Taipei: Taiwan is taking significant steps to bolster its defenses against potential artificial intelligence (AI) interference in the upcoming local elections on November 28. This move comes in response to growing concerns that China might seek to influence the electoral process using advanced AI technologies. Analysts have shared with Radio Free Asia that China is likely to employ AI to manipulate voter groups, impersonate Taiwanese social media users, and spread fabricated recordings or scandals just before the election day. According to Radio Free Asia, Taiwan's ruling Democratic Progressive Party (DPP) issued a warning this month about the possible use of AI by Beijing to disrupt the democratic process in Taiwan. The Taiwanese government has firmly rejected China's sovereignty claims, maintaining that the island's future should be determined solely by its citizens. Researchers have pointed out that AI technology can be utilized to rapidly produce content, manage networks of fake accounts, and tailor mess ages to individual voters, posing a significant threat to the integrity of the elections. Experts, including Sze-Fung Lee, an independent researcher specializing in Chinese hybrid warfare, have expressed concerns that AI will play a more prominent role in the November polls. The use of AI has been observed in previous elections, manifesting as deepfakes, coordinated inauthentic behavior, and sophisticated microtargeting. William Yang, a Northeast Asia analyst at the Belgium-based International Crisis Group, highlighted China's long-standing history of interfering in Taiwan's elections, emphasizing that advanced AI technology could intensify these efforts. Beijing has consistently denied any involvement in Taiwan's electoral processes, accusing the DPP of using tensions with China for political gain. However, the DPP's warning comes as the party faces a challenging local election. The opposition Kuomintang (KMT) has historically secured a majority of mayoral and county chief positions, and current projection s suggest they might win more counties in the upcoming elections. Despite this, the DPP remains vigilant, monitoring signs of disinformation that may undermine its electoral prospects and tracing any connections to China. The KMT advocates for increased engagement with Beijing but distances itself from being labeled as pro-China. In contrast, the DPP seeks closer ties with countries like the United States and Japan. While cross-straits relations are a point of contention, Taiwan's local elections are primarily driven by domestic issues. Despite the DPP's emphasis on national security in 2022, it suffered its worst local election result in years. The DPP's recent warning also highlighted AI-generated material linked to Taiwan's 2024 presidential election, including videos promoting controversial content and suspected fake recordings. Research by Taiwan-based civil society group Doublethink Lab revealed that generative AI was used to conduct large-scale spam campaigns and aid inauthentic accounts in varying t heir posts. Attributing such activities to China remains a challenge, as AI-generated material does not directly indicate its creator. Researchers must analyze account networks, posting behavior, and connections with state-backed media to establish a link. AI's ability to mimic local expressions and issues complicates attribution further. In response, Taiwan's Central Election Commission is setting up a specialized task force and an online platform for candidates to report suspected deepfakes. The election preparations align with broader concerns about Chinese cyber activity, with Taiwan's National Security Bureau reporting an increase in cyberattacks on critical infrastructure. While the cyberattack figures do not directly demonstrate AI-enabled election interference, they support the government's assessment of sustained Chinese pressure. Sze-Fung Lee noted that the data underscores the frequency of cyberattacks and cognitive warfare targeting Taiwan, reinforcing the government's cautious approach to elec tion security.