War has always needed a story. 

Long before algorithms, viral videos, or artificial intelligence, the United States and other belligerent powers understood that winning a war was not only about territory, but perception. In the twentieth century, the U.S. refined this understanding into a sophisticated system of narrative production, creating institutions and strategies to shape public opinion across the World Wars, Vietnam, and the Iraq Wars and transforming posters, cinema, and mass media into tools designed to align entire populations with the logic of war.

Over time, this system evolved into something more subtle and more powerful: the fusion of geopolitics and entertainment. Through films, television, and digital media, the United States has consistently presented itself as a defender of freedom, a reluctant warrior, and, ultimately, a global saviour. These strategies of military propaganda still inhabit the collective psyche of the West. People, including me, are deeply familiar with the large-scale cinematic productions and video games portraying “Yankee heroes” at war and with the “iconic Uncle Sam” posters urging citizens to enlist. This is not incidental. It reflects what propaganda, in its classical sense, seeks to achieve: the systematic shaping of beliefs and attitudes through symbols, narratives, and emotional resonance.

For decades, this narrative dominance remained largely uncontested. But the digital era has altered the conditions of storytelling. The emergence of AI-driven content, meme culture, and decentralized media production has opened space for new actors to challenge established narratives. In this context, phenomena such as “Explosive Media” in Iran and its leading figure, known as “Mr. Explosive”, represent more than isolated cases: they signal a structural shift in how propaganda is created, distributed, and consumed.

Propaganda Game Changers

The traditional model of propaganda relied on resources: institutional backing, professional production, and control over distribution channels. Today, that model has been disrupted. Generative artificial intelligence has drastically reduced the cost and time required to produce persuasive content, giving rise to what scholars describe as Slopaganda: AI-generated material designed to influence beliefs at scale, rapidly and with minimal resources.

Explosive Media illustrates this transformation with particular clarity. Its “Lego-style” videos merge humor and political messaging into a format that is immediately recognizable and highly shareable. As described in The New Yorker, “the clips have accumulated millions of views and many enthusiastic comments from Western audiences,” while “the political messaging on display in the videos is as blunt and cartoonish as the blocky Lego characters.” This combination of visual familiarity, satire, narrative framing, and the remarks on social injustices allows the content to embed references to global events and controversial issues within a broader storyline that often portrays the United States as an oppressive force, reaching audiences well beyond the immediate geopolitical context in which the videos are produced.

What makes this development especially significant is not only the content itself, but its cultural adaptability. AI tools (largely trained on Western datasets) enable the production of content that feels native to Western digital culture. This allows such material to circulate more easily across global platforms, blending seamlessly with entertainment, satire, and everyday online communication. At the same time, the actors behind this content have acknowledged links to state interests, revealing the increasingly blurred boundary between grassroots digital expression and coordinated political influence, as the university of Manchester remarked, “These videos contain disinformation and antisemitic tropes but do not look or feel like state propaganda—despite the spokesperson for Explosive Media admitting to the BBC that the Iranian government is a client”.

This creates a new kind of “informational asymmetry”. While the United States has historically dominated global narratives through both institutional and cultural power, new actors can now challenge that dominance using tools that are cheaper, faster, and more adaptable. However, this challenge does not occur on neutral ground. The infrastructure through which these narratives circulate (Google, YouTube, Meta: Facebook, Instagram, etc., X & TikTok) is itself governed by a small number of private actors, primarily based in the United States, the “Techno-bros”.

This raises a critical issue: even if the production of propaganda has been democratized, its distribution remains centralized.

Why is this important for human rights?

From a legal and ethical perspective, the rise of AI-driven propaganda exposes a fundamental tension between freedom of expression and the prevention of harm.

On one hand, the ability to produce and share content is a core component of democratic participation. AI tools can empower individuals and groups to communicate more effectively, challenge dominant narratives, and engage in global-scale debates. On the other hand, the same tools can be used to amplify disinformation, normalize violence, and spread hate speech at an unprecedented scale and speed.

This duality is particularly relevant in conflict contexts, where narratives are not neutral but are directly tied to questions of legitimacy, responsibility, and human rights. The portrayal of one actor as a victim and another as an aggressor can influence international opinion, policy responses, and even the perception of what constitutes a war crime.

Current platform regulations attempt to address these risks, but they remain fragmented and inconsistent, as we saw in Iran’s “Explosive Media” case (most of them banned the content, except X and other non-big platforms, but the motives behind the ban remain subjective and do not quite fit the platforms’ own enforcement policies). For example, YouTube requires disclosure of synthetically generated or significantly altered content when it appears realistic. But, according to the Jerusalem Post, the reason for banning the videos was the violation of the spam, deceptive practices, and scams policies, which resulted in the termination and removal of the channel from the platform. Meta has shifted toward a model that prioritizes transparency through labeling rather than removal, except in cases where other policy violations occur, That is why the Instagram account was temporarily blocked for a few hours, but was later reinstated, according to The National. X, meanwhile, restricts deceptive synthetic media likely to cause harm and may apply labels to provide context, however, none of the restrictions established by X were applied in the case of Explosive Media, perhaps because the potential for harm from synthetically generated videos is not presumed, or due to the well-known recent rivalry between the platform’s owner Elon Musk and the President of the United States Donald Trump.

These approaches reflect different interpretations of the same dilemma: should harmful AI-generated content be removed, labeled, or allowed to circulate in the name of free expression? 

In practice, the lack of a unified standard creates regulatory gaps. Content removed from one platform can reappear on another. Accounts can be suspended and recreated. Narratives can be reshaped and redistributed faster than any moderation system can respond. The result is an environment in which governance is reactive, while the production of content is proactive and continuous.

Based on the evidence, it can be argued that this is where the real issue lies. The problem is not simply that propaganda exists (it always has). The problem is that the technological conditions under which it operates have changed, and regulatory frameworks are still catching up. Although California’s AB 2013 now mandates training data transparency,  and the EU AI Act has banned manipulative AI (European Union, 2024, Article 5), we remain in a “regulatory lag and fragmentation” where the tech is fully operational, yet comprehensive global labeling mandates will not be fully enforceable until August 2026 because it is the “Day Zero” for AI transparency in the Western world. It marks the moment the European Union’s AI Act moves from being a set of guidelines to a strictly enforced legal mandate for content disclosure. Nevertheless, this patchwork of laws falls short of a universal standard for protecting human rights against the potential harms of AI-driven misinformation.

Structural transformation

For much of modern history, the United States perfected the art of narrative dominance in war, using media, culture, and communication strategies to frame itself as a force of order, justice, and liberation. That dominance is now being contested and not necessarily by more powerful states, but by more agile and intelligent forms of communication.

AI has changed the economics of influence; it has made propaganda faster, cheaper, and more adaptable. It has blurred the line between entertainment and political messaging. And it has exposed the limitations of existing regulatory systems.

But perhaps the most important change is conceptual, because we can no longer think of war as something that happens only on the ground. It also happens on the internet: the stories people believe, the images they share, and the narratives that shape their understanding of reality. 

This is why the question is no longer whether modern propaganda should exist, but how it should be governed.

We need to establish clearer and more consistent rules for this new informational environment. Rules that ensure a level playing field, where no actor can unilaterally control visibility. Rules that protect freedom of expression while setting meaningful limits on manipulation and harm. Rules that recognize that digital platforms are not neutral intermediaries, but central actors in the construction of global narratives.

Because in the end, the most powerful weapon in modern conflict may not be a missile or a drone, it may be the story (and the words) that survive.

Victor Manuel Haro Gutiérrez: A Mexican lawyer and human rights activist graduated from the University of Guadalajara, with certifications and experience in litigation and politics. He is currently pursuing the European Master in Law, Data, and Artificial Intelligence (EMILDAI).