الصورة الرئيسية:Fotographiya Wedding Photography / Pexels

Deepfake Gandhi Family Videos: Two Booked in India

Two individuals have been charged in India for sharing sexually explicit AI-generated deepfake videos of members of the Gandhi political family, highlighting the escalating use of synthetic media in India’s polarized political climate. The cases underscore gaps in enforcement and the rapid spread of harmful content despite legal provisions targeting deepfakes.

Indian authorities have booked two people for publishing obscene deepfake videos targeting the Gandhi family, a development that spotlights the growing weaponization of AI-generated synthetic media in political disinformation campaigns. This synthesis examines what two independent outlets reported about the case, the nature of the alleged offenses, the legal context, and the broader implications for India’s digital public sphere. The reporting from Hindustan Times and ThePrint provides overlapping but distinct accounts of the incident, the police response, and the political fallout, revealing both corroboration and gaps in public information.

Background: Rise of Deepfake Misinformation in Indian Politics

India has emerged as a global hotspot for AI-driven disinformation, particularly around elections and high-profile political figures. Synthetic media—including deepfake audio and video—has been used to impersonate leaders, fabricate speeches, and spread defamatory content, often targeting opposition figures in a charged electoral environment. The proliferation of low-cost generative AI tools and the ubiquity of social media platforms have lowered the barrier to creating and distributing such content, enabling rapid, large-scale dissemination before platforms can detect or remove it.

While the Gandhi family has been a frequent target of online disinformation campaigns in recent years, the use of deepfake technology represents a qualitative escalation in the sophistication and potential harm of such attacks. Unlike traditional manipulated media, deepfakes can convincingly simulate a person’s voice and appearance, making them more likely to be believed and shared, even when debunked. This trend has prompted calls from civil society, tech platforms, and legal experts for stronger detection mechanisms, faster takedowns, and clearer legal accountability.

What the Two Outlets Reported: A Cross-Comparison of Hindustan Times and ThePrint

Both Hindustan Times and ThePrint reported on August 28, 2026, that two individuals had been booked under sections of the Indian Penal Code (IPC) and the Information Technology Act for posting obscene deepfake videos targeting the Gandhi family. However, the two outlets diverged in their emphasis and level of detail, particularly regarding the identities of the accused, the specific legal sections invoked, and the platforms involved in the spread of the content.

Hindustan Times identified the accused as “two individuals” and stated they were booked under Section 67 of the IT Act (publishing obscene material in electronic form)وSection 509 of the IPC (word, gesture, or act intended to insult a woman’s modesty). The report also mentioned that the videos were shared on WhatsApp and Telegram, platforms frequently used for viral misinformation in India due to end-to-end encryption and closed-group dynamics.

In contrast, ThePrint provided additional context about the political targeting, noting that the videos were part of a broader campaign against Congress party leaders, including Rahul Gandhi, Priyanka Gandhi Vadra, and references to Rajiv and Sonia Gandhi. ThePrint also specified that the accused were booked under Section 67A of the IT Act (publishing sexually explicit material in electronic form), which carries stricter penalties than Section 67. ThePrint did not explicitly name the platforms in its initial report but emphasized the “obscene” nature of the content and its intent to defame.

Both outlets agreed on the core facts: the creation and dissemination of AI-generated deepfake videos targeting the Gandhi family, the obscene content, and the subsequent police action. However, ThePrint’s account placed greater emphasis on the political targeting and the specific legal section addressing sexually explicit material, while Hindustan Times focused on the platforms used for distribution and the broader legal framework under the IPC and IT Act.

Discrepancies and Gaps

One notable discrepancy lies in the legal sections cited. Hindustan Times referenced Section 67 and Section 509, which address obscenity and insult to modesty, respectively. ThePrint, however, cited Section 67A, which specifically targets sexually explicit content. This difference may reflect evolving interpretations of the law or variations in how police officials register cases. It also highlights the ambiguity in applying existing statutes to AI-generated synthetic media, where the intent to defame or humiliate may overlap with charges of obscenity.

Neither outlet provided the names of the accused, citing police anonymity or ongoing investigations. Both also lacked detail on whether the videos were created using publicly available tools or custom AI models, or whether any arrests had been made beyond the initial booking. These gaps underscore the opacity that often surrounds early-stage digital investigations in India, where police may file cases without immediately disclosing full details to prevent interference or further spread of the content.

The Alleged Offense: Nature of the Obscene Deepfake Videos

According to Hindustan Times, the deepfake videos depicted members of the Gandhi family in sexually explicit scenarios, using AI to superimpose their faces onto fabricated bodies or actions. Such content is designed to humiliate, intimidate, and discredit the individuals portrayed, leveraging societal taboos to amplify outrage and viral spread. The use of sexualized deepfakes against women politicians is a particularly insidious tactic, as it combines gendered harassment with technological manipulation to inflict reputational and psychological harm.

ThePrint’s reporting reinforced the obscene and defamatory nature of the videos, emphasizing their intent to target Congress leaders within a broader political context. While ThePrint did not provide visual or technical details about the deepfakes, the reference to “Congress leaders” suggests the content was part of a coordinated campaign rather than isolated incidents. This framing aligns with a pattern observed in previous deepfake campaigns in India, where synthetic media is used to smear opposition figures ahead of elections or during periods of heightened political tension.

Both reports avoided publishing or linking to the videos, adhering to ethical guidelines that prohibit amplifying harmful or non-consensual content. However, they noted that the videos had circulated widely on encrypted messaging apps before being flagged by users and reported to authorities. The rapid spread of such content—often within hours of creation—demonstrates the challenge of detection and moderation in closed, encrypted ecosystems where content is not publicly indexed.

Technical and Ethical Dimensions

The creation of such deepfakes typically involves training AI models on publicly available images and videos of the target individuals, followed by fine-tuning to produce realistic outputs. While tools like DeepFaceLab, FaceSwap, and voice cloning models have become more accessible, high-quality deepfakes still require technical skill and computational resources. The fact that these videos were produced and distributed suggests either the involvement of technically proficient actors or the use of user-friendly platforms that automate the process.

Ethically, the use of deepfakes to fabricate sexually explicit content is a form of gendered disinformation, disproportionately targeting women in politics. Studies on online harassment of female politicians in India have shown that sexualized abuse is a common tactic to silence women and discourage their participation in public life. The deployment of deepfakes amplifies this harm by making the abuse appear “real,” thereby eroding trust in the targets and the platforms that host the content.

Legal Framework: How Indian Law Addresses Deepfake Defamation

India’s legal response to deepfakes is fragmented, relying on a patchwork of provisions under the IPC, the IT Act, and rules issued by the Ministry of Electronics and Information Technology (MeitY). The most commonly invoked sections in cases involving synthetic media include:

• Section 67 and 67A of the IT Act: These sections criminalize the publication, transmission, or creation of obscene or sexually explicit material in electronic form. Section 67A, in particular, was introduced to address the distribution of sexually explicit content and carries a maximum penalty of five years’ imprisonment and a fine.
• Section 509 of the IPC: This section penalizes any word, gesture, or act intended to insult the modesty of a woman, a charge frequently used in cases of online harassment and defamation targeting female politicians.
• Section 499 and 500 of the IPC: These sections deal with defamation, though their application to AI-generated content remains legally untested in higher courts. Defamation cases involving deepfakes often hinge on proving intent and harm, which can be difficult in the absence of clear precedents.
• Section 66D of the IT Act: This section addresses cheating by personation using communication devices, which could apply if a deepfake is used to impersonate a person for fraudulent purposes.
• Intermediary Guidelines (2021): Under these rules, social media platforms are required to remove content reported as fake or defamatory within 36 hours of a government or court order. However, enforcement remains inconsistent, and platforms often cite privacy or encryption concerns when resisting takedown requests.

While these provisions provide a legal basis for prosecuting creators and distributors of deepfakes, their effectiveness is limited by several factors. First, the rapid pace of technological change outstrips legislative updates, leaving gaps in how laws are interpreted. Second, law enforcement agencies often lack the technical expertise to investigate AI-generated content, leading to delays or misclassification of cases. Third, the anonymity afforded by digital platforms and encrypted messaging services complicates identification and prosecution of offenders.

In this case, the police invoked sections addressing obscenity and insult to modesty, reflecting a preference for charges that are easier to prove in the short term rather than pursuing defamation or impersonation claims, which may require more complex evidence. This approach aligns with a broader trend in India, where law enforcement tends to prioritize charges that align with immediate public outrage or political pressure, even if they do not fully capture the harm caused by synthetic media.

Jurisdictional and Enforcement Challenges

Another challenge lies in jurisdiction. Deepfake content often crosses state and even national borders, requiring coordination between multiple law enforcement agencies and platforms. In India, the lack of a centralized cybercrime unit with nationwide jurisdiction can delay investigations and allow perpetrators to evade accountability. Additionally, platforms like WhatsApp and Telegram operate under different legal regimes than public social media sites like Facebook or X (formerly Twitter), making coordinated responses difficult.

Despite these challenges, the booking of individuals in this case signals a willingness by authorities to apply existing laws to AI-generated content. However, without clearer guidelines, faster investigative processes, and stronger platform accountability, such cases may remain isolated incidents rather than a deterrent to future deepfake campaigns.

How the Content Spread: Platforms, Virality, and Algorithmic Amplification

Both Hindustan Times and ThePrint noted that the deepfake videos were shared on encrypted messaging platforms, particularly WhatsApp and Telegram. These platforms are widely used in India for political communication due to their privacy features and low data usage, but they also serve as fertile ground for the spread of misinformation and synthetic media. The closed nature of these groups—often limited to trusted contacts or political affiliates—reduces the likelihood of early detection by automated content moderation systems, which primarily scan public-facing content.

ThePrint’s emphasis on the political targeting of Congress leaders suggests that the videos were likely disseminated within closed networks aligned with opposing political factions. Such networks are designed to reinforce existing beliefs and mobilize supporters, making them ideal vectors for disinformation campaigns. The use of deepfakes in these contexts is particularly effective because the content appears to come from a trusted source within the group, increasing the likelihood of sharing and engagement.

Hindustan Times highlighted the role of WhatsApp and Telegram in the spread, noting that the videos circulated widely before being flagged. This aligns with research on misinformation in India, which shows that encrypted messaging apps are often the first stage of viral spread, followed by amplification on public platforms once the content gains traction. The delay between creation and detection—often measured in hours or days—gives perpetrators a window to maximize harm before countermeasures can be deployed.

Platform Responsibility and Detection Gaps

Neither outlet reported on whether the platforms themselves were notified or took action to remove the content. Under India’s Intermediary Guidelines, platforms are required to remove content flagged as fake or defamatory within 36 hours of a government or court order. However, the guidelines do not mandate proactive detection or removal of synthetic media, leaving a significant gap in accountability.

Platform Type	Detection Capability	آلية التنفيذ	Gap Identified
Public Social Media (e.g., Facebook, X)	Moderation tools scan public posts for policy violations	Can remove content or accounts under IT Act and community standards	Limited effectiveness in detecting deepfakes without metadata or watermarks
Encrypted Messaging (e.g., WhatsApp, Telegram)	No scanning of message content due to end-to-end encryption	Relies on user reports and platform takedowns after content spreads publicly	No proactive detection; content spreads rapidly in closed networks
Cloud Storage and File-Sharing Services	Limited scanning for synthetic media	Removal only after explicit takedown requests	Deepfakes often hosted on third-party sites before being shared

The table above illustrates the structural limitations of current platform-based detection and enforcement. While public social media platforms have some capacity to identify and remove deepfakes, their tools are not foolproof and often rely on user reports. Encrypted messaging platforms, by design, cannot scan message content, leaving them largely reactive. This asymmetry creates opportunities for bad actors to exploit closed networks for initial dissemination, only for the content to resurface on public platforms once it has gained viral momentum.

العلامات الحمراء وقائمة التحقق من دحضها: كشفDeepfakes السياسية التي تم إنشاؤها بواسطة الذكاء الاصطناعي

Detecting deepfakes requires a combination of technical awareness and critical media literacy. While AI-generated content is becoming increasingly sophisticated, certain inconsistencies and anomalies can reveal synthetic origins. Below is a checklist of red flags and verification steps to help identify potential deepfakes:

• الحركات الوجهية غير الطبيعية: Look for inconsistencies in blinking, eye movement, or facial expressions. Deepfakes often struggle to replicate natural micro-expressions or blinking patterns.
• المعايرة المرئية الصوتية: Check if the lip movements align with the spoken words. AI-generated audio may not perfectly sync with the video, especially in languages with complex phonetics.
• إضاءات وظلال Inconsistent lighting or shadows on the face or background can indicate a deepfake, as AI may not accurately render real-world lighting conditions.
• Unusual Backgrounds or Artifacts: Look for pixelation, blurring, or distortions around the edges of the face or in the background, which can reveal manipulation.
• تحليل البيانات الوصفية Use tools like InVID or Google Reverse Image Search to check the origin of the video. Metadata such as creation date, device type, or editing software can provide clues.
• مصدر التحقق: Check if the video is being shared from a verified account or a known disinformation source. Cross-reference with reputable news outlets or fact-checking organizations.
• Behavioral Anomalies: Does the person in the video appear to act out of character? Politicians, for example, rarely make sexually suggestive remarks in public speeches.
• Reverse Image Search for Faces: Use tools like Yandex or PimEyes to search for similar faces online. If the same face appears in unrelated contexts, it may be a deepfake.
• الاختلافات السياقية: Does the video align with the person’s known schedule, public statements, or recent events? If not, it may be fabricated.
• سلوك المنصة: Be wary of videos shared in closed groups or forwarded multiple times without context. Encrypted platforms are common vectors for deepfake dissemination.

If you encounter a potential deepfake, avoid sharing it and report it to the platform and fact-checking organizations such as Boom Live, Alt News, or Vishvas News. Document the content with timestamps and screenshots before it is removed, as this can aid in investigations. Users should also familiarize themselves with the reporting mechanisms of platforms like WhatsApp, Telegram, and Facebook, which often include options to flag misinformation or synthetic media.

Institutional Response: Police Action and Political Reactions

Both Hindustan Times and ThePrint reported that police in India had booked two individuals in connection with the deepfake videos. However, neither outlet provided details on the location of the incident, the investigating agency, or whether additional arrests were expected. The lack of transparency around the investigation reflects broader challenges in India’s digital law enforcement, where cases are often filed under anonymity to prevent interference or retaliation.

ThePrint’s emphasis on the political targeting of Congress leaders suggests that the case may have broader implications for the party and its leadership. While neither outlet quoted political figures directly, the framing implies that the videos were part of a coordinated campaign rather than isolated incidents. This aligns with a pattern observed in previous elections, where synthetic media has been used to smear opposition candidates and influence public opinion.

Police action in such cases is often reactive, triggered by public outrage or media reports rather than proactive detection. This reactive approach can limit the effectiveness of investigations, as perpetrators may have already dispersed or deleted evidence. Additionally, the use of sections addressing obscenity and insult to modesty—rather than defamation or impersonation—may reflect a strategic choice by law enforcement to secure quicker convictions, even if the charges do not fully capture the harm caused by the deepfakes.

Silence from Political Parties

Neither report included statements from Congress party leaders or representatives regarding the incident. This silence may stem from legal advice to avoid amplifying the content or from internal party strategies to downplay the impact of the deepfakes. However, the absence of public condemnation or calls for stronger action also highlights the normalization of digital disinformation as a political tool, where parties may prioritize electoral strategy over addressing the underlying issue.

In contrast, previous deepfake incidents in India have sometimes elicited stronger institutional responses, including public denials, legal threats, and demands for platform accountability. The relatively muted reaction in this case may indicate a shift in how such incidents are perceived—as routine rather than exceptional—or it may reflect a strategic decision to avoid drawing further attention to the content.

Original Analysis: A Pattern of AI-Driven Defamation in India’s Electoral Landscape

Taken together, the reports from Hindustan Times and ThePrint reveal a troubling pattern: the weaponization of AI-generated deepfakes as a tool of political defamation in India, particularly against high-profile opposition figures. The use of sexually explicit content is not merely an attempt to mislead; it is a deliberate tactic to humiliate, intimidate, and silence women in politics, leveraging societal taboos to amplify harm. The fact that two individuals were booked under laws addressing obscenity and insult to modesty suggests that authorities recognize the severity of the offense, but the legal framework remains ill-equipped to address the full scope of the harm.

The reliance on encrypted messaging platforms for dissemination underscores a critical gap in content moderation. While public social media platforms have some capacity to detect and remove deepfakes, encrypted apps operate outside the reach of automated scanning, allowing misinformation to spread unchecked in closed networks. This asymmetry creates a fertile environment for bad actors to exploit, particularly during election seasons or periods of heightened political tension.

Moreover, the legal response in this case—focusing on obscenity and insult to modesty—reflects a broader trend in India’s approach to digital defamation. Rather than pursuing charges that address the unique harms of synthetic media (such as impersonation or coordinated disinformation campaigns), law enforcement often defaults to sections that are easier to prove but may not fully capture the intent or impact of the offense. This approach risks underestimating the long-term damage to democratic discourse, where trust in institutions and individuals is eroded by the proliferation of convincing fakes.

Finally, the lack of transparency around the investigation and the muted political response suggest that deepfake defamation is becoming normalized as a political tactic. If parties and institutions treat such incidents as routine rather than exceptional, the deterrent effect of legal action will diminish, emboldening future perpetrators. Without stronger platform accountability, clearer legal guidelines, and proactive detection mechanisms, AI-driven disinformation is poised to become an even more pervasive threat to India’s electoral integrity.

What to Do: Reporting, Prevention, and Platform Accountability

For individuals who encounter potential deepfakes, the first step is to avoid sharing or engaging with the content. Sharing such material, even with debunking intent, can amplify its reach and lend it undeserved legitimacy. Instead, document the content with timestamps and screenshots, then report it to the platform and a fact-checking organization. In India, organizations like Boom Live, Alt News, and Vishvas News regularly debunk deepfakes and provide guidance on verification tools.

For platforms, the challenge lies in balancing privacy with accountability. Encrypted messaging services like WhatsApp and Telegram cannot scan message content due to end-to-end encryption, but they can implement proactive measures such as user education, in-app warnings, and partnerships with fact-checkers to identify and mitigate the spread of deepfakes. Public platforms like Facebook and X can improve their detection algorithms by incorporating metadata analysis, reverse image search tools, and collaboration with AI researchers to identify synthetic media.

For policymakers, the priority should be updating the legal framework to explicitly address AI-generated synthetic media. This includes clarifying liability for platforms, establishing faster takedown mechanisms, and ensuring that law enforcement agencies have the technical training and resources to investigate such cases. Additionally, public awareness campaigns—targeting both urban and rural populations—can help build media literacy and reduce the likelihood of deepfakes being believed and shared.

Civil society organizations and media literacy initiatives play a crucial role in this ecosystem. By training journalists, educators, and community leaders to identify and debunk deepfakes, these groups can help inoculate the public against disinformation. In India, where digital literacy varies widely, such initiatives are essential to ensuring that citizens can critically evaluate the content they encounter online.

Steps for Institutions and Journalists

• Adopt Verification Protocols: Newsrooms should implement standardized verification processes for videos and images, including reverse image search, metadata analysis, and source triangulation.
• Collaborate with Fact-Checkers: Partner with organizations like Boom Live or Alt News to debunk deepfakes quickly and reduce their viral spread.
• Educate Audiences: Use social media and community outreach to teach users how to spot deepfakes and report them effectively.
• Advocate for Platform Transparency: Push for greater transparency from platforms about the prevalence of deepfakes on their services and the measures they are taking to address them.

الأسئلة الشائعة

What legal sections were invoked in the Gandhi family deepfake case?

According to Hindustan Times, the accused were booked under Section 67 of the IT Act and Section 509 of the IPC. ThePrint reported that they were charged under Section 67A of the IT Act, which specifically addresses sexually explicit material in electronic form. The discrepancy reflects variations in how police classify such cases.

Which platforms were used to spread the deepfake videos?

Hindustan Times identified WhatsApp and Telegram as the platforms where the videos were shared. ThePrint did not explicitly name the platforms but emphasized the closed, encrypted nature of the networks involved.

كيف يمكنني التحقق مما إذا كان الفيديو مزيفًا عميقًا؟

Check for unnatural facial movements, audio-visual mismatches, inconsistent lighting, and artifacts around the edges of the face. Use tools like InVID, Google Reverse Image Search, or Yandex to analyze the video’s origin and metadata. Reverse image search for faces can also reveal inconsistencies.

ماذا أفعل إذا واجهت تزويرًا عميقًا؟

Avoid sharing or engaging with the content. Document it with timestamps and screenshots, then report it to the platform and a fact-checking organization such as Boom Live, Alt News, or Vishvas News. If the content is sexually explicit or defamatory, consider filing a police complaint under relevant sections of the IT Act or IPC.

Why are deepfakes particularly harmful in politics?

Deepfakes can convincingly simulate real people, making them more believable than traditional manipulated media. In politics, they are often used to humiliate opponents, spread disinformation, and erode trust in institutions. The use of sexually explicit deepfakes against women politicians is a particularly insidious tactic that combines gendered harassment with technological manipulation.

المصادر والمراجع

• Hindustan Times — Two people booked for posting obscene deepfake videos targeting Gandhi family
• ThePrint — Two people booked for posting obscene deepfake videos targeting Cong leaders