Copywriting Danish People Against Deepfakes

The Danish government is planning a major legal shift to let people claim copyright over their own body, facial features, and voice, in what it says is the first European attempt to systematically tackle the threat posed by deepfakes.

A Legal Response to a Rapidly Growing Threat

Deepfakes, which are highly realistic synthetic media generated using artificial intelligence (AI), have become one of the most pressing digital threats of the past five years. By mimicking a person’s appearance, voice, and movements, these AI-generated videos, images or audio clips can convincingly impersonate individuals without their consent. Initially used for novelty and satire, they’re increasingly tied to malicious uses including fraud, harassment, and disinformation.

Massive Rise

According to a 2024 report from cybersecurity firm Sumsub, the number of detected deepfake videos worldwide rose by over a massive 700 per cent in a single year, with Europe seeing the sharpest spike. Consequently, the European Union’s law enforcement agency, Europol, has warned that deepfakes are “a significant threat to democracy and trust in institutions,” particularly around elections and public figures. However, individuals are also at risk, e.g. from revenge porn to financial scams where a cloned voice is used to impersonate a relative or company executive.

While many countries are beginning to introduce narrow legislation to deal with specific uses of deepfakes, Denmark is now attempting something broader.

What Denmark Is Proposing

Under the new proposals announced by Denmark’s Ministry of Culture in late June 2025, citizens would be granted copyright over their physical appearance, voice, and other personal traits. The hope is that this would allow them to demand the removal of AI-generated content that imitates them without permission (regardless of context) and seek compensation where harm has occurred.

Treated As A Creative Work

One important aspect of this new legal approach is that it would not rely on proving defamation or reputational damage, as is often required under existing European law. Instead, it would actually treat a person’s likeness as a creative work, similar to how a photograph or piece of music is protected. The law would apply to both private individuals and public figures, including artists and performers.

Culture Minister Jakob Engel-Schmidt described the legislation as a “bold step to protect personal identity in the age of AI,” noting that current legal protections lag behind technical capabilities. “Human beings can be run through the digital copy machine and misused for all sorts of purposes,” he said in a statement. “We are not willing to accept that.”

Timing, Process and Political Backing

The proposed changes will be submitted for public consultation before the Danish parliament breaks for summer recess, with formal legislation expected to be introduced in the autumn. Given the political climate, it’s highly likely to pass. For example, around 90 per cent of MPs reportedly support the reform, following widespread concern about the use of AI-generated content in political misinformation and online abuse.

Would Be A European First

The law would make Denmark the first European country to explicitly codify individual ownership of biometric traits for the purpose of combatting generative AI misuse. It is expected to take effect in early 2026 if passed.

What It Means in Practice

If enacted, the law would essentially give Danes the legal right to request takedowns of deepfake content from online platforms if it replicates their image, voice or body in a “realistic, digitally generated imitation.” The rule would apply whether or not the content was created with malicious intent.

Platforms that fail to comply with takedown requests could face “severe fines,” according to Engel-Schmidt. There’s also potential for EU-level action if enforcement proves challenging, particularly during Denmark’s upcoming EU presidency in 2026, when it plans to raise the issue with member states.

Includes Key Exceptions

Crucially, the proposal includes exceptions for parody and satire, which are protected under free expression rules. These carve-outs are intended to ensure that political cartoonists, satirical shows, and legitimate artistic works aren’t caught by the law.

Performances Too

The reform would also extend to artists’ performances. For example, musicians would have legal grounds to object if their voice or performance style is cloned by AI without consent, which has been a growing concern in the music industry as AI-generated songs imitate the voices of famous performers.

Why Businesses and Platforms Should Take Note

For technology companies, particularly those that operate online platforms or generate AI models, Denmark’s proposal could have far-reaching consequences.

In practical terms, businesses hosting user-generated content, such as social media platforms, image generators, or AI voice apps, may soon be legally obligated to implement mechanisms for recognising and responding to takedown requests based on biometric misuse. This could involve new detection systems, moderation processes, and audit trails to demonstrate compliance.

It also raises questions around liability. Under current EU law, platforms benefit from limited liability for illegal content they host, provided they act promptly when notified. Denmark’s new copyright-based approach might test the limits of that framework, especially if it leads to conflicts over enforcement or definitions of consent.

For creative industries, including advertising, film, and gaming, the law could restrict the use of AI tools trained on real individuals without licensing agreements. While this may increase costs and licensing complexity, supporters argue it could also encourage more ethical use of synthetic media.

From a business reputation standpoint, being seen to respect biometric rights could become a key trust signal for users and customers. A 2023 survey by the European Commission found that 79 per cent of EU citizens want stronger legal safeguards on the use of AI-generated likenesses.

How Other Countries Are Approaching the Issue

Globally, it seems, few countries have gone as far as Denmark is proposing, but some are moving in the same direction.

For example, in the United States, several states have passed deepfake-specific laws, mostly focused on election interference and non-consensual pornography. California, Texas, and New York, for instance, have made it illegal to create or distribute deepfakes that impersonate political candidates within 30 to 60 days of an election. However, there is no federal law yet, and a new budget proposal being debated in Congress could strip states of their authority to regulate AI for 10 years.

In China, deepfake creators must label synthetic media clearly and obtain consent from the people being replicated. Failure to comply can result in heavy fines. South Korea is also considering similar legislation, particularly to address deepfake abuse in online pornography, which has become a major social issue there.

Within Europe, the EU’s AI Act (adopted in 2024) includes provisions requiring deepfakes to be labelled as such, but it does not go as far as granting individuals copyright over their features. That’s why Denmark’s move is seen as a potential model for broader reforms.

What Challenges Remain?

Despite strong domestic support, Denmark’s proposal is not without critics. For example, some legal scholars have raised questions about how biometric copyright would be enforced across borders, especially on platforms based outside the EU. Others argue that tying personal identity to copyright, a system traditionally designed to protect creative works, may lead to unintended legal consequences.

There are also practical concerns, e.g. identifying a deepfake is not always straightforward, and takedown systems are often slow or ineffective. If enforcement relies heavily on users flagging violations, the burden may fall disproportionately on individuals without the resources or knowledge to pursue their rights.

For now, however, Denmark appears determined to lead the way by betting that stronger individual protections are the only way to restore trust in a digital landscape where seeing is no longer believing.

What Does This Mean For Your Business?

If Denmark succeeds in passing this reform, it could change how personal identity is treated under copyright law, not just nationally, but across Europe. By legally enshrining the right to control one’s own voice, face, and likeness, the country is effectively trying to redraw the boundary between creative freedom and personal protection in the age of synthetic media. For individuals, this could offer an unprecedented tool to fight back against misuse, without needing to prove reputational harm or navigate complex defamation law.

For UK businesses, particularly those in tech, media, and advertising, Denmark’s approach may offer a glimpse of what’s to come. If other EU countries follow suit, companies that operate across borders could face new compliance demands, from biometric consent processes to proactive takedown mechanisms. At the same time, businesses that adopt strong safeguards now, such as consent-driven AI use policies, may gain a competitive advantage by building trust with customers and clients. For those in the creative sector, for example, the move could also help clarify the grey area around training AI models on real human traits, especially in performance-heavy fields like music, voiceover, or influencer marketing.

However, enforcement remains a key challenge. For example, without international alignment, cross-border takedowns could prove difficult, and smaller platforms may struggle to implement the necessary safeguards. There’s also a risk that applying copyright principles to human identity could lead to unintended consequences, particularly if courts are left to interpret the balance between personal rights and creative expression.

Even so, Denmark’s proposed law appears to reflect a broader global reckoning with the risks of generative AI. It signals that governments are no longer willing to let platforms set the terms of engagement when it comes to biometric misuse. With deepfakes set to become more sophisticated and widespread, that signal may be just as important as the legal details that follow.