Digital Fog of War: How AI-Generated Fakes of Iranian Conflict Are Deceiving Millions and Reshaping Online Reality

The Viral Deception: AI’s Shadow Over the Iranian Conflict

A photograph of an Iranian schoolgirl graveyard, seemingly bombed and overflowing with freshly dug graves, resonated across the globe, sparking outrage and grief. The Guardian’s headline questioned the image’s authenticity: “A photo of Iran’s bombed schoolgirl graveyard went around the world. Was it real, or AI?” This viral moment underscored a disturbing trend, as headlines from CNN and The New York Times have also highlighted: AI-generated fakes are impacting the war with Iran, causing chaos online, and reshaping our understanding of reality. The spread of such deceptive visuals isn’t limited to Iran; a separate incident saw fake AI images circulating online, falsely depicting the main civilian terminal on fire during a militant attack on Niamey airport in Niger.

The Technology: How Generative AI Creates Convincing Wartime Fakes

Generative AI has significantly amplified the capability of state actors to fabricate highly convincing satellite imagery during conflicts. Consider the fake satellite image disseminated by the state-aligned Iranian daily, Tehran Times, which purported to show a devastated US base in Qatar. Researchers quickly exposed it as an AI-manipulated version of a Google Earth image of a US base in Bahrain from the previous year. The telltale signs included cars parked in identical positions in both the authentic and the manipulated images. Another AI-generated image surfaced, claiming Israeli-US jets had targeted a painted aircraft silhouette in Iran; clues revealing its falsity included gibberish coordinates embedded within the image. In a crucial development for detection, Agence France-Presse (AFP) has successfully identified SynthID, an invisible watermark from Google, on some of these fake images, definitively marking them as AI-created.

The Impact: Undermining Truth and Fueling Propaganda

The proliferation of these AI fakes carries profound implications, primarily by undermining truth and fueling propaganda. Disinformation agents are actively exploiting the credibility of Open-Source Intelligence (OSINT) by fabricating imposter accounts and distributing manipulated images. This tactic preys on the public’s trust in seemingly objective visual evidence. As Bo Zhao, a professor from the University of Washington, observed, “When a satellite image is presented as visual evidence in the context of war, it can easily influence how people interpret events”. This erosion of trust makes it increasingly difficult for the public to discern reality from fabricated narratives. Melanie Smith, ISD Senior Director, echoed this concern, stating, “The inability to get access to verified and credible information in times like this — it’s getting harder and harder to do that.” The problem isn’t confined to static images; misinformation also manifests in videos, such as a widely circulated clip claiming to show an Iranian plane strike that was, in fact, footage from a video game.

The Response: Efforts by Platforms and Fact-Checkers to Combat Disinformation

Social media platforms are beginning to react to this surge in AI-generated disinformation. X (formerly Twitter), for instance, recently announced a new policy: creators who post undisclosed AI-generated war videos will face a 90-day suspension from its revenue-sharing program. This measure aims to disincentivize the spread of unverified AI content. Despite this new policy, researchers note that online feeds remain heavily “flooded with AI content about the war,” indicating the persistent challenge in controlling the flow of such material. Fact-checkers and independent researchers continue to play a critical role, meticulously analyzing digital breadcrumbs like identical car placements or gibberish coordinates to expose the fakes. However, the sheer volume and increasing sophistication of AI-generated content demand a more robust, multi-faceted approach from platforms, governments, and users alike.

---

Frequently Asked Questions

Q1: How can I identify an AI-generated fake image?
A1: Look for subtle inconsistencies such as repeated patterns (e.g., identical cars in different locations), blurry or distorted details in the background, unnatural lighting, odd angles, or gibberish text in what should be legible signs or coordinates. Tools like Google’s SynthID can also help detect invisible watermarks on AI-generated content.

Q2: Are only images being faked, or other forms of media too?
A2: No, misinformation extends beyond images. Videos are also being faked, with examples including clips purporting to show real conflict events that are actually sourced from video games. The rise of generative AI affects various media types, making it harder to distinguish between authentic and fabricated content.

Q3: What role does Open-Source Intelligence (OSINT) play in this “digital fog of war”?
A3: OSINT, traditionally a valuable tool for verifying information during conflicts, is now being exploited by disinformation agents. They create imposter accounts and manipulate images to spread false narratives, preying on the inherent credibility that OSINT once offered. This makes the work of legitimate OSINT researchers even more critical but also more challenging.

---

What responsibility do individual social media users have in preventing the spread of AI-generated misinformation during wartime?

---

Related Topics: Artificial Intelligence, Iran, Disinformation