A wave of disinformation has spread online since the beginning of Israeli strikes on Iran, with dozens of posts reviewed by BBC Verify seeking to amplify the effectiveness of the Iranian response using AI-generated videos, in the first widespread use of generative artificial intelligence during a conflict.
Article contents:
- Introduction
- AI disinformation
- Viral fake clips
- Focus on F-35 jets
- Pro-Israeli content
- Official sources and disinformation
- Grok problems and verification
- Conclusion
- Frequently asked questions
Introduction
A wave of disinformation has spread online since the beginning of Israeli strikes on Iran last week, with dozens of posts reviewed by BBC Verify seeking to amplify the effectiveness of Tehran’s response. The top 3 fake videos found by BBC Verify together garnered more than 100 million views across multiple platforms.
Pro-Israeli accounts also shared disinformation online, primarily by recirculating old clips from protests and gatherings in Iran, falsely claiming they showed growing opposition against the government and support among Iranians for the Israeli military campaign.
AI disinformation
Israel launched strikes on Iran on June 13, leading to several rounds of Iranian missile and drone attacks on Israel. One organization analyzing open-source imagery described the volume of online disinformation as “staggering” and accused some “engagement farmers” of seeking to profit from the conflict by sharing misleading content designed to attract attention online.
GeoConfirmed, an online verification group, wrote on X: “We are seeing everything from irrelevant footage from Pakistan, to recycled videos from the October 2024 strikes—some of which have garnered over 20 million views—as well as gaming clips and AI-generated content presented as real events.”
Certain accounts became “super-spreaders” of disinformation and were rewarded with significant growth in their follower counts. A pro-Iranian account with no clear ties to authorities in Tehran—Daily Iran Military—saw its followers on X grow from just over 700,000 on June 13 to 1.4 million by June 19, a 100% increase in less than a week.
It is one of several mysterious accounts that have recently appeared in people’s feeds. All of them have blue checks, are active in messaging, and frequently post misinformation. Because some use official-sounding names, some people have assumed they are authentic accounts, but it remains unclear who actually runs these profiles.
Emmanuel Saliba, chief investigative officer at the analytical group Get Real, described the flood of disinformation to BBC Verify as “the first time we have seen generative AI used on a large scale during a conflict.”
Focus on F-35 jets
Accounts reviewed by BBC Verify heavily shared AI-generated images that appeared to seek to amplify the success of Iran’s response to the Israeli strikes. One image, which received 27 million views, depicted dozens of missiles falling on the city of Tel Aviv.
Another video claimed to show a missile strike on a building in the Israeli city late at night. Ms. Saliba said the clips often depict nighttime attacks, making them particularly difficult to verify.
The AI fakes also focused on claims that Israeli F-35 fighter jets—advanced US-made aircraft capable of striking ground and air targets—had been destroyed. If the flurry of clips were real, Iran would have destroyed 15% of Israel’s fighter jet fleet, Lisa Kaplan, CEO of the analytical group Alethea, told BBC Verify. We have not yet been able to document any footage of an F-35 being shot down.
Pro-Israeli content
Disinformation is also being posted by well-known accounts that previously intervened in the Israel-Gaza war and other conflicts. Their motives vary, but experts said some may be trying to monetize the conflict, with some major social media platforms offering payouts to accounts that generate high view counts.
Conversely, pro-Israeli posts focused largely on suggestions that the Iranian government faces growing opposition as strikes continue. Among them was a widely circulated AI-generated video falsely claiming to show Iranians chanting “We love Israel” in the streets of Tehran.
Official sources and disinformation
Official sources in both Iran and Israel shared some fake images. State media in Tehran shared fake strike footage and an AI-generated image of a downed F-35 jet, while a post shared by the Israel Defense Forces received a community note on X for using old, irrelevant footage of a missile barrage.
Much of the disinformation reviewed by BBC Verify was shared on X, with users frequently turning to the platform’s AI chatbot—Grok—to determine the authenticity of posts.
Grok problems and verification
However, in some instances, Grok insisted that AI-generated videos were real. One video showed an endless stream of trucks carrying ballistic missiles emerging from a mountain complex. Markers of AI-generated content included rocks in the video moving on their own, Ms. Saliba said.
Yet, in response to X users, Grok repeatedly insisted the video was real and cited reports from media outlets including Newsweek and Reuters. “Check trusted news for clarity,” the chatbot concluded in several messages. X did not respond to a request from BBC Verify for comment on the chatbot’s actions.
Conclusion
While the motivations of those creating online fakes vary, many are shared by ordinary social media users. Matthew Fachiani, a researcher at the University of Notre Dame, suggested that disinformation can spread more rapidly online when people face binary choices, such as those provoked by conflicts and politics. This points to the broader social and psychological issue of people wanting to reshare things if they align with their political identity, as well as the general rule that more emotionally provocative content will spread faster online.
Frequently asked questions
1. What is the scope of disinformation in the Israeli-Iranian conflict?
The top 3 fake videos garnered more than 100 million combined views, making this the first widespread use of generative AI during a conflict.
2. What types of fake content are most prevalent?
They include AI-generated videos showing fake missile strikes, false claims regarding the destruction of F-35 jets, and recycled clips from past events.
3. How does this disinformation affect public opinion?
It spreads more rapidly when it aligns with people’s political identities, exploiting emotions and divisions to drive views and engagement.
