AI-generated video intended to spread disinformation
In an age when a person's voice and face can be borrowed without consent, India's Ministry of External Affairs stepped forward to reclaim the truth: a senior Army officer had been made to say things he never said, his image weaponized to suggest India had chosen sides in one of the world's most combustible conflicts. The fabrication was sophisticated, the correction swift, and the episode a quiet reminder that the integrity of public discourse now depends as much on institutional vigilance as on the honesty of those who speak.
- An AI-generated deepfake placed fabricated geopolitical declarations in the mouth of Major General CS Mann, falsely claiming India was aligning with Israel and the U.S. against Iran.
- The video was polished enough to circulate convincingly on social media, threatening to mislead domestic and international audiences about India's official foreign policy stance.
- India's MEA Fact Check unit moved quickly, publishing a side-by-side comparison on X that exposed the manipulation by contrasting the fake clip with the original July 4 statement on drone security.
- The real statement contained no geopolitical claims whatsoever — only technical discussion of drone capabilities, emergency procurement, and vulnerabilities posed by Chinese drone technology.
- The government's rapid response controlled the narrative, but the incident raises an unresolved question: whether institutional fact-checking can keep pace with the accelerating volume of AI-generated disinformation.
On a Friday in March, India's Ministry of External Affairs issued a warning that cut through a spreading falsehood: a video on social media purporting to show Major General CS Mann, Additional Director General of the Army Design Bureau, had been fabricated using artificial intelligence. In the altered clip, Mann appeared to describe a sweeping regional strategy involving Pakistan, Afghanistan, Bangladesh, and Nepal, and claimed India was actively supporting Israel and the United States against Iran — statements he never made.
The MEA's Fact Check unit responded by publishing a direct comparison on X, placing the deepfake beside the authentic recording. The contrast was unambiguous. The real statement, delivered on July 4, was entirely technical in nature — focused on India's drone and counter-drone capabilities, emergency procurement procedures, and security concerns about Chinese drone technology. Nothing in it touched on geopolitics or foreign military alignment.
The ministry's public message was blunt: the video was AI-generated disinformation, and citizens were urged to remain alert. The warning carried a deeper implication — that the technology for fabricating convincing statements attributed to real, senior officials had matured into a genuine threat to public understanding of military and diplomatic affairs.
What distinguished this incident was its target and its ambition. Rather than a generic hoax, the deepfake attempted to reframe India's position on one of the world's most volatile conflicts, a claim that could have inflamed tensions or distorted perceptions of official policy if left unchallenged. The speed of the government's response suggested active monitoring of social media for precisely this kind of manipulation.
The episode left an open question for those watching the intersection of technology and information warfare: whether rapid institutional fact-checking can sustain its effectiveness as AI-generated disinformation grows faster and more voluminous than any single body's capacity to answer it.
On Friday, India's Ministry of External Affairs released a warning that stopped a false narrative in its tracks: a video circulating on social media purporting to show Major General CS Mann, Additional Director General of the Army Design Bureau, making sweeping geopolitical claims about India's military alignment with Israel and the United States against Iran was entirely fabricated.
The deepfake, constructed using artificial intelligence, had placed words in the officer's mouth that he never spoke. In the altered version, Mann appeared to outline a strategy involving Pakistan, Afghanistan, Bangladesh, and Nepal, and claimed India was actively assisting Israel and the U.S. in confronting Iran. The video was designed to look authentic enough to fool casual viewers scrolling through social media feeds.
The MEA's Fact Check unit responded by publishing a side-by-side comparison on X, showing the fabricated clip alongside the actual recording. The contrast was stark. The real statement, delivered on July 4, contained no geopolitical declarations whatsoever. Instead, Mann had discussed India's drone and counter-drone capabilities, the government's emergency procurement procedures for security equipment, and vulnerabilities the country faced—including specific concerns about Chinese drone technology. The substance was technical and defensive, not strategic and inflammatory.
The ministry's warning was direct: "This is an AI-generated video intended to spread disinformation! Please stay alert against such fake videos and content on social media." The message carried an implicit acknowledgment that the technology for creating convincing false statements attributed to real people had matured enough to pose a genuine threat to public discourse, particularly around sensitive military and diplomatic matters.
What made this incident noteworthy was not merely that a deepfake existed—such fabrications have become increasingly common—but that it targeted a senior military official and attempted to reframe India's foreign policy stance on one of the world's most volatile geopolitical questions. The false video suggested India had taken a definitive position on the Israel-Iran conflict, a claim that could have inflamed tensions or misled audiences about official Indian policy if left unchallenged.
The speed of the MEA's response suggested the government was monitoring social media for exactly this kind of manipulation. By publishing the comparison themselves, they controlled the narrative and provided citizens with the evidence needed to recognize the fabrication. The incident underscored a new reality: in an era when artificial intelligence can convincingly mimic a person's voice and likeness, institutional credibility depends partly on the ability to quickly expose and debunk false content before it spreads beyond recovery.
For those watching the intersection of technology and information warfare, the episode served as both a cautionary tale and a demonstration of defensive capability. The deepfake had been created with enough sophistication to fool many viewers, yet the original statement remained preserved and retrievable. The question moving forward was whether such rapid-response fact-checking could keep pace with the accelerating production of AI-generated disinformation, or whether the sheer volume of false content would eventually overwhelm institutional capacity to respond.
Citas Notables
This is an AI-generated video intended to spread disinformation. Please stay alert against such fake videos and content on social media.— India's Ministry of External Affairs Fact Check unit
La Conversación del Hearth Otra perspectiva de la historia
Why would someone go to the trouble of creating a deepfake of a military officer talking about Israel and Iran? What's the goal?
To shift how people perceive India's foreign policy without India actually saying anything. If the video spreads widely enough before being debunked, it plants doubt and confusion about what India's real position is.
But the MEA caught it quickly. Doesn't that mean the system works?
It caught this one. But there are probably dozens more being created right now that no one has noticed yet. The MEA can only fact-check what they see or what gets reported to them.
The original statement was about drones and procurement—pretty mundane stuff. Why would anyone bother faking that?
Exactly. That's what makes it effective. The real statement is boring and technical, so most people wouldn't share it. But a statement about India secretly helping Israel against Iran? That spreads like wildfire.
So the deepfake is more interesting than the truth.
Much more interesting. And in a world where people skim headlines and share without reading, interesting wins.
What happens to the people who already saw the fake video and believed it?
Most of them probably never see the correction. They've already formed an opinion and moved on.