A Chinese influence campaign that has tried for years to promote Beijing’s interests is now using artificial intelligence and a network of social media accounts to amplify American discontent and division in the run-up to the US presidential election, according to a new report.
The campaign, known as Spamouflage, hopes to spark disenchantment among voters by denigrating the United States as being plagued by urban decay, homelessness, fentanyl abuse, gun violence and crumbling infrastructure, according to the report released Thursday by the Institute for Strategic. Dialogue, a non-profit research organization based in London.
Another goal, according to the report, is to convince the international public that the United States is in a state of chaos.
Artificially generated images, some of them also photoshopped with tools like Photoshop, pushed the idea that the November vote would damage and potentially destroy the country.
An article on Other images showed the two men facing each other, cracks in the White House or the Statue of Liberty, and terminologies such as “CIVIL WAR,” “INTERNAL CONFLICT” and “COLLAPSE OF AMERICAN DEMOCRACY.”
The stories do not appear to have overtly partisan leanings, although Mr. Biden has been the target of multiple negative portrayals, including references to his son. Hunter Biden’s legal troubles and claims the president is a drug user. Spamouflage’s attitude toward Mr. Trump was more ambiguous; messages claiming that his “antihero status makes him unstoppable” could be interpreted as flattering. Both men were described as being too old to govern.
In America’s “hyper-polarized divide,” China sensed an opportunity, said Elise Thomas, a senior analyst at the institute who wrote the report. Spamouflage’s focus on social conflict and antagonism in the U.S. presidential race could also indicate how Beijing hopes to shape the future. many other important elections which will take place around the world this year.
“In this narrative universe, American democracy is presented as a source of discord and weakness,” Ms. Thomas said in a statement. “They seek to create the impression of a sclerotic superpower in disarray, incapable of resolving its internal problems and unfit to act as a leader on the international stage. »
Spamouflage has been active since at least 2017, Ms. Thomas wrote in the report, adding that the campaign is “infamous among researchers both for its sprawling size and for its inability to generate notable engagement from real users of the spam. social media “. Meta, which owns Facebook and Instagram, said last summer that it had deleted thousands of social media accounts and hundreds of campaign-related pages. Meta researchers linked the campaign to Chinese law enforcement.
Thursday’s report focused on Spamouflage posts on social media”, it spread much more easily on X.
In their report on digital risks for this year’s elections, researchers at the NYU Stern Center for Business and Human Rights warned that the main threat in the 2024 elections comes less from AI-generated content than from the spread of false, hateful and malicious content. violent material. THE reportreleased Wednesday, said this type of content was more prevalent because many social media platforms, including Facebook and YouTube, had backed away from some of their past commitments related to election integrity.
The researchers pointed the finger at X, saying that since Elon Musk took power in late 2022, much of the platform’s election team has been eliminated, toxic content rose and other social media companies used the volatility as an excuse to let their guard down.
The researchers also noted that political polarization in the United States was likely to prompt China and other countries to try to sow confusion among voters.