Seeing Isn’t Believing: Disinformation and the Collapse of Verification in the Iran War

Contrary to the old adage, “seeing is believing”, disinformation has created a structurally disorienting environment where this no longer holds. Instead, we are witnessing the steady collapse of verification.

Typically, this verification relies on open-source intelligence, journalistic standards and institutional processes that involve[v] looking for visual inconsistences in images and videos, digital watermarks by using tools such as SynthID, and finding the source via reverse image searching to see where the content first appeared online.

Some social media companies have introduced safeguarding measures[vi] such as changing their community guidelines to prevent unverified information from being promoted to users’ feeds and flagging this content before users send it to others.

Similarly, the EU AI Act[vii], which becomes broadly applicable in August 2026, is the world’s first comprehensive legal framework regulating AI, imposing strict requirements on transparency and risk management. This includes informing users that they are interacting with a machine when using AI systems such as chatbots.

Much of the tech-focused analysis of the conflict has highlighted the presence of AI-generated visual content in spreading disinformation. Fabricated images of U.S. troops surrendering[viii] to Iranian forces, the destruction[ix] of critical infrastructure in Gulf cities, and videos of the aircraft carrier USS Abraham Lincoln[x] burning at sea provide examples.

Admittedly, the proliferation of AI-generated content already began blurring the lines between real and artificial content prior this war, during the 2022 Russian invasion of Ukraine[xi] and the Sudan civil war.[xii] The defining feature of the Iran War is not the mere presence of falsehood, but the erosion of mechanisms that once enabled one to distinguish fact from fiction.

Disinformation was systematically institutionalised by the KGB as a core tool of statecraft during the Cold War. It ran several active disinformation campaigns[xiii] that influenced many across the world throughout the 1970s and 80s, using forged documents, planted media narratives and proxy outlets to shape perceptions of the U.S., and the West more broadly.

However, through rigorous intelligence analysis and investigative reporting, these narratives were eventually exposed as a fabrications[xiv] and removed from credible discourse. The crucial point here is that the information environment in which disinformation spread was structurally different from what we experience today where verification still served its purpose as a corrective mechanism.

Most recently, several Republican politicians were misled into disseminating an AI-generated image[xv] that falsely depicted the rescue of the pilot of a downed U.S. warplane. Most striking is that the image retained credibility long enough for it to heavily influence and mould public and political discourse. By the time verification processes debunked the image and a warning was issued stating, “this photo is probably AI generated”, the factual status of the image had already become secondary to its narrative impact.