Fear, war, and algorithms: the limits of regulating AI-generated campaign speech in Hungary
This article examines the regulatory challenges posed by AI-generated, fear-based political communication through the example of a Hungarian election campaign video depicting a fictional wartime scenario. Widely disseminated on social media, the video employs emotionally intense imagery and narrative framing to present electoral choice as a matter of survival, raising questions about the limits of lawful political speech in digitally mediated environments. The analysis argues that such content reveals a structural gap in contemporary speech regulation. While existing legal frameworks—at the national, European Union, and platform levels—are designed to address clearly defined harms such as incitement, deception, or voter interference, they remain largely unable to respond to emotionally manipulative yet formally lawful political communication. The Hungarian case illustrates this limitation clearly: despite its disturbing nature, the video does not appear to fall within existing sanctionable categories under Hungarian law, appears to fall outside the jurisdiction of media authorities, and would be unlikely to trigger removal obligations under EU law or platform standards. This gap is especially significant in relation to the protection of minors, as content that would ordinarily be subject to safeguards in media regulation may circulate freely during election campaigns. The article situates this problem within broader debates on digital speech regulation and political polarization, arguing that regulatory approaches are better equipped to address falsehood than emotional manipulation. It concludes that future responses should focus on complementary mechanisms—such as platform governance measures such as age-gating, warning screens, and reduced algorithmic amplification—rather than expanding content-based restrictions on political speech.