Hero image: Adi Rajpoot / Pexels

Pakistani Propaganda Fact Check Video

India’s Press Information Bureau has debunked a viral deepfake video purporting to show Prime Minister Narendra Modi threatening protesters at Delhi’s Jantar Mantar. Independent fact-checkers at Livemint and Times Now have corroborated the PIB’s findings, revealing coordinated disinformation tactics linked to transnational propaganda networks. The episode underscores how synthetic media is weaponized in regional geopolitical conflicts.

India’s Press Information Bureau (PIB) has formally debunked a widely circulated video that falsely depicts Prime Minister Narendra Modi threatening protesters at Delhi’s Jantar Mantar. The video, which spread rapidly across social media platforms, was flagged as a deepfake by government fact-checkers and independent media outlets. This incident is the latest in a series of disinformation campaigns that Indian authorities and journalists have attributed to coordinated efforts originating from Pakistan-based propaganda networks. Rather than treating the video as an isolated hoax, the episode reveals a broader pattern in which synthetic media is deployed to inflame domestic tensions in India and to shape international perceptions of Indian governance. This synthesis examines reporting from Livemint and Times Now alongside the PIB’s official statement to assess the video’s provenance, the mechanics of its fabrication, and the institutional responses it has provoked.

Introduction to Pakistani Propaganda Tactics

Pakistan has long been identified by Indian authorities and independent researchers as a source of state-linked disinformation campaigns aimed at destabilizing domestic politics in India. These operations typically exploit social media ecosystems, leveraging both automated bot networks and human-operated accounts to amplify divisive narratives. According to India’s Ministry of Electronics and Information Technology (MeitY), such campaigns often target religious minorities, political protests, and high-profile public figures to stoke communal tensions and erode public trust in institutions. The Jantar Mantar deepfake incident follows a documented playbook: a fabricated video is seeded on fringe platforms, then amplified through coordinated sharing by inauthentic accounts, before being picked up by mainstream media and political commentators who amplify it without verification. This cycle of disinformation is not unique to India; it mirrors tactics observed in other regional conflicts, including Ukraine and Taiwan, where synthetic media is used to blur the line between satire, propaganda, and state-sponsored information warfare.

The Indian government has increasingly framed these operations as part of a broader “Pakistani propaganda” strategy, citing evidence from digital forensic reports and social media analytics. While Pakistan denies direct involvement, Indian cybersecurity agencies have traced the infrastructure of several disinformation campaigns to servers located in Pakistan or operated by entities with known links to Pakistani intelligence or affiliated groups. These campaigns are not always overtly political; some focus on economic disinformation, such as false claims about Indian currency devaluation or stock market manipulation, designed to undermine investor confidence. The Jantar Mantar deepfake, however, represents a direct attempt to influence public perception of a domestic protest movement by falsely implicating the prime minister in a coercive act.

What Livemint and Times Now are Reporting on the Digitally Altered Video

Both Livemint and Times Now conducted independent forensic analyses of the video in question, which purports to show Prime Minister Narendra Modi addressing a gathering near Jantar Mantar and threatening protesters with unspecified consequences. Livemint’s report emphasizes the role of the Press Information Bureau (PIB) in issuing a formal fact-check, noting that PIB’s Fact Check Unit (FCU) concluded the video was “digitally altered” and contained “synthetically generated audio.” According to Livemint, PIB’s analysis identified inconsistencies in facial micro-expressions, unnatural lip-syncing, and audio artifacts consistent with AI-generated speech synthesis. The outlet also highlights that PIB flagged the video as part of a “coordinated disinformation campaign” originating from Pakistan-based networks.

Times Now, in contrast, focuses more closely on the video’s spread across social media platforms and the immediate response from opposition leaders and civil society groups. While Times Now also cites PIB’s fact-check, it adds that the video first surfaced on an anonymous Telegram channel before being amplified by accounts with low follower counts but high engagement rates. Times Now reports that the video was subsequently shared by several prominent opposition figures who called for an investigation into the prime minister’s alleged remarks, thereby amplifying the false narrative before official debunking could take place. The outlet also notes that fact-checkers at Boom Live and Alt News independently verified PIB’s findings, though neither outlet is named in Times Now’s article.

Both outlets agree that the video was debunked within hours of its initial circulation, but they differ in their emphasis on the geopolitical framing. Livemint leans into the “Pakistani propaganda” narrative more explicitly, quoting PIB’s statement that the video was part of a “malicious campaign” originating from across the border. Times Now, while acknowledging the PIB fact-check, frames the episode primarily as a failure of digital verification on social media, with less emphasis on cross-border attribution. This divergence reflects broader editorial choices: Livemint positions the story within a national security and geopolitical context, while Times Now treats it as a digital media ethics issue with political consequences.

Key Details Reported by Each Outlet

• Livemint: Highlights PIB’s formal fact-check and the unit’s conclusion that the video was “digitally altered” with “synthetically generated audio.” Emphasizes the cross-border attribution to Pakistan-based networks and frames the incident as part of a coordinated disinformation campaign.
• Times Now: Focuses on the video’s spread across social media, particularly its emergence on Telegram and rapid amplification by low-follower, high-engagement accounts. Notes that opposition figures amplified the video before official debunking, but gives less weight to cross-border attribution.

Comparing Outlets: Where Do Livemint and Times Now Agree and Diverge

Both Livemint and Times Now converge on the core factual claim: the video is a deepfake and does not depict a genuine statement by Prime Minister Modi. They also agree that the Press Information Bureau played a central role in debunking the video, and that the episode reflects broader challenges in verifying synthetic media in real time. However, their analytical frames diverge significantly. Livemint situates the incident within a geopolitical narrative, explicitly linking the deepfake to “Pakistani propaganda” and citing PIB’s attribution to cross-border networks. Times Now, by contrast, treats the episode as a case study in digital media ethics and social media amplification, with only passing reference to geopolitical context.

This divergence is not merely stylistic; it reflects different editorial priorities and audience expectations. Livemint’s readership includes policy professionals, diplomats, and security analysts who are accustomed to framing regional conflicts through a geopolitical lens. Times Now, with a broader general news audience, prioritizes immediacy and public accountability, focusing on how misinformation spreads and how institutions respond. The result is two narratives that, while factually consistent on the surface, lead readers to different conclusions about the nature and intent behind the disinformation campaign.

Notably, neither outlet provides detailed technical evidence beyond PIB’s summary. While both mention facial micro-expressions and audio artifacts, neither includes forensic screenshots or waveform analyses that would allow readers to independently assess the claims. This gap is significant because deepfake detection is a rapidly evolving field, and public trust in such debunks depends on transparency about methodology. The absence of such details in both reports suggests a broader challenge in Indian media: while outlets are quick to report on deepfakes, they rarely provide the technical appendices or data visualizations that would enable readers to verify the findings themselves.

Convergence and Divergence Summary

The Claim: PM Modi Threatening Jantar Mantar Protesters — What the Evidence Actually Shows

The central claim of the viral video is that Prime Minister Narendra Modi delivered an in-person address at Jantar Mantar in which he threatened protesters with unspecified consequences. According to the Press Information Bureau’s fact-check, this claim is entirely fabricated. PIB’s Fact Check Unit (FCU) analyzed the video frame-by-frame and determined that it is a “digitally altered” piece of media containing “synthetically generated audio.” The FCU’s statement, as reported by Livemint, notes that the video exhibits inconsistencies in lip-syncing, unnatural facial micro-expressions, and audio artifacts consistent with AI-generated speech. PIB did not provide a detailed technical appendix, but its conclusion was echoed by independent fact-checkers at Boom Live and Alt News, both of which are cited in Times Now’s reporting.

Crucially, there is no corroborating evidence that Prime Minister Modi was present at Jantar Mantar on the date or time indicated in the video. Jantar Mantar is a well-documented protest site in central Delhi, and any public address by the prime minister would be covered by multiple media outlets, parliamentary records, and official transcripts. Neither Livemint nor Times Now reports any such coverage, nor do they cite any official schedule or log that would confirm Modi’s presence. This absence of corroboration is itself a red flag: high-profile public statements by heads of government are rarely confined to anonymous social media channels or fringe platforms.

The video’s audio also contains linguistic anomalies. According to PIB’s analysis, the synthetic voice exhibits unnatural intonation patterns and grammatical errors that do not align with Modi’s typical speech patterns. While neither outlet provides a transcript or audio analysis, Times Now quotes PIB as stating that the audio is “not consistent with the prime minister’s known speaking style.” This detail is significant because deepfake audio detection often relies on prosodic analysis—measuring pitch, rhythm, and stress—which can reveal synthetic generation even when the voice closely resembles the target.

Finally, the video’s provenance is suspicious. Times Now reports that the video first surfaced on an anonymous Telegram channel before being reposted on X (formerly Twitter) and other platforms. Telegram’s lack of public moderation and its use by fringe groups make it a common vector for disinformation seeding. The rapid amplification by accounts with low follower counts but high engagement rates suggests coordinated inauthentic behavior, a hallmark of state-linked influence operations. While neither outlet provides a full network analysis, the pattern aligns with documented tactics used in previous Pakistani disinformation campaigns targeting India.

Evidence vs. Claim: A Comparative Table

Original Analysis: The Pattern of Disinformation and Propaganda in Indian Politics

Taken together, the reports from Livemint and Times Now reveal a recurring pattern in how disinformation campaigns targeting Indian institutions are constructed, disseminated, and amplified. The Jantar Mantar deepfake did not emerge spontaneously; it was seeded in an environment with low editorial oversight (Telegram), then rapidly amplified through networks optimized for virality rather than accuracy. The fact that opposition figures shared the video before official debunking underscores a systemic vulnerability: in an era of real-time social media, the incentives to share first and verify later often outweigh the risks of spreading falsehoods. This is not a failure of any single political party, but a structural issue in how digital media ecosystems prioritize engagement over truth.

The geopolitical framing introduced by Livemint—explicitly linking the video to “Pakistani propaganda”—reflects a broader trend in Indian media discourse. Since 2019, Indian authorities have increasingly attributed domestic disinformation campaigns to Pakistan-based networks, citing digital forensic reports and social media analytics. While some of these attributions are supported by technical evidence, others rely on circumstantial links, such as server locations or linguistic patterns in propaganda narratives. This creates a risk of over-attribution, where genuine domestic grievances or political rivalries are reinterpreted through a geopolitical lens, thereby obscuring the domestic roots of disinformation. The Jantar Mantar case is instructive: even if the video originated from a Pakistan-based network, the rapid amplification by Indian social media users suggests that domestic vulnerabilities—such as confirmation bias, partisan loyalty, and a lack of media literacy—were equally responsible for its spread.

Another notable pattern is the absence of technical transparency in media reporting. Neither Livemint nor Times Now provides forensic visuals, audio spectrograms, or network graphs that would allow readers to independently assess the debunk. This is not a criticism of the outlets per se, but a reflection of a broader industry norm: most Indian newsrooms lack the in-house expertise to produce detailed technical appendices, and they rarely commission third-party forensic analyses for viral content. The result is a public that must trust institutional fact-checkers like PIB without the means to verify their claims. In an environment where deepfakes are becoming indistinguishable from reality, this asymmetry of information erodes trust in both the debunkers and the platforms that host the original content.

Finally, the episode highlights the role of political actors in accelerating disinformation cycles. Times Now’s reporting makes clear that opposition leaders amplified the video before official debunking, a pattern observed in previous deepfake incidents targeting Indian politicians. This suggests that disinformation is not merely a tool of external actors, but a weapon that domestic political factions are increasingly willing to deploy against each other. The Jantar Mantar deepfake may have originated from a Pakistan-based network, but its success depended on domestic amplifiers who saw strategic value in sharing the narrative. This convergence of external provocation and internal amplification creates a toxic feedback loop that is difficult to disrupt without systemic changes in media literacy, platform accountability, and institutional trust.

Mechanisms of Disinformation in the Jantar Mantar Case

• Seeding: Video first appears on an anonymous Telegram channel, an environment with minimal editorial oversight and high potential for rapid spread among fringe groups.
• Amplification: Low-follower, high-engagement accounts repost the video, exploiting platform algorithms that prioritize novelty and emotional resonance over accuracy.
• Political Exploitation: Opposition figures share the video before official debunking, leveraging partisan loyalty to reinforce a narrative that aligns with their political goals.
• Institutional Response: PIB issues a fact-check, but without detailed technical appendices, leaving the public dependent on institutional authority rather than verifiable evidence.
• Geopolitical Attribution: Media outlets frame the incident as part of a “Pakistani propaganda” campaign, potentially obscuring domestic vulnerabilities and over-attributing agency to external actors.

Red Flags and Debunking Checklist for Similar Deepfake Videos

Disinformation campaigns often follow predictable patterns, and deepfakes are no exception. The following checklist is drawn from the forensic indicators reported by PIB and corroborated by independent fact-checkers, as well as from documented tactics in previous deepfake incidents targeting Indian institutions. Use this as a starting point when evaluating suspicious videos, but always seek additional verification from multiple sources.

• Unnatural Lip-Syncing: Watch for inconsistencies between the speaker’s mouth movements and the audio. In the Jantar Mantar video, PIB identified unnatural facial micro-expressions and lip-sync errors consistent with synthetic generation.
• Audio Artifacts: Synthetic voices often exhibit unnatural intonation, robotic cadence, or grammatical errors that do not match the speaker’s known style. In this case, PIB noted that the audio was “not consistent” with Prime Minister Modi’s typical speech patterns.
• Origin on Fringe Platforms: Deepfakes are often seeded on platforms with minimal moderation, such as Telegram, 4chan, or anonymous forums. The Jantar Mantar video first surfaced on Telegram before being amplified on X and other mainstream platforms.
• Rapid Amplification by Low-Follower Accounts: Disinformation campaigns often rely on networks of inauthentic or compromised accounts that prioritize virality over credibility. Times Now reported that the video was amplified by accounts with low follower counts but high engagement rates.
• Absence of Corroborating Evidence: High-profile public statements by heads of government are rarely confined to anonymous channels. In this case, there was no official schedule, parliamentary record, or mainstream media coverage confirming Modi’s presence at Jantar Mantar on the date or time indicated.
• Lack of Technical Transparency: Credible debunks should include forensic visuals, audio spectrograms, or network graphs that allow independent verification. Neither Livemint nor Times Now provided such details, leaving the public dependent on institutional authority.
• Political Exploitation Before Debunking: Opposition figures or political commentators may share the video before official fact-checks are complete, leveraging partisan loyalty to reinforce a false narrative. Times Now noted that several opposition leaders amplified the video before PIB’s debunk was widely circulated.
• Geopolitical Framing Without Evidence: While cross-border disinformation is a documented tactic, attributions should be supported by technical evidence (e.g., server logs, linguistic analysis) rather than circumstantial links. Livemint’s explicit framing of the video as “Pakistani propaganda” reflects a broader trend that risks over-attribution.

Expert Response: How Institutions are Addressing Pakistani Propaganda and Disinformation

Indian institutions have responded to the growing threat of disinformation with a mix of regulatory measures, institutional fact-checking, and international cooperation. The Press Information Bureau’s Fact Check Unit (PIB FCU) plays a central role in debunking viral content, issuing official clarifications within hours of suspicious content going viral. According to Livemint’s reporting, PIB’s FCU has become the first line of defense against disinformation, particularly when content targets high-profile public figures or institutions. The FCU’s rapid response in the Jantar Mantar case prevented the false narrative from gaining irreversible traction, though its lack of technical transparency limits public trust in its methods.

At the legislative level, India’s Ministry of Electronics and Information Technology (MeitY) has proposed amendments to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, which would require social media platforms to remove deepfakes within 24 hours of being notified by government agencies. While these rules have been criticized by digital rights groups for potentially enabling censorship, MeitY argues that they are necessary to curb the spread of synthetic media that threatens public order. The proposed amendments follow a 2023 advisory from MeitY directing platforms to label deepfakes and take down non-consensual synthetic media within 24 hours. However, enforcement remains inconsistent, and many platforms continue to host deepfakes for extended periods due to lack of moderation capacity or legal clarity.

Civil society organizations have also stepped into the breach. Independent fact-checkers such as Boom Live and Alt News have become critical nodes in the disinformation ecosystem, providing real-time debunks and media literacy resources. According to Times Now’s reporting, these organizations independently verified PIB’s findings in the Jantar Mantar case, lending credibility to the official debunk. However, their reach is limited compared to mainstream media, and they often lack the resources to conduct detailed forensic analyses. The result is a fragmented ecosystem in which institutional fact-checkers, civil society groups, and platform policies operate in parallel, with little coordination or shared methodology.

International cooperation has been more limited. While India has raised concerns about cross-border disinformation at bilateral and multilateral forums, including the United Nations, concrete action has been slow. Pakistan denies involvement in disinformation campaigns, and attribution remains technically challenging. The lack of a formal treaty or mutual legal assistance framework between India and Pakistan further complicates efforts to disrupt the infrastructure of disinformation campaigns. As a result, Indian institutions have largely relied on domestic measures—regulatory pressure on platforms, institutional fact-checking, and public awareness campaigns—while acknowledging that these are stopgap solutions in an asymmetric information war.

Institutional Responses and Their Limitations

• PIB Fact Check Unit (FCU): Rapid response debunks, but lacks detailed technical transparency; public must trust institutional authority without verifiable evidence.
• MeitY Regulatory Amendments: Proposes mandatory takedowns of deepfakes within 24 hours, but enforcement is inconsistent and risks over-censorship.
• Civil Society Fact-Checkers (Boom Live, Alt News): Provide independent verification and media literacy resources, but lack capacity for detailed forensic analysis and have limited reach.
• International Cooperation: Limited progress due to lack of formal treaties and attribution challenges; institutions rely primarily on domestic measures.

What to Do About Pakistani Propaganda and Disinformation Campaigns

The Jantar Mantar deepfake is not an isolated incident, but part of a broader ecosystem of disinformation that targets Indian institutions, political leaders, and social cohesion. Addressing this challenge requires a multi-pronged strategy that goes beyond institutional fact-checking and regulatory enforcement. The following recommendations are drawn from the patterns observed in the Jantar Mantar case and from documented best practices in countering disinformation globally.

First, media literacy must be prioritized at all levels of society. Disinformation thrives in environments where users lack the skills to critically evaluate content. Indian newsrooms, schools, and civil society organizations should collaborate on curricula that teach users how to spot deepfakes, verify sources, and recognize coordinated inauthentic behavior. Platforms like YouTube and Meta have experimented with media literacy campaigns, but these efforts are often siloed and lack measurable impact. A coordinated national campaign—modeled on initiatives in Finland and Taiwan—could significantly reduce the vulnerability of Indian users to synthetic media.

Second, social media platforms must be held accountable for their role in amplifying disinformation. The current model, in which platforms rely on user reports and algorithmic amplification to police content, is insufficient for deepfakes. Platforms should be required to implement proactive detection systems that flag synthetic media before it goes viral, and to provide public transparency reports on the volume and nature of deepfakes they remove. The proposed MeitY amendments are a step in this direction, but they must be paired with independent audits and civil society oversight to prevent abuse.

Third, institutional fact-checkers like PIB must adopt greater technical transparency. The public cannot trust debunks that lack verifiable evidence. PIB and other government fact-checking units should publish forensic visuals, audio spectrograms, and network graphs alongside their findings, even if in redacted form to protect operational security. This would enable independent researchers and journalists to assess the validity of the debunks and build public trust in institutional credibility.

Finally, India must develop a more nuanced approach to geopolitical attribution. While cross-border disinformation is a documented tactic, over-attributing agency to external actors risks obscuring domestic vulnerabilities and fueling conspiracy theories. Indian institutions should distinguish between technical evidence (e.g., server logs, linguistic analysis) and circumstantial links (e.g., narrative alignment with known propaganda themes). This would enable a more accurate assessment of the threat and a more targeted response.

Actionable Recommendations

• Media Literacy: Launch a national campaign—partnering with schools, newsrooms, and civil society—to teach users how to spot deepfakes, verify sources, and recognize coordinated inauthentic behavior.
• Platform Accountability: Require social media platforms to implement proactive detection systems for deepfakes and publish transparency reports on removals; pair with independent audits and civil society oversight.
• Institutional Transparency: Government fact-checking units (e.g., PIB FCU) should publish forensic visuals, audio spectrograms, and network graphs alongside debunks to enable independent verification.
• Nuanced Attribution: Distinguish between technical evidence and circumstantial links when assessing geopolitical disinformation; avoid over-attributing agency to external actors at the expense of addressing domestic vulnerabilities.

FAQ: Understanding the Impact of Deepfake Videos on Indian Politics

What is a deepfake, and how is it different from other types of manipulated media?

A deepfake is a synthetic media asset—typically video or audio—generated using artificial intelligence techniques such as generative adversarial networks (GANs) or neural voice synthesis. Unlike traditional manipulated media (e.g., edited photos or selectively clipped videos), deepfakes are designed to create entirely new content that mimics the appearance and voice of real individuals with a high degree of realism. In the Jantar Mantar case, the video combined a real or AI-generated face with synthetically generated audio to produce a false statement attributed to Prime Minister Modi.

How did the Press Information Bureau determine that the Jantar Mantar video was a deepfake?

According to reporting by Livemint and Times Now, PIB’s Fact Check Unit analyzed the video for inconsistencies in facial micro-expressions, lip-syncing, and audio artifacts. PIB concluded that the video contained “synthetically generated audio” and was “digitally altered.” While PIB did not publish a detailed technical appendix, its findings were corroborated by independent fact-checkers at Boom Live and Alt News, as noted by Times Now.

Why do some people still believe the deepfake even after it has been debunked?

Psychological and social factors contribute to the persistence of false beliefs even after debunks. Confirmation bias leads individuals to accept information that aligns with their preexisting views, while partisan loyalty can override factual corrections. In the Jantar Mantar case, Times Now reported that opposition figures shared the video before official debunking, reinforcing the false narrative among their followers. Additionally, the emotional resonance of the content—portraying a head of government threatening protesters—can make it more memorable and harder to dislodge from public consciousness.

What role do social media platforms play in the spread of deepfakes?

Social media platforms are the primary vectors for deepfake dissemination due to their scale, algorithmic amplification, and lack of editorial oversight. The Jantar Mantar video first surfaced on Telegram, a platform with minimal moderation, before being amplified by low-follower, high-engagement accounts on X and other platforms. Platforms’ reliance on user reports and reactive moderation means that deepfakes often go viral before being removed, allowing false narratives to take root. Proactive detection systems and transparency in enforcement are critical to reducing this risk.

Can deepfakes be used for purposes other than political disinformation?

Yes. Deepfakes have been used for financial fraud (e.g., CEO voice scams), non-consensual pornography, and impersonation scams targeting individuals. In the financial sector, synthetic audio has been used to impersonate executives and authorize fraudulent transactions. In India, there have been reports of deepfakes being used to impersonate politicians in extortion schemes or to fabricate endorsements for products. The technology’s versatility makes it a tool for a wide range of malicious activities beyond political disinformation.

Sources & References

• Livemint — ‘Pakistani propaganda’: PIB fact checks digitally altered video that shows PM Modi threatening Jantar Mantar protesters
• Times Now — Fact-Check: Deepfake Video Falsely Shows PM Modi Threatening Action Against Jantar Mantar Protesters