YouTube Censors Pro-Iranian ‘Lego’ Video Mocking Trump Amid Escalation of AI Propaganda War
Iran has condemned the removal of a YouTube channel that used AI technology to create Lego animated parody videos featuring President Trump (who was depicted as a bumbling lego man). The Iranian Foreign Ministry spokesperson stated the ban was an attempt to hide the “truth” about the Israeli-U.S. war and protect what they described as the “American Administration’s Narrative”.
YouTube cited its violation of violent content policy for removing the channel; however, YouTube did not provide specific examples of videos that caused the removal. According to reports from France 24, the removal of the channel represents a new phase of what they describe as a “slopaganda battle”, which is the first global information war waged with low-cost AI technologies.
In addition to being unusual for its transparency, the creators of the Explosive Media channel recently admitted that the Iranian Government is directly paying for their work. This admission occurred after weeks of speculation surrounding the source of funding for the channel due to the high quality and speed of production of the channel. The producers typically release at least three videos daily.
The videos fall somewhere in the middle of a range between children's toy commercials and political attack ads. In each scenario, Trump is portrayed as a clumsy Lego figure attempting to navigate mazes (labeled Middle East Policy); he is pursued by toy missiles; or he sits confused as his advisors – created out of blocks – argue around him. It appears that choosing an iconic children's toy brand to convey political messages deemed too violent by YouTube for their platform, is intentional.
According to Marina Koren, Digital Conflict Researcher at Stanford University's Internet Observatory, state actors are discovering what YouTubers have understood for years – that ironic detachment and meme aesthetics can be much farther reaching than traditional forms of propaganda. Additionally, she states, “using a popular children's toy brand as a way to disguise their intentions allows them to claim plausible deniability regarding whether their content is satirical or threatening.”
Get the latest model rankings, product launches, and evaluation insights delivered to your inbox.
This timing may be significant. The videos were released during a period when relations between Iran and Israel were escalating rapidly. Both countries have engaged in recent military operations and diplomatic exchanges. Explosive Media positioned themselves as counter programming against Western media coverage; however, their approach presents a completely new paradigm in terms of state sponsored messaging.
AI-generated toy soldiers instead of traditional news broadcasts.
Although YouTube appears to have struggled in determining whether these videos represented legitimate satire or genuine threats; YouTube's automated content moderation systems may not have fully accounted for the unique aspects of this format. Specifically, YouTube's spam and deceptive practices policies specifically prohibit content that is technically altered or fabricated in such a manner that it deceives users. However, as reported by Ground News, the videos referenced above openly admitted that they were generated using AI.
Within days of YouTube banning Explosive Media, multiple other channels were established featuring similar Lego-style parodies of international politics. For example, within hours of YouTube's ban, similar types of videos began showing up on TikTok, Telegram, and other social media outlets. Many of these videos referenced the ban and encouraged viewers to watch “The Videos That YouTube Does Not Want You To Watch,” among other names. The distinctive visual aesthetic of these AI-generated characters has become synonymous with this emerging form of political content.
Iran's reaction demonstrates why there is so much at stake. The country's foreign ministry characterized YouTube's actions as part of a larger western effort to control narratives surrounding current conflicts. By doing so, the Iranian government has positioned AI-generated content as a means of achieving narrative sovereignty -- i.e., allowing state actors to circumvent traditional media gatekeepers.
The technical specifics are important here. There is evidence that Explosive Media uses common off-the-shelf AI video creation tools. More than likely, they combine these with some level of animation software. The Lego style provides an advantage -- the simple and blocky nature reduces many of the typical AI-created video problems (i.e., temporal inconsistencies and morphing objects).
State actors are beginning to utilize meme aesthetics and irony that were previously utilized by independent creators; therefore, creating confusion concerning attribution and purpose. Furthermore, platforms' attempts to moderate AI-generated content that openly acknowledges its artificial nature, yet continues to disseminate political messaging creates further challenges. Low barriers to entry make banning propagandistic AI-generated content easy and quick to replicate across all platforms. Ultimately, once these visual styles (e.g., Lego propaganda) become easily recognized branding beyond individual channels or creators -- platforms will need to establish methodologies for moderating synthetic content that acknowledge its inherently distinct differences from both actual footage and traditional animation.
As additional state and non-state actors begin to realize the propaganda capabilities inherent in AI-generated content -- platforms will be required to make increasingly difficult moderation choices.