On the eve of an election, the social media platform X quietly suspended about two dozen accounts linked to a Russian propaganda syndicate known as Storm-1516. On the surface, that sounds like good news—a last-minute crackdown, a win for the defenders of democracy. But the closer you look, the more complicated it becomes. The suspensions did not happen in a vacuum. They came as a wave of fabricated stories was washing through the feeds of voters in three states, targeting Democratic candidates in competitive Senate races. James Talarico in Texas and Sherrod Brown in Ohio were both hit with false sexual assault allegations. Troy Jackson in Maine was accused of workplace violence. None of these claims survived even basic fact-checking; they collapsed almost immediately. Yet the lies had already been shared, screenshotted, turned into memes, and boosted by accounts with enormous followings. Disinformation researchers who tracked the campaign say it was orchestrated by Storm-1516, a Russian operation that specializes in using artificial intelligence, bot networks, and paid social media personalities to manufacture outrage. The fact that X removed two dozen of those accounts is real progress, but it is not the whole story. Many more accounts tied to the same syndicate remain active, and the false narratives they helped create are still circulating. As researchers put it, once disinformation takes hold, it’s like a stain that won’t come out. You can scrub one corner, but the rest spreads. There is a reason disinformation researchers describe their work as whack-a-mole. Every time a network is exposed, it fractures and reappears under new names. Sometimes the same people come back with new accounts. Sometimes the algorithms promote the content before the platform ever gets around to reviewing it. And sometimes the platform doesn’t want to act because the accounts are popular, profitable, or likely to cause a feud over censorship. The result is an uneven, inconsistent response that leaves voters exposed.
French disinformation researcher David Colon, a professor at Sciences Po in Paris, has been documenting the group’s activity for months. On Friday, he identified 20 accounts associated with Storm-1516 that had been suspended from X. That might seem like a decisive action, but Colon’s findings also include dozens of other accounts linked to the same network that are still live. Together, those surviving accounts have more than 2.4 million followers. That is not a small operation. It’s a substantial audience, larger than many mainstream news outlets. And here’s the part that makes this so difficult to police: most of those accounts belong to real people, not bots. They have real faces on their profiles, real life histories, real followers, and real opinions. Some of them are being paid to push talking points. Some are true believers who have convinced themselves that their lies are necessary. Some are just chaos agents who enjoy watching the world burn. Colon said the accounts that got suspended were either fake accounts or accounts amplified by coordinated inauthentic activity. But the ones that remain are harder to classify. “Those that have been suspended are either fake accounts or accounts amplified by coordinated inauthentic behaviors” in violation of X’s rules, he said. “It’s complicated when it’s not a bot account because you can get a backlash from the people whose accounts are suspended.” The same network isn’t just trying to influence Americans. It’s also active in France, where hundreds of thousands of people have taken to the streets in youth-led protests over education costs and the cost of living. In other words, this isn’t a one-off election trick. It’s a global disinformation machinery, tailored to whatever tensions already exist. The coordination across borders is one of the strongest signs that this is not just scattered online craziness. The same phrases, the same imagery, and the same AI-generated faces appear in different countries at the same time. It suggests a playbook that can be adapted quickly, with local influencers inserted into the machinery.
X’s head of safety, Michael O’Herlihy, didn’t respond to requests for comment, and the silence is telling. For weeks, X’s official channels have been talking about their efforts to protect the platform. In mid-October, the company’s global government affairs team announced that it had suspended more than 13,000 accounts involved in suspected foreign influence operations, including networks linked to Russia, China and Iran. X framed this as an essential defense of democracy. “Free speech is the bedrock of a functioning democracy,” the platform said in a post, while also promising to make sure conversations on the site are “authentic, open and safe.” The banned accounts, according to X, were “inauthentic accounts” used to “post and amplify content about US politics, targeting candidates and public figures across the political spectrum.” That language is carefully neutral—it doesn’t single out any country or party—and that’s part of the problem. Enforcing rules evenly is difficult, especially when the rules are designed to be broad. Many of the accounts that researchers have linked to Storm-1516 and that remain active are designed to look like the kind of conservative Americans you might meet at a diner or a football game. They post about the Constitution, about faith, about family. One account with more than 144,000 followers says it exists to defend U.S. “constitutional principles.” Another, with more than 113,000 followers, describes itself as a “single, independent believer in truth.” If you didn’t know better, you’d think they were just passionate citizens. That’s exactly the point. Colon says the hardest part is determining authenticity. “You can’t say for sure that it’s true or fake,” he said. “How can you be sure, you know?” It’s a question that hangs over everything: how do you stop a lie from spreading without punishing the people who are simply repeating it?
At its core, this is the era of weaponized ambiguity. The goal of a group like Storm-1516 is not necessarily to convince you that a particular candidate is a monster. It’s to make you unsure whether anything you read is real. It’s to sow so much distrust that you stop believing in elections altogether. The old model of Russian disinformation relied on troll farms posting from gray offices in St. Petersburg, clumsily pretending to be Americans. That approach still exists, but it’s been refined. The new model is more distributed, more human, and harder to detect. Storm-1516 uses a network of real collaborators across international borders. According to the Gnida Project, a research initiative that monitors Russian disinformation, the group appears to have an “international network of real collaborators to publish and promote their narratives.” That’s a fancy way of saying that people are being paid to pretend they stumbled across a story and decided to share it. The Gnida Project says the solution is obvious: X could “ban anyone posting their content for money.” But that would require the platform to trace payments, identify patterns, and make judgment calls about intent. So far, X hasn’t done that. Colon says the financial incentives are pretty straightforward. Storm-1516 recruits influencers by offering “a few hundred bucks for each publication” of disinformation on social media, though the rate “depends on the number of followers you have.” A small account might get a few hundred dollars. A big account could command thousands. And there’s another layer: X itself pays creators for viral posts. That means an influencer can make money from the propaganda group and from the platform at the same time, effectively turning disinformation into a side hustle. The more outrageous the post, the more engagement it gets. The more engagement, the more money. The more money, the more outrageous the post. It’s a perfect feedback loop, and the only thing being burned is the truth.
One of the most visible faces of this system is an account called “Johnny Midnight,” which has more than 760,000 followers. On its profile, it describes itself as a MAGA parody account and openly advertises “paid promotions.” That should be a red flag, but in the current online ecosystem, it’s practically a business model. Researchers say Johnny Midnight has a history of spreading pro-Kremlin stories and was among the accounts that pushed false allegations against both Sherrod Brown and Troy Jackson. In the final days before the election, the account was busy, not just amplifying the fake scandals but also posting about voter fraud, election integrity, and the need for real Americans to stay vigilant. Another figure identified by Colon is Joey Mannarino, an internet personality with more than 779,000 followers. Mannarino’s bio says he runs The Overton Institute, which sounds like a think tank but appears to be a made-up organization designed to give him an air of credibility. He is also a host for NewsForce, an outlet that claims to be an independent news network but that Colon says is a front for distributing propaganda. A review of NewsForce’s website using the detection tool GPTZero found that many of its articles appear to be entirely AI-generated—no reporters, no sources, just algorithmically produced text designed to look like news. Colon says Mannarino is “probably a paid-for-hire influencer and also somebody ideologically involved in extreme-right propaganda.” In the days before the election, Mannarino was outspoken in his support for Republicans and his contempt for Democrats and libertarians. He wrote that “we have a country to save” and begged followers to vote Republican. He also made baseless claims about voter fraud. “Tomorrow will have more voter fraud attempts than any day in the history of America,” he tweeted on Monday morning. “Be vigilant for these instances and report them IMMEDIATELY!” There is no evidence this was true. But the tweet was widely shared, because fear travels faster than fact. Mannarino didn’t respond to messages seeking comment. Instead, he announced that he and the NewsForce crew would be “going LIVE” for an election-night special called “THE BATTLE FOR AMERICA,” sponsored by Kalshi. It was all part of the performance.
Storm-1516 is not the only operation trying to influence the outcome. Graphika, a U.S.-based company that tracks online narratives, says another influence network linked to the notorious group Spamouflage has been active in recent days. Spamouflage has long been associated with Chinese influence operations, and its latest posts closely resemble the propaganda it pushed ahead of the 2024 presidential election. According to Graphika, the network includes accounts that “post content questioning election integrity and the legitimacy of U.S. democracy.” If there’s a common thread between Russian and Chinese operations, it’s not ideology. It’s cynicism. The goal is not to make you vote for a particular candidate; it’s to make you believe the whole system is rigged. It’s to make you so exhausted, so angry, and so confused that you either stay home or accept whatever outcome happens because you never trusted it anyway. For ordinary voters, this is terrifying. It can feel like being trapped in a room where every mirror is distorted and every stranger might be an actor reading a script. The suspension of 24 accounts might seem like a small victory, but it’s also a warning: the people behind these campaigns are already building the next wave. They know how to hide. They know how to recruit. They know that the platforms will respond slowly and inconsistently. They know that a lie can travel around the world in seconds while the truth is still tying its shoes. Researchers like Colon and the teams at Graphika and the Gnida Project are doing what they can, but they are outnumbered and underfunded compared to the forces they’re fighting. That means the rest of us have a role to play. Not by becoming paranoid, but by becoming more deliberate. Look at the source before you share. Ask whether the account has a history of being wrong. Be suspicious of posts that make you feel rage or disgust, because those are the posts designed to be shared. The truth is often less exciting than the lie, but it is also more durable. In a world full of manufactured chaos, the most radical thing you can do is demand evidence. The propaganda machine depends on your willingness to click first and think later. Refuse to be that easy. Because the real battle for America—and for every other democracy—isn’t just about elections. It’s about whether we can still tell the difference between what is real and what is paid for.