Taiwan's National Security Bureau told the Legislative Yuan on April 7, 2026, that China's government service network had been attacked more than 173.28 million times in the first three months of the year alone. In the same briefing, the bureau said it had identified roughly 13,000 suspicious internet accounts and 860,000 disputed messages tied to Beijing's cognitive-warfare campaign, alongside 420 incursions by Chinese military aircraft into Taiwan's airspace in the same quarter, coordinated with ten separate naval "joint combat readiness patrols." The target of all this, according to the bureau, is not a presidential race. It is the nine-in-one local elections scheduled for November 28, 2026 — a vote for mayors, county magistrates and more than 11,000 village and township seats that Taiwan's political system has historically treated as insulated from national security debates.
That insulation is exactly what analysts say is breaking down. According to reporting by the Liberty Times, later relayed by the Global Taiwan Institute, Chinese officials designated interference in the November elections a top priority at the Communist Party's annual Taiwan Work Conference in February 2026, discussing a dedicated task force for what internal Chinese messaging calls "united front work in cyberspace" against Taiwan independence forces — the party's standing shorthand for the governing Democratic Progressive Party. Because the conference is closed-door, the claim traces to a single Taiwanese media report, however widely it has since been repeated; it should be read as credible but not independently confirmed by a second, differently sourced outlet.
What Changed Since 2024
The tactics themselves aren't new. During Taiwan's 2024 presidential race, PRC-linked accounts amplified public anger over an egg shortage and circulated forged documents alleging secret cash payments from Taipei to Paraguay, aimed at damaging then-candidate Lai Ching-te. In the 2022 local elections, Taiwan's Doublethink Lab tracked a shift toward more decentralized, harder-to-trace campaigns, including a false rumor that the ruling party planned to sell chipmaker TSMC to the United States. What's different heading into 2026 is documented in the bureau's own January 2026 report: a 60 percent increase in inauthentic social media accounts between 2024 and 2025, and more than 2 million individual instances of disinformation last year, a 74 percent rise since 2023.
A leaked database compiled by a Chinese influence-for-hire firm sorted at least a thousand named Taiwanese political figures into categories — "Hardliners," "Moderates," "Swing Voters" — built for micro-targeting, not blanket propaganda.
The clearest evidence of a qualitative shift came out of a 2025 leak. Vanderbilt University's Institute of National Security and Doublethink Lab jointly analyzed internal documents from GoLaxy, a Chinese influence-operations firm, and found it had compiled a database of 50,000 Taiwan-related news items, with political figures — reportedly including Lai Ching-te, Ko Wen-je and Su Tseng-chang — sorted into targeting categories for tailored messaging. It's a level of individualized profiling that older, blunter content-farm operations never approached. Taiwan's prosecutors have since named AI-generated disinformation as one of four priority areas for election-related investigations in 2026, and researchers at Taiwan.md have identified more than 120 spoofed news websites built ahead of the vote.
The most striking new precedent isn't cross-strait at all — it's intra-party. In October 2025, Kuomintang commentator Jaw Shaw-kong alleged that AI-generated content and newly created accounts with foreign IP addresses had targeted moderate KMT leadership candidate Hau Lung-bin while boosting Cheng Li-wun, a rival seen as more conciliatory toward Beijing, who went on to win the party's chairmanship race. The National Security Bureau later said it had identified more than 1,200 TikTok and YouTube videos about the race originating from overseas accounts. If confirmed as sustained policy, it would mean Chinese influence operations are no longer simply trying to elect the more Beijing-friendly of Taiwan's two major parties — they are trying to shape who leads the less Beijing-friendly one, too.
Public opinion complicates any simple story about the effect of all this. Chicago Council on Global Affairs polling found 60 percent of Taiwanese favor maintaining the political status quo and 76 percent support a US commitment to defend the island — far above the 38 percent of Americans who say the same in matching surveys. But support for higher defense spending splits sharply by party identity: over 80 percent among supporters of the pro-Taiwan-identity "green" camp, under 20 percent among "blue" opposition supporters, and under 25 percent among the smaller Taiwan People's Party base, according to polling presented at the East Asia Democracy Forum. That is not a electorate uniformly hardening against Beijing — it is one where the interference and the domestic argument about how to respond to it now run along the same partisan line, which is precisely the kind of division a foreign influence campaign can widen rather than overcome.