Deepfakes and the Law: Emerging Legal Protections

Deepfakes challenge our sense of reality by appearing startlingly real, making it increasingly difficult to separate fact from fabrication. A deepfake is a digitally manipulated video, audio or image that appears real, but is actually generated or altered by artificial intelligence (AI).

Deepfakes pose significant legal and ethical challenges within the realm of cybersecurity. These challenges arise from the unethical use of deepfake technology to generate content involving individuals without their consent, often with the intent to cause harm. For example, deepfake technology has been weaponized to create and spread non-consensual pornography, disproportionately targeting women.

Evolving Legal Responses to Deepfakes

A few jurisdictions are taking legal steps to address the misuse of deepfake technology. For example, in the United States, Congress passed the Take It Down Act in the spring, which criminalizes the nonconsensual sharing of intimate images under certain circumstances, like digitally manipulated content such as deepfakes. Likewise, the European Union’s AI Act, which came into effect in August 2024, requires deepfake content be clearly labeled as artificially generated or manipulated.

Denmark’s Proposal to Combat Deepfakes

Recently, the Danish Minister of Culture announced a proposed amendment to Denmark’s copyright law intended to combat the growing issue of deepfakes.

Should the proposal take effect, it would grant individuals copyright-like protections over their own image, voice and likeness, enabling them to request the removal of unauthorized digital reproductions and pursue compensation for misuse. Under the proposal, protections would apply for the duration of the subject’s lifetime and continue for 50 years after their death. In line with standard copyright exceptions, Danish law would continue to allow the use of deepfakes for parody, satire, and caricature.

Arguably, Denmark’s proposal surpasses efforts in other jurisdictions to combat deepfake misuse by substantially expanding the rights of individuals. Rather than focusing solely on the creators and consumers of deepfakes, the proposal empowers the individuals depicted, granting them the ability to pursue legal action, demand the removal of unauthorized content and seek compensation for misuse. This marks a shift from criminalizing deepfake creation, as seen in the US, toward a civil framework that provides victims with enforceable remedies.

Limited Legal Protections for Deepfake Victims in Canada

As of now, Canada does not recognize copyright protection for personality rights, so deepfake victims must rely on common law claims. Privacy law in some provinces protects against unauthorized use of a person’s name, likeness or voice, while trademark law prohibits using a person’s portrait or signature in a business context. The tort of passing off may apply where a deepfake falsely suggests an individual’s endorsement or affiliation with a product. The tort of misappropriation of personality protects an individual’s right to control the commercial use of their personality rights. Interestingly, the tort of false light, which is recognized only in Ontario and has been explicitly rejected in Nova Scotia, applies when a person is portrayed falsely in a highly offensive way with knowledge or reckless disregard of the truth.

Although these legal remedies may offer some protection against deepfake content, their effectiveness is limited. To date, no notable civil deepfake case has reached the courts in Canada, leaving this a developing and unsettled area of law.

With respect to criminal law, in November 2025, Ontario Court Justice Brian Puddington ruled that distributing a fake nude image, while “morally reprehensible,” was not a crime under the Criminal Code.