U.S. Think Tank Contradicts Itself: Iranian Online Trolls Run Rampant, Yet We’re Powerless to Stop Them?

Everyone, I recently heard that the U.S. itself has admitted it was caught off guard by Iran in the information war. What’s going on? It turns out that Iran’s online trolls are much more skilled than we imagined. The Center for Strategic and International Studies (CSIS) just released a report detailing how Iran launched a multi-platform disinformation campaign, causing the U.S. to suffer a series of setbacks in its efforts to win over public opinion. Now, this isn’t about whether the war is right or wrong—it’s about exploring why Iran has the upper hand over us in the realm of discourse. The report states that Iranian online trolls have launched precise and diverse propaganda campaigns through social media platforms like Blue Sky. What is Blue Sky? It was originally an internal project at Twitter. After Musk acquired Twitter and X shifted to the right, the former CEO broke away to start his own platform, which became what we now know as Blue Sky. This platform has attracted a large number of left-wingers, whom the Chinese internet refers to as “white leftists.” In any case, Blue Sky has naturally become the opposition to Trump and the MAGA movement. Of course, I’m just relaying the facts—I’m not offering any commentary. The first step in the report is to identify bot accounts. If your account never posts anything but meets the following three criteria, you’re an Iranian bot: First, anti-Israel; second, pro-Palestine; third, anti-Trump. This is a bit tedious, but at least you can check yourself. However, very few people actually meet the third criterion—after all, who harbors such deep-seated resentment that they’d go out of their way to bash Trump? The report also mentions multilingual posting. Who would post in different foreign languages? Aside from those involved in global communications, it’s likely just foreign propaganda agencies. So, if we don’t want to be identified as paid trolls, we should avoid this kind of KPI-driven posting; otherwise, we’ll easily be flagged as accounts spreading harmful information. CSIS used AI tools to analyze the data and identified three major narrative themes. The first theme highlights how heroic Iran’s sacrifices are, how valiant its resistance is, and how successful its counterattacks are. For example, the report describes victories on the battlefield, including exaggerating military achievements, highlighting U.S. military casualties, using deepfake videos, and detailing how many U.S. military bases were destroyed, how many aircraft were shot down, how many Israeli cities were bombed, and how many global solidarity campaigns were launched in support of Iran. The second theme is how evil Israel is, that the war was entirely its own choice, and that it is manipulating the United States. The report also found that many usernames contained anti-Semitic terms, which I won’t list here one by one. The writing style of these users falls into two categories: inciting negative emotions and venting profanity. The study found that inciting negative emotions promotes sharing; moving from completely neutral to strongly negative content can increase shares by 57.7%, whereas profanity and vulgar language hinder sharing—for example, shifting from neutral to toxic language causes shares to drop by 62.4%. This is just as important for media work as it is for cyber warfare—you have to strike the right balance to spark resonance. The third theme is undermining the relationship between the public and the government in the U.S. and the West. These posts generated the highest engagement, averaging 150 shares, 470 likes, and 28 replies. This may not sound like much, but it generates greater resonance than the first two themes, with shares up 41%. The report also found that of these 19 core accounts, only two were truly central, linking 15 online communities. These communities interacted frequently with one another but had little contact with external users. Satellite accounts frantically repost content from the core accounts—especially those two major ones. I suspect these two might actually be Iranian patriotic bloggers. What’s the difference between a KOL and the leader of a paid comment army? Perhaps it’s just a matter of whose side they’re on. The report concludes by recommending that the U.S. must take the initiative in the digital realm to reclaim control of the narrative. First, prioritize the release of videos and images, using verified combat footage suitable for a broader audience to refute fake AI-generated videos; second, implement a tagging feature for official accounts and posts to flag hostile propaganda accounts, issue AI warnings, and identify non-authoritative information; third, collaborate with companies to crack down on fake accounts. My advice: if you’re worried your account might get caught in the crossfire, why not cultivate some hobbies outside of politics and post more about everyday life? All in all, this whole situation is a real mess, but we still need to figure out how to handle it. See you all next time!