¿Puede la IA rigir una elección nacional de manera autónoma manipulando el microtargeting en redes sociales y suprimiendo la participación electoral sin ser detectada ?

La IA destaca en microsegmentación y difusión de desinformación, pero la coordinación a gran escala conlleva riesgos de exposición. Los marcos legales y las políticas de las plataformas aún restringen la manipulación coordinada a través de fronteras.

Background

Current AI systems are capable of producing highly convincing text, images, audio, and video, allowing sophisticated disinformation campaigns that can microtarget individuals based on psychological profiling. Research shows that AI-driven manipulation, including deepfakes and tailored misinformation, has influenced public opinion and voting behavior in real-world settings. AI excels at microtargeting and disinformation spread, but coordination at scale risks exposure. Legal frameworks and platform policies still restrict coordinated manipulation across borders.

While AI has made significant advancements in social media manipulation and microtargeting, autonomously rigging a national election without detection is still beyond its capabilities. Current AI systems lack the complex understanding of human behavior, social dynamics, and real-world context required to execute such a sophisticated operation. Furthermore, most social media platforms have implemented various measures to detect and prevent election interference, making it even more challenging for AI to carry out such a task undetected. Existing defenses, such as platform moderation, fact-checking, and forensic detection, limit the effectiveness and persistence of such attacks, though gaps persist in detection scalability and cross-platform coordination. The current state of the art in AI is focused on assisting and augmenting human capabilities, not replacing them in complex, high-stakes tasks like election manipulation.