SUMMARY Brigitte Bardot, who died in 2025, never engaged with AI, but her refusal to commercialize her image has made her a key figure in debates about AI and human identity. French law treats likeness as part of personal dignity, unlike U.S. publicity laws, which focus on commercial harm. Bardot’s legacy highlights the need for AI to respect consent and autonomy, showing that human identity is not mere raw material for technology.
Brigitte Bardot died on December 28, 2025. She never touched artificial intelligence. Yet her legacy sits at the center of some of the most important debates about how AI should treat human identity.
As generative AI systems become capable of producing realistic images, videos and voices, the question is no longer whether this can be done, but whether it should be done at all. Few public figures illustrate that tension better than Bardot.
Unlike many film icons, Bardot walked away from cinema early and spent decades actively resisting the commercialization of her image. She declined advertising offers, refused promotional retrospectives and consistently opposed unauthorized reproductions of her likeness. That refusal is not merely biographical detail; it aligns closely with how French law conceptualizes personal identity. Under the French droit à l’image, an individual’s image is treated as an extension of personality and dignity rather than as a freely tradable economic asset. French legal commentary has long noted that Bardot’s litigation and public stance helped crystallize modern image-rights doctrine, reinforcing that unauthorized reproductions may constitute a personal intrusion even in the absence of commercial harm, as discussed in “Et Bardot créa le droit à l’image” published by Actu-Juridique.
This framing has taken on renewed significance in the AI era. An AI-generated image of Bardot is not just a hypothetical licensing opportunity gone astray; it represents a technological override of an explicit, long-standing refusal. That distinction helps explain why European regulators approach synthetic identity with caution. In France, this concern has translated into legislative proposals requiring the labeling of AI-generated images on social networks and strengthening oversight of manipulated content, reflecting an emphasis on protecting reputation and human dignity rather than merely preventing consumer deception, as summarized by the U.S. Library of Congress’s Global Legal Monitor.
The Bardot example has become a useful reference point in AI governance discussions precisely because it presents a “hard case.” If AI systems are capable of convincingly reproducing the likeness of someone who has consistently rejected such use for decades, then the systems are not neutral tools. They are making implicit normative choices about consent, autonomy and the limits of technological power.
The contrast with U.S. celebrity AI practices is striking. In the United States, control over name, image and likeness is governed primarily by state right-of-publicity laws, which are largely commercial in orientation and vary widely in scope, duration and remedies. There is no unified federal framework, and enforcement typically depends on after-the-fact litigation rather than safeguards embedded in technology. Major law firms have noted that U.S. right-of-publicity claims often hinge on commercial exploitation and economic harm, leaving significant gaps in protection against nonconsensual AI replicas, as outlined in Cravath’s analysis of AI deepfakes and publicity rights.
Recent controversies illustrate this reactive posture. Unauthorized AI depictions of U.S. celebrities have circulated widely before being challenged, prompting public backlash and calls for regulation. When an AI-generated video resembling Scarlett Johansson went viral without her consent, Johansson publicly urged lawmakers to impose limits on AI, emphasizing how current law forces individuals to respond only after harm has already occurred, as reported by People magazine. Similar issues have arisen with AI-generated chatbots and endorsements. A Reuters investigation revealed that Meta created conversational AI personas modeled on celebrities such as Taylor Swift without permission, highlighting the ease with which AI systems can appropriate identity absent clear legal guardrails.
At the federal level, proposed legislation such as the NO FAKES Act seeks to establish a nationwide right protecting an individual’s image, voice and likeness from unauthorized digital replicas. While these proposals acknowledge the inadequacy of existing frameworks for generative AI, legal analysis suggests they remain fundamentally transactional, treating consent as something to be licensed or monetized rather than as a durable expression of personal autonomy, as discussed in Mondaq’s overview of the bill. Parallel efforts such as the TAKE IT DOWN Act, which targets nonconsensual intimate deepfakes, further underscore the piecemeal nature of the U.S. response.
Technology companies have responded unevenly. Some platforms allow public figures or estates to opt out of certain AI uses, while others encourage celebrities to license their likenesses for synthetic content. This opt-in model fits U.S. commercial norms but sits uneasily with the Bardot paradigm, where the most consequential choice is the decision never to participate at all.
That is why Brigitte Bardot continues to matter in AI discourse. She represents a category that AI systems and legal regimes often struggle to accommodate: a globally recognizable figure who has consistently and publicly said no. Her influence on AI is not that of a muse or innovator, but of a constraint. She forces policymakers, lawyers and technologists to confront a question that AI cannot answer on its own—whether technological capability should ever be allowed to override human refusal.
In an industry driven by scale, speed and nostalgia, Bardot’s legacy serves as a reminder that human identity is not raw material. As AI governance evolves, it may ultimately be judged less by what systems can generate than by what they are designed to respect.
You may also enjoy:
- AI and Scarlett Johansson
- AI Fair Use Landmark: Authors v. Anthropic
- Council of Europe Adopts AI Convention
and if you like what you read, please subscribe below or in the right-hand column.