Microsoft says a China-backed group is using AI misinformation to sway foreign elections

Microsoft is sounding the alarm about China’s use of AI to create misinformation campaigns and sway foreign elections.

Microsoft Threat Intelligence released a report Friday saying that China “will, at a minimum, create and amplify AI-generated content that benefits their positions” ahead of high-profile elections in the U.S., South Korea, and India. The company’s team of cyberthreat experts and researchers found that an online operation backed by the Chinese Communist Party, known as “Spamouflage” or “Dragonbridge,” has begun using AI in attempt to influence public opinion in Taiwan and Canada.

Ahead of the Taiwan presidential and legislative elections on Jan. 13, Spamouflage used AI to make fake audio clips of a former candidate, who had dropped out of the race months earlier, endorsing someone else.

“This was the first time that Microsoft Threat Intelligence has witnessed a nation state actor using AI content in attempts to influence a foreign election,” the researchers wrote.

The China-backed group — somewhat confusingly given another name by Microsoft, “Storm-1376” — also used tools from TikTok parent company ByteDance to make videos with an AI-generated news anchor saying the U.S. and India were responsible for unrest in Myanmar. Microsoft said Spamouflage created AI-enhanced videos to target Canadian members of parliament and posted AI-generated photos claiming the U.S. government caused the wildfires in Maui while testing a “weather weapon.”

“While the impact of such content in swaying audiences remains low, China’s increasing experimentation in augmenting memes, videos, and audio will continue—and may prove effective down the line.” — Microsoft Threat Intelligence Apr. 5 report

A U.S.-backed independent review board blamed Microsoft in a report released earlier this week for security failures that allowed a Chinese hacking operation — this one called “Storm-0558” — to access emails of top U.S. officials.