How Generative AI Is Scaling Transnational Repression

Generative AI is being used to scale digital transnational repression by enabling state-aligned operators and proxies to identify, monitor, and target diaspora activists and exiled dissidents across borders. A detailed Threat Intelligence report documents cases in which an AI assistant was used to translate messages, craft recruitment scripts in local dialects, geolocate individuals, profile vulnerabilities (such as financial hardship or family ties in the home country), and turn fragmented social-media and public-data traces into actionable intelligence. Operators relied on models to substitute for language skills and analyst expertise, develop persona systems and campaign plans, automate daily surveillance briefings, and map social networks - amplifying reach while lowering operational costs. In at least one instance, a commercial intelligence platform analyzing diaspora activity was disrupted before deployment.

These AI-enabled workflows produced reconnaissance useful for offline coercion: informing informant recruitment, physical surveillance, and disruption of civic activity, with examples including compiled surveillance files, venue floor plans, and protest-route reconnaissance. While the report represents one company’s limited visibility and some attributions are uncertain, its operational detail shows a meaningful expansion in pace and scale of cross-border repression. It recommends that frontier AI firms treat transnational repression as a distinct threat category, publish incident reporting, adopt detection protocols, notify affected communities, and preserve investigatory evidence.