The explosion of AI-generated misinformation surrounding the U.S.-Israel war with Iran has reached levels that digital media experts and fact-checking organizations are calling unprecedented, raising urgent questions about the reliability of online content and the tools designed to detect it.
Digital media expert Timothy Graham told the BBC that the barrier to creating convincing synthetic conflict footage has effectively collapsed. “What used to require professional video production can now be done in minutes with AI tools,” he said.
Sofia Rubison, a senior editor at NewsGuard, an organization that rates the reliability of global news sources, said the volume of fake videos and photos being spread online represents a clear increase from the past. “I definitely think so,” she said when asked whether the current level of AI-generated video is new, speaking on the podcast Question Everything.
The challenge extends to the tools available for detecting AI-generated content. Rubison noted that Grok, the AI tool integrated into the social media platform X, is “actually one of the biggest spreaders of false claims on this platform,” adding that X does not claim its model can accurately fact-check content or reliably detect AI-generated material. Despite that, Grok’s responses can feel like definitive verdicts to users.
A detector from the company Hive, considered among the more accurate tools available, still returned a greater than 95% likelihood that a recent video of Israeli Prime Minister Benjamin Netanyahu at a cafe was AI-generated — a determination that turned out to be wrong. Reuters had verified the video by examining stock footage from the cafe that matched the background, and the cafe itself posted corroborating photos and videos on social media.
Rubison said NewsGuard uses Hive for initial assessments but never relies solely on one platform when determining whether a video is AI-generated. The Netanyahu case illustrates why: even the most accurate automated detectors are not infallible, and thorough fact-checking requires human verification across multiple independent sources.
The danger is amplified in conflict contexts, where images and videos carry particular persuasive power. Research has shown that people are far less skeptical of content they believe they have seen with their own eyes, making synthetic conflict footage a potent tool in what experts describe as a growing “misinformation economy” — the monetization of false content at scale.
International Fact-Checking Day, observed annually on April 2, was established as a counterweight to the spread of misinformation. Experts say the day serves as a reminder to think critically about the images and videos that populate social media feeds and to incorporate fact-checking resources into regular information consumption.
