 

Frequently Asked Questions