Generative AI at the Service of Propaganda

In the contemporary digital landscape, generative artificial intelligence is redefining the boundaries of content creation, extending into the domain of political communication and propaganda. An emblematic example is Explosive Media, a group of pro-Iran activists who have leveraged these technologies to produce a series of animated Lego-style videos targeting former U.S. President Donald Trump. These sophisticated contents emerged in a context of geopolitical tensions, shortly after a statement by Trump regarding the potential elimination of an “entire civilization.”

The use of AI to generate such animations underscores a growing trend: the ability to create complex visual narratives with relatively limited resources. This approach allows political messages to be conveyed in an engaging and viral manner, exploiting the familiarity and immediacy of popular formats like Lego bricks. The speed of production and scalability offered by generative AI represent a significant advantage for actors aiming to influence public opinion on a large scale.

Technical Details and Content Impact

Explosive Media's videos demonstrate a clear mastery of generative AI techniques applied to animation. One cited example includes a Trump mini-figure colluding with leaders from Gulf states, Iranian officials pressing a large red button labeled “back to the stone age,” and Trump throwing a chair at U.S. generals. These scenarios, while stylized, are designed to be immediately recognizable and provocative, amplifying the underlying political message through satirical humor and caricature.

The choice of the Lego style is not accidental: it offers a playful aesthetic that can mask the seriousness of the topics addressed, making the content more digestible and shareable for a broad audience. The ability to rapidly generate animated sequences with custom characters and environments is a testament to the advanced capabilities of current Large Language Models (LLM) and multimodal generative models. These tools allow textual scripts to be translated into visual outputs, drastically reducing the time and costs that would be required for traditional animation production.

Geopolitical Context and Dissemination

Since their debut, Explosive Media's videos have garnered millions of views across various mainstream platforms, demonstrating remarkable effectiveness in reaching a wide audience. While Iranian government-linked accounts have posted Lego-style videos in the past, Explosive Media's content stands out for its greater sophistication and more elaborate scripting. This suggests a more refined level of production and communication strategy, likely due to the expertise of a team of young pro-Iranian creators with deep knowledge of the internet and American culture.

The dissemination of such content raises important questions about data sovereignty and information management in digital environments. Platforms face the challenge of distinguishing between satire and propaganda, and of moderating AI-generated content that can be difficult to attribute with certainty. Allegations, made by some critics, of ties between the group and the Iranian government, add another layer of complexity, highlighting the need for robust mechanisms for verifying the origin and authenticity of digital media.

Future Prospects and Digital Security Challenges

The Explosive Media episode is a clear indicator of how generative AI is becoming an increasingly accessible and powerful tool for content creation, with significant implications for digital security and geopolitical stability. The ease with which sophisticated synthetic media can be produced makes it harder for the public to discern truth and for authorities to counter disinformation. This scenario necessitates a critical reflection on AI deployment strategies, not only in terms of efficiency and TCO, but also ethical responsibility and governance.

For organizations evaluating the deployment of LLMs and other generative models, whether on-premise or in hybrid environments, it is crucial to consider not only technical capabilities but also the potential social and political repercussions of using these technologies. The ability to generate persuasive, mass-produced content, while not directly linked to hardware specifications or throughput benchmarks, is a crucial aspect that will influence future decisions regarding AI infrastructure and usage policies. The need to ensure data sovereignty and control over models becomes even more pressing in an era where AI can be used to shape the perception of reality.