By Sureena Sokhal
When Russia invaded Ukraine, social media flooded with fabricated imagery, photos that “were presented as fresh images of the war” but were either doctored originals or “mislabeled clips from video games, movies, past incidents or unrelated news coverage.”[1] That same phenomenon is now playing out with the war with Iran.[2]
Professor Hany Farid, a digital forensics expert at UC Berkeley, puts it plainly: ten years ago, a deepfake circulating online was obvious and could be quickly debunked.[3] However today, the sheer volume of manipulated content and the realism now achievable with generative AI tools make it far easier for people to accept what they are seeing as real.[4] These altered images and videos have become “hard[er] to detect to the untrained or naked eye.”[5]
Shayan Sardarizadeh, a Senior Journalist at BBC Verify, has noted that AI-generated fakes “racked up millions of views on social media” in just the first two weeks of the conflict.[6] The content ranges widely: videos falsely depicting Iranian missiles striking Tel Aviv, footage showing panicked crowds at Tel Aviv’s airport, and clips purporting to show Iranian military facilities being destroyed.[7] Some of this footage is recycled from unrelated events last year, while other videos are digitally altered to make old clips appear current.[8] New fakes appear online far faster than platforms can take them down. Some of this content may be deliberate propaganda pushed through state-linked accounts; however, finding a motive is rarely straightforward.[9]
Nikita Bier, Head of Platform at X, stated that paid content creators who share AI-generated videos without disclosing them as such will be “suspended from the payment program for 90 days and then permanently suspended if they commit additional violations.”[10] Even so, that policy only covers a narrow slice of users and the vast majority of people spreading this content are not paid or sponsored creators, and no such consequences apply to them.[11]
Much of the responsibility for moderating misleading digital content falls on social media platforms themselves.[12] Under the Communications Decency Act, online platforms are generally shielded from liability for content posted by their users.[13] Section 230 provides that an interactive computer service cannot be treated as the publisher or speaker of information provided by another content provider.[14] Courts have interpreted this provision broadly to protect platforms from liability arising from most user-generated content.[15] Although the statute was originally designed to promote innovation and free expression on the internet, it also means that platforms are often not legally responsible when manipulated images or AI-generated videos are shared by users.[16] As a result, the spread of synthetic war footage frequently falls outside the reach of traditional legal liability, leaving platforms themselves as the primary moderators of misleading digital content.[17]
Deepfakes can pose a significant threat to democratic discourse because they allow fabricated media to appear authentic and circulate widely online.[18] Scholars further note that these risks become especially pronounced during moments of geopolitical crisis or armed conflict, when misinformation may spread rapidly and influence public perception.[19] As generative AI tools become more accessible, the potential for deceptive media to shape narratives surrounding international conflicts will likely continue to grow.[20]
For readers trying to “avoid being duped,” experts suggest a few practical steps. Look for visual red flags like “misplaced limbs” or “extra fingers,” which are common artifacts in AI-generated images.[21] Pay attention to audio that seems out of sync with the visuals, or background details that look slightly off.[22] Most importantly, get news from “credible journalistic outlets” rather than from accounts that surface while scrolling through a feed.[23] If something feels staged or too dramatic to be real, that instinct is often worth trusting.[24]
Sureena Sokhal is a 2L at Vanderbilt Law School. She plans on general corporate transactional work after law school.
[1] Daniel Dale, Fake Explosions, Fake Missiles, Fake Troops: AI Videos and Images of Iran War Spread Widely on Social Media, CNN Politics (Mar. 11, 2026, at 02:00 ET), https://www.cnn.com/2026/03/11/politics/fake-ai-images-videos-iran-war.
[2] See id.
[3] Id.
[4] Id.
[5] Id.
[6] Id.
[7] Id.
[8] Id.
[9] See id.
[10] Id.
[11] Id.
[12] See Valerie C. Brannon & Eric N. Holmes, Section 230: An Overview, Cong. Rsch. Serv. R46751 (2024).
[13] Communications Decency Act of 1996, 47 U.S.C. § 230(c)(1) (2024).
[14] Id.
[15] Zeran v. Am. Online, Inc., 129 F.3d 327, 330–31 (4th Cir. 1997).
[16] Jeff Kosseff, The Twenty-Six Words That Created the Internet 7–9 (2019).
[17] Brannon & Holmes, supra note 12.
[18] Id. at 1758–60.
[19] Robert Chesney & Danielle Citron, Deep Fakes and the New Disinformation War: The Coming of Age of Post-Truth Geopolitics, Foreign Affs. (Dec. 11, 2018), https://www.foreignaffairs.com/articles/world/2018-12-11/deepfakes-and-new-disinformation-war.
[20] Danielle Keats Citron & Robert Chesney, Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security, 107 Calif. L. Rev. 1753, 1760–62 (2019).
[21] Dale, supra note 1.
[22] Id.
[23] Id.
[24] See id.