Taiwan’s National Security Bureau has gone public with a stark warning about the digital battlefield unfolding around the island. In a report submitted to lawmakers this week, the agency said that Beijing has flooded Taiwan with more than 2.5 million pieces of controversial or false information in the first nine months of 2026. That marks a jump of roughly 48 percent from the same period a year earlier, when the figure stood at more than 1.69 million. Behind these numbers lies a sobering reality: Taiwan’s information space is being treated as a target by a sophisticated and increasingly aggressive operation aimed at shaping public opinion and influencing the outcome of elections. With Taiwan’s local elections scheduled for November, the timing of the surge is particularly alarming. The report does not treat these messages as random internet noise. Instead, it describes a coordinated push to exploit social media, state-controlled news outlets, and even artificial intelligence to create confusion, deepen social divisions, and nudge voters in directions that benefit Beijing. The National Security Bureau’s findings were prepared for elected officials, but their meaning reaches far beyond government corridors. For ordinary Taiwanese, the report is a reminder that what they see on their screens may not be what it seems, and that the fight for Taiwan’s political future is increasingly taking place in the feeds, forums, and comment sections they scroll through every day.
The content of these operations is as varied as the platforms they use. According to the intelligence report, China-backed messaging has focused on energy policy, food safety, defense issues, and even the images and records of specific candidates. The goal is to influence Taiwan’s November elections, in which voters will choose city mayors and county magistrates across the island. These are not national-level races, but they carry real political weight. Local governments control resources, shape public discourse, and often serve as stepping stones for larger political ambitions. The National Security Bureau said the messages have spread mainly through Facebook, Threads, YouTube, and online forums, the spaces where Taiwanese citizens regularly gather to share news, debate politics, and entertain themselves. The aim, the report suggests, is not necessarily to convince everyone to adopt a pro-Beijing view. Rather, it is to fan the flames of discontent, amplify anti-government sentiment, and make it harder for society to reach consensus. Beijing has repeatedly denied that it interferes in Taiwan’s elections or spreads disinformation. Its Taiwan Affairs Office did not immediately respond to a request for comment on the new report. China continues to insist that Taiwan is part of its territory, a position Taipei firmly rejects, and it has long accused Taiwan’s government of using such accusations to drum up anti-China sentiment.
Taiwanese officials have for years described China’s tactics as “grey zone” operations, actions that fall somewhere between ordinary diplomacy and open conflict. These include daily military drills, cyberattacks, and now information warfare. The latest report fits neatly into that framework. It says Beijing often begins by creating narratives through official media outlets, then amplifies those narratives using what the agency calls “troll armies” and public relations agencies. The objective is to sow discord among the Taiwanese public and erode trust in democratic institutions. The methods are familiar but increasingly sophisticated: content farms producing large volumes of posts, AI-generated videos designed to look authentic, fake accounts that pretend to be real citizens, and coordinated campaigns that flood comment sections with identical or similar messages. In practice, this can look like a sudden barrage of angry responses to a political post, a viral video that turns out to be manufactured, or a steady stream of doom-scrolling headlines that leaves people anxious and exhausted. The report paints a picture of an information environment where authenticity is hard to verify and where even the most organic-looking social media trend may have been tested, timed, and launched with strategic intent. That is a difficult challenge for any democracy, but especially for Taiwan, which faces constant pressure from a much larger neighbor and holds elections that Beijing watches closely.
Artificial intelligence has become a key weapon in this campaign, according to the National Security Bureau. The report states plainly that China has expanded its use of AI to manipulate information, and that this technology is being used to gather intelligence about Taiwanese current events, candidate policy positions, and polling data. That information is then fed into influence operations designed to shape voter preferences and election outcomes. It is not hard to imagine how this works in practice. AI can help generate enormous amounts of tailored content in a short time, mimic local dialects and cultural references, and even produce realistic videos of politicians saying things they never said. It can also analyze public reactions in real time, allowing operators to adjust their messages quickly. The report warns that this kind of AI-driven misinformation is not just about spreading lies. It is about creating a constant hum of doubt, making it harder for people to know whom to trust, and pushing voters toward choices that may not reflect their true interests. The intelligence agency’s language is careful, but the message is clear: the next few months will be a test of whether Taiwan’s citizens can recognize and resist these digital manipulations. For those working in media, civil society, and political campaigns, the report is a call to be more transparent, more fact-checked, and more aware that foreign influence can hide behind seemingly ordinary online interactions.
Alongside the flood of disinformation, the report also documents a dramatic increase in cyberattacks against Taiwan’s government. The National Security Bureau said Chinese cyberattacks against Taiwanese government agencies averaged 2.58 million per day over the first nine months of 2026. The number of detected threats rose sharply beginning in June, reaching 185.6 million in July and 106.4 million in August. Those are staggering figures, and the agency suggests they may be more than random hacking attempts. The report says it cannot be ruled out that these attacks were part of a preliminary operation to gather information and prepare disruptive actions aimed at interfering with Taiwan’s year-end local elections, known locally as the “nine-in-one” elections. Such activities could be used to steal sensitive data, undermine public confidence in the electoral system, or simply overload government systems at critical moments. Cybersecurity experts often compare this kind of pressure to a probing process, where attackers test defenses, map vulnerabilities, and wait for the right moment to strike. The report’s tone suggests Taiwan’s intelligence community is taking that possibility very seriously. Combined with the increase in disinformation, the cyberattacks point to a two-front war, one aimed at the minds of voters and the other at the machinery of government itself.
For Taiwanese citizens, the report is more than an intelligence briefing, it is a reminder that democracy in the digital age requires vigilance. The idea that foreign powers might try to influence elections is not new, but the scale, sophistication, and openness of China’s operations have changed the game. The National Security Bureau’s report also echoes findings from a Reuters investigation conducted with the Taipei-based research group IORG, which found that Chinese state media have increasingly amplified critical voices in Taiwan, including influencers and politicians linked to the opposition Kuomintang. That kind of influence can be harder to detect than overt fake news because it often uses real people, real complaints, and real political grievances. It blurs the line between legitimate criticism and foreign manipulation. In this environment, ordinary Taiwanese are left to navigate a landscape where every post, video, and comment could be part of a larger strategy. The human cost is measured not only in hacked accounts or viral misinformation, but in trust: trust in institutions, trust in the media, and trust in fellow citizens. The report does not offer easy solutions, but it does make one thing clear. The information war is already here, and the defense against it starts with awareness, skepticism, and a shared commitment to protecting the democratic process from forces that would prefer to see it fail.