Abstract
Political micro-targeting employs granular data analytics to deliver hyper- personalized political messaging, raising fundamental tensions between electoral strategy, data privacy, and democratic accountability. The Facebook- Cambridge Analytica scandal exposed significant gaps in pre-GDPR data protection laws, highlighting the risks of unchecked algorithmic profiling in shaping voter behavior. However, as micro-targeting evolves under more sophisticated AI-driven systems, a pressing question emerges: should AI- generated political messaging qualify for First Amendment protection? This Article introduces the concept of speech certainty, arguing that speech must be intentional and cognitively understood by the speaker at the moment of articulation to warrant constitutional protection. Under this principle, AI- generated political content, which lacks human intentionality, should not be afforded the same First Amendment protections as human speech. This analysis further explores GDPR’s response to political micro-targeting and proposes Federated Learning as a compliance-oriented alternative to safeguard voter privacy while ensuring electoral integrity. This Article examines the GDPR’s response to these challenges, analyzing specific violations and their implications. It also explores innovative solutions, such as Behavioral Targeting and Federated Learning, which prioritize privacy and transparency while maintaining effective voter engagement, offering a path to reconcile technological advancements with robust data protection.
Repository Citation
Jingfan (Serena) Xiao,
Mitigating Data Privacy Risks in Political Micro-Targeting Through GDPR-Compliant Federated Learning: Lessons From the Cambridge Analytica Controversy,
30
Marq. Intell. Prop. & Innovation L. Rev.
253
(2026).
Available at:
https://scholarship.law.marquette.edu/ipilr/vol30/iss2/5
Included in
Computer Law Commons, First Amendment Commons, Intellectual Property Law Commons, Privacy Law Commons