Recent reports are revealing China's strategic use of psychological warfare tactics aimed at shaping public perceptions and beliefs without engaging in direct military conflicts. This emergent concept underscores the evolving landscape of modern warfare, emphasizing the manipulation of cognitive processes as a central tenet of this approach. Notably, China is leveraging artificial intelligence-produced disinformation, including realistic yet fabricated audio recordings and videos, as a core component of its strategy. A significant development is the unprecedented involvement of state-backed entities in influencing foreign elections, enabled by AI technologies that offer newfound opportunities not available just a year ago, thus presenting a convenient means for such operations to be executed.

The intensifying China-Taiwan rivalry has brought the notion of cognitive warfare to the forefront, showcasing China's recent shift towards influence-centric military operations over the past two decades. Faced with intricate geopolitical dynamics and practical constraints, China seems to have calculated that direct military intervention in Taiwan could incur substantial costs. Consequently, reports indicate that China's experimental use of AI-generated content during Taiwan's general election was a trial run, with future plans to extend these tactics to influence electoral processes in countries like the United States, South Korea, and India, as outlined in a recent Microsoft report. Furthermore, there are indications that North Korea might collaborate with China to target upcoming U.S. elections through their state-supported cyber entities, reflecting a growing nexus of cyber and influence operations.

Projections suggest that China would disseminate AI-generated content via social media platforms to manipulate electoral outcomes in their favor. Although the current impact of fabricated content on swaying public opinions remains moderate, the evolving technological landscape and China's continued experimentation could amplify their influence significantly in the future. State-sponsored cyber groups, such as the Chinese-affiliated Storm 1376, are at the forefront of these campaigns, with their activities prominently noticed during the Taiwan presidential election. This group, also known by aliases like Dragonbridge or Spamouflage, was responsible for circulating a fake audio clip of candidate Terry Gou, who ultimately withdrew from the election race. Despite YouTube's subsequent removal of the clip following reports, it is presumed that the disinformation reached a substantial audience before being taken down.

Further instances include the targeted dissemination of AI-generated memes portraying opposition candidate William Lai as a corrupt figure misappropriating state funds due to his perceived anti-China stance. Concurrently, AI-created TV anchor figures emerged, spreading false allegations such as Lai fathering illegitimate children. Notably, the generation of these AI anchors was facilitated by a tool called Capcut, developed by Chinese tech conglomerate ByteDance, owner of popular platforms like TikTok. Moreover, an investigation by the Institute for Strategic Dialogue revealed the operation of deceptive accounts on the X platform attributing false claims to US figures like President Biden and the CIA, with an intent to influence public perception. These deceptive accounts, feigning authenticity by adopting Western or diverse profiles, have been flagged as part of a broader disinformation campaign orchestrated by Chinese actors.

In response to these activities, major social media platforms like Meta, encompassing Facebook, Threads, and Instagram, have taken steps to remove numerous suspicious accounts allegedly linked to groups like Storm 1376. However, newer accounts are adopting more sophisticated strategies, blending with authentic content to garner followers and projecting a human-operated facade to evade detection. Amidst mounting evidence of disinformation campaigns originating from Chinese entities, China's official stance remains one of denial, disavowing any involvement in activities aimed at influencing elections in foreign territories. Despite these assertions, the covert campaigns orchestrated by the aforementioned groups persist, raising concerns about the pervasive impact of AI-powered strategies on shaping global public opinions and electoral outcomes.