Exploitation of Public Figures' Faces by AI Platforms: Unjust Enrichment and Layered Accountability under Indonesian Law

The study argues that a public figure's face has shifted within generative AI ecosystems from a marker of identity into an intangible economic asset sustaining platform valuation. Where prior discussion tends to place the face within personality-rights or privacy frameworks, this work repositions it as an asset with distributable economic value, moving the question from personality protection toward benefit allocation. A conceptual reconstruction within normative legal research, developed through systematic and teleological interpretation of the Copyright Law, the Electronic Information and Transactions Law, the Personal Data Protection Law, and Article 1365 of the Civil Code; it contains no empirical data or sample.

The study finds that existing legal regimes are fragmented rather than complementary on the use of faces: copyright refuses to treat a face as a protectable work, and data protection law leans toward privacy rather than economic interest. The judgment identifies the misalignment between the two regimes as the very condition enabling unjust enrichment, rather than as a mere legislative gap. Based on systematic and teleological interpretation of the four legal instruments, argued at the level of statutory text.

The study holds that the four elements of unjust enrichment—enrichment, loss, causation, and absence of legal basis—can all be satisfied in this setting, and that the conduct may be qualified as unlawful under Article 1365 of the Civil Code. It uses unjust enrichment as an economic-analytical framework for generative AI's exploitation of identity and connects it to a pathway for finding unlawful conduct. A subsumption-style normative argument whose persuasiveness depends on the interpretation of the four elements rather than on empirical testing.

The study proposes a layered accountability model consisting of profit restitution, compensation, dataset transparency, and explicit consent. It expands the remedial design from a single form of liability into a multi-layered arrangement and brings dataset transparency and explicit consent into the accountability structure. A normative institutional proposal; the text offers no empirical support on implementation effects or comparative-law experience.

The conclusions apply to normative analysis within the Indonesian legal system and are addressed to legislators, judges, and researchers concerned with generative AI, personality rights, and data governance. The layered accountability model offers a starting point for later refinement of profit restitution, compensation, dataset transparency, and explicit consent in specific statutory provisions, and can serve as a reference framework for comparative legal research.

As normative legal research, the study provides no empirical data or case testing, so how the four unjust-enrichment elements would be subsumed in concrete disputes remains to be tested in judicial practice. How dataset transparency and explicit consent in the layered accountability model would be implemented, who would supervise them, and how they would connect with existing data protection mechanisms are not developed in the text. In addition, the study is situated in Indonesian law, and its applicability in other jurisdictions would require separate argument.