risk · R27

Unauthorized AI Appropriation of Identity and Persona

Generative AI is enabling the unauthorized replication of individuals' voices, likenesses, and creative styles for commercial or malicious purposes. This form of digital appropriation extends beyond data privacy, constituting a new type of identity and intellectual property theft that affects both the living and the deceased. The ease of access to these powerful tools is rapidly outpacing legal and ethical safeguards.

Updated 2026-09-20

Observatory assessment: 85 / 100. Critical

How assessments are made

Deep dive

The primary driver of this risk is the rapid advancement and democratization of sophisticated generative AI models, including voice synthesis, diffusion models for images, and large language models capable of mimicking writing styles. These systems can be trained on vast amounts of publicly available data—interviews, social media posts, artistic portfolios, and film appearances—to create highly convincing digital replicas or 'deepfakes' of a person's persona. The low cost and accessibility of these tools mean that this capability is no longer confined to specialized effects studios but is available to individual developers, marketers, and malicious actors, creating a fertile ground for exploitation.

The implications are profound and multifaceted. For creators, artists, and public figures, it represents a direct threat to their livelihood and control over their own identity, leading to economic harm through unauthorized endorsements or the creation of competing works. Beyond financial loss, it can cause significant psychological distress and reputational damage, particularly through the creation of non-consensual pornographic material or defamatory content, with recent incidents showing this harm extending from public figures to ordinary citizens, including children. For society, this trend erodes trust in digital media, complicates the enforcement of intellectual property rights, and raises complex ethical questions about posthumous rights and the commercialization of deceased individuals' personas without consent from their estates.

The trajectory of this risk is one of rapid escalation. The technology is continuously improving, making synthetic media increasingly indistinguishable from reality. Current legal frameworks, such as copyright law and the 'right of publicity', were not designed to address the nuances of generative AI and are proving inadequate. While technical solutions like digital watermarking are being developed, they are caught in a perpetual arms race with methods designed to circumvent them. This creates a significant governance gap where the potential for harm grows much faster than the ability of institutions to mitigate it.

Ultimately, the unauthorized appropriation of identity challenges our fundamental concepts of ownership, consent, and authenticity in the digital age. It is not merely about data privacy but about the right to control one's own digital existence and creative essence. Without robust legal, ethical, and technical guardrails, we risk entering an era where personal and artistic identity can be digitally hijacked, commodified, and manipulated on a massive scale, with accountability remaining dangerously diffuse.

Trajectory and projections

Near term

A wave of high-profile lawsuits from artists and actors will force initial legislative and judicial responses to AI-driven voice and likeness cloning. Expect major platforms to introduce stricter policies on synthetic media, though enforcement will be inconsistent.

Mid term

The first comprehensive 'Digital Identity Rights' laws will be enacted in some jurisdictions, establishing clearer consent and compensation models. A legitimate market for licensed AI personas will grow alongside a persistent and sophisticated black market.

Long term

Advanced cryptographic provenance and digital identity management will become standard for public figures, but the underlying technology for appropriation will be so powerful and widespread that it remains a persistent national security and personal safety threat.

Mitigation strategies

• Development of 'Digital Identity Rights' legislation and regulatory frameworks (e.g., EU AI Act) to establish clear consent and usage rules.
• Litigation and legal challenges by affected individuals (artists, public figures) to establish legal precedents for AI-generated likenesses.
• Platform-level policies for labeling synthetic media, content moderation against harmful deepfakes, and user consent/opt-out for AI training data.

Top evidence