AI-Driven Election Interference: China’s New Weapon Against U.S., South Korea, and India

On the 6th, foreign media outlets, such as The Guardian, reported on a Microsoft warning that China might use artificial intelligence (A.I.) to disrupt elections in South Korea, the U.S., and India.

Microsoft has warned that China, having run a trial using AI content during the Taiwanese presidential opinion polls, intends to disrupt elections in South Korea, the U.S., and India this year.

According to a report released last Friday by Microsoft’s Threat Intelligence Team, the company anticipates that a cyber group supported by the Chinese government will target significant elections in 2024, with potential involvement from North Korea.

As the populations of South Korea, the U.S., and India head to the polls, there’s a high possibility of encountering cyber attackers and influential actors from China, and cyber actors from North Korea are also expected to target these elections.

Speculation has arisen that China will create and distribute AI-generated content that aids its position through social media during crucial elections in South Korea, the U.S., and India. The report warns that while the impact of AI-generated content may be minimal now, the situation could drastically change.

Microsoft’s report also includes that China had already attempted a false information campaign using A.I. during last January’s Taiwanese presidential election. According to Microsoft, this is the first instance of a state-owned company using AI-created content to influence foreign elections.

The Chinese hacker group known as Storm 1376, also known as Spamouflage or Dragonbridge, was highly active during the Taiwanese election.

Their attempts to influence the election included posting a fake audio on YouTube of resigned candidate Terry Gou endorsing another candidate last November. Microsoft identified the clip as AI-generated, and YouTube reportedly removed the content before it reached many users.

The hacker group, known to be supported by China, eventually released a series of AI-generated memes against William Lai, a candidate likely to succeed and a sovereignty supporter opposed by China. The group normalized baseless claims that Lai embezzled national funds as if they were true through these memes.

The use of AI-generated TV news anchors has also increased, a tactic used in Iran. Using the anchor’s voice, the hacker group created controversy by making baseless claims about Lai’s private life, including rumors of being an illegitimate father.

Microsoft revealed that the A.I. news anchor was created using CapCut, a tool developed by ByteDance, the Chinese company that owns TikTok.

According to the report, Chinese hacker groups are continuing influential campaigns in the U.S. It has been pointed out that the hackers supported by China are using social media accounts to raise divisive issues to divide American voters.

It is speculated that hackers are making such attempts to gather information and accuracy about key voting demographics ahead of the U.S. presidential election.

This report was released after an official review committee appointed by the White House criticized Microsoft for repeated errors that allowed Chinese cyber groups to infiltrate the email accounts of U.S. high-ranking officials. Last month, the U.S. and U.K. governments criticized Chinese-supported hackers for running a years-long cyber campaign targeting politicians, journalists, businesses, and even U.K. election monitors.

Microsoft analyzed that “the impact of content created by Chinese A.I. on the audience is still low, but China will continue to conduct more experiments to enhance memes, videos, and audio” and “it could prove effective in the future.”