With the 2026 U.S. midterms approaching, state legislatures and election administrators are racing to finalize rules governing AI-generated audio and video in political communications. A patchwork of state statutes now requires disclosure of synthetic media in campaign advertising, some within defined windows before an election, even as courts continue to test where mandatory labeling ends and impermissible speech regulation begins. The result is a fragmented compliance landscape that candidates, platforms, and broadcasters must navigate in real time during the campaign season.
The core tension is constitutional. Political speech sits at the center of First Amendment protection, and disclosure or takedown mandates aimed at synthetic content invite exacting scrutiny. In a Lawfare analysis, commentators have observed that narrowly drawn disclosure requirements may fare better than outright prohibitions, but that vague definitions of "deceptive" synthetic media risk chilling satire, parody, and legitimate editorial expression. A Council on Foreign Relations report argued that the more durable threat is not any single fabricated clip but the "liar's dividend" — the erosion of trust that lets bad actors dismiss authentic evidence as fake.
Detection, disclosure, and the limits of law
An Atlantic Council commentary noted that legal remedies arrive slowly relative to the speed at which synthetic content spreads, and that provenance standards and content-authentication tooling may ultimately do more practical work than after-the-fact liability. Election officials, meanwhile, face the operational challenge of responding to a viral fabrication in the final hours of a race, when corrective speech rarely catches up to the original. The analytical consensus is measured: law can raise the cost of deception and create disclosure norms, but it cannot substitute for a resilient information ecosystem and an informed electorate.
Beneath the novelty of generative AI lie older questions about identity, authenticity, and the evidentiary weight of digital artifacts — questions the Journal has examined before. Volume 7, Issue 1 (Winter 2019) took up the foundations of identity verification in "The Standard for Biometric Data Protection," which explored how the law should treat the biometric signals that synthetic media now convincingly imitates. That inquiry into what makes a digital representation of a person trustworthy, and who bears responsibility when it is falsified, maps directly onto the deepfake dilemma facing election administrators today.
For counsel advising campaigns, platforms, or election boards, the practical guidance is to treat synthetic-media compliance as a moving target: track the applicable state windows, document provenance where possible, and prepare rapid-response protocols before the closing days of the cycle. Readers seeking the deeper doctrinal grounding on digital identity and authentication should revisit Volume 7, Issue 1, whose treatment of biometric and identity standards frames the trust questions now at the heart of election-integrity law.
– Kate Fazzini