#FactCheck-AI-Manipulated Video Falsely Claims PM Modi Promised Free Gold for Women

Introduction

For years, when Meta's automated systems flagged suspected child sexual abuse material on Facebook or Instagram involving an Indian child, that information did not go straight to an Indian police station. It went first to a private, US based nonprofit, the National Center for Missing & Exploited Children (NCMEC), which would then decide how and when to route the relevant details back to law enforcement in the country where the crime actually happened. That indirect chain has now been shortened. In mid September 2026, senior Indian government officials confirmed that Meta had agreed, for the first time among major social media platforms operating in India, to report child sexual abuse cases and details of repeat offenders directly to the cybercrime division of the Indian Cyber Crime Coordination Centre, commonly known as I4C, under the Ministry of Home Affairs. Google followed within days with a similar commitment of its own.

How the reporting actually worked before this

Under US federal law, Meta and every other American headquartered platform is legally obligated to report CSAM instances to NCMEC, facing criminal liability, financial penalties, and the loss of legal safe harbour protections if they fail to do so. NCMEC would then coordinate with law enforcement agencies in the relevant country. In India's case, that meant a 2019 memorandum of understanding between NCMEC and the National Crime Records Bureau, through which more than 69 lakh CyberTipline reports had reportedly been shared with Indian states and union territories by early 2024. The structural problem was not the volume of reports; it was the layer of institutional distance built into every single one of them. As one government official put it plainly, an Indian law enforcement agency was, in effect, made to wait on a private American nonprofit for information about a crime committed on Indian soil against an Indian child, and NCMEC itself has little to no operational presence in India to manage that handoff quickly.

That distance had real, documented costs. Officials directly linked slow data sharing to delayed CSAM investigations, describing situations where exploitation was actively ongoing while Indian agencies remained one step removed from the account level details, IP addresses, and identifying information that could have accelerated a rescue or an arrest.

What actually changed, and why now

This shift did not emerge from a routine policy review. It followed sustained government pressure through 2026, including a formal inquiry launched by the National Commission for Protection of Child Rights into Meta India's child safety policies, multiple summons issued to Meta India head, and a July 2026 controversy in which Meta was found to have run paid Instagram advertisements allegedly directing users toward Telegram channels distributing CSAM. 

Union Minister Ashwini Vaishnaw directly raised the direct reporting demand with Meta's Chief of Global Affairs, Joel Kaplan, during a visit to New Delhi in August 2026, and separately questioned whether Meta could still claim to be a "simple intermediary" given that it profits from advertisement placement and exercises systems level control over ad distribution. Under the new arrangement, account level details, including names, associated email addresses, and IP addresses, will now flow directly to I4C, which officials say should meaningfully compress the time between detection and actionable law enforcement response.

Why this is a welcome step, but not a solved problem

There is genuine reason to treat this as meaningful progress. For years, reporting routed through NCMEC meant Indian agencies were effectively once removed from data concerning crimes committed on Indian soil, against Indian children. Direct reporting to I4C shortens that distance and gives investigators quicker access to information that matters most in time sensitive cases, where every day of delay can mean continued access, continued distribution, or a victim who remains unidentified.

That said, this should not be described as a finished problem, and its real value will depend on at least three things.

The first is the depth of data actually shared. Account level details, a name, an email address, an IP address, are a starting point for an investigation, not an end point. Whether the new arrangement includes the kind of contextual detail investigators actually need to build a prosecutable case, rather than simply a faster version of the same limited data packet previously routed through NCMEC, remains to be seen in practice rather than in the announcement itself.

The second is institutional capacity on the receiving end. I4C and state cyber cells will now be positioned to receive a higher volume of reports arriving faster than before, but a faster pipeline only produces better outcomes if the infrastructure receiving it can actually convert that inflow into timely, actionable results on the ground. India's cyber cells have historically been understaffed and unevenly resourced across states, with some districts operating with only a handful of trained personnel handling cybercrime alongside a broad range of other digital offences. A genuinely improved reporting channel could, in that context, simply shift the bottleneck downstream from Washington to a district cyber cell rather than eliminate it, unless the surge in report volume is matched by a corresponding investment in trained staff, forensic capacity, and case management systems capable of prioritising the most time sensitive reports first.

The third is consistency across the industry. Meta and Google have now committed to this model, but whether it extends reliably across every major platform operating in India, rather than remaining an arrangement secured through targeted pressure on one or two companies at a time, will determine whether this becomes a systemic fix or a set of isolated exceptions.

The quieter tension nobody has fully resolved

There is also a more technical reality worth naming honestly. As platforms adopt end to end encryption more broadly, extending it across messaging services and, in some cases, considering it for other content types, the pool of CSAM content that can actually be automatically detected narrows, even as the reporting pipeline for whatever is detected improves. WhatsApp itself has argued in Indian courts, including in its ongoing challenge to the IT Rules' traceability requirements, that end to end encryption and message traceability are close to mutually exclusive, a position echoed by cryptography experts including Signal's Moxie Marlinspike, who has stated plainly that identifying a message's originator at the very least undermines, and likely breaks, the purpose of end to end encryption altogether. Faster reporting channels do not, by themselves, compensate for reduced visibility into content that a platform's own encryption prevents it from scanning in the first place. That tension, between strengthening user privacy through encryption and strengthening child safety through detection, sits underneath this entire announcement and deserves sustained attention from technologists, platforms, and child safety practitioners alike, rather than being treated as resolved simply because reporting speed has improved.

India's National Human Rights Commission has already flagged this directly in its own advisory, recommending that platforms using end to end encryption be required to devise additional protocols or technology specifically to monitor CSAM circulation, an unresolved and technically difficult ask that sits alongside, rather than instead of, the reporting improvements announced this month.

What this actually represents

The most accurate way to characterise this development is as a meaningful first step in platform accountability toward children in India, not a finish line. Meta agreeing to report directly, rather than through a private American intermediary, closes a real structural gap that officials say genuinely delayed investigations. But the task now shifts to India itself: building the institutional capacity at I4C and in state cyber cells to make full use of what is finally being offered, extending the same commitment across the rest of the industry rather than platform by platform, and confronting the encryption question honestly rather than treating faster reporting as a substitute for solving it.

A judicial note worth remembering

It is worth noting that Indian courts pushed for this same sensitivity well before it became routine practice. The Supreme Court, in Just Rights for Children Alliance v. S. Harish, 2024, directed that the term "child pornography" be replaced with "child sexual abuse and exploitation material" across legal and public usage, holding that the word "pornography" wrongly implies consent and normalises what is, in fact, a recorded act of abuse. That judicial insistence on precise language is a useful reminder that reporting frameworks like this one will ultimately be judged not just on speed, but on whether they treat the material, and the children in it, with the gravity courts have already demanded.

References

1. https://www.business-standard.com/industry/news/meta-agrees-to-share-csam-details-with-indian-law-enforcement-agencies-125091500052_1.html
2. https://www.business-standard.com/technology/tech-news/google-india-to-share-csam-case-details-with-govt-cybercrime-division-126092101062_1.html
3. https://www.storyboard18.com/advertising/ncpcr-summons-meta-india-md-again-over-child-abuse-ads-ws-l-110781.htm
4. https://www.superprep.io/current-affairs/meta-report-csam-directly-to-indias-cybercrime-portal
5. https://theprint.in/india/specialised-central-police-unit-use-of-technology-to-proactively-detect-csam-nhrc-advisory/1822223/
6. https://www.livelaw.in/news-updates/whatsapp-delhi-high-court-traceability-end-to-end-encryption-privacy-risk-174743
7. https://www.forbesindia.com/article/take-one-big-story-of-the-day/traceability-and-endtoend-encryption-cannot-coexist-on-digital-messaging-platforms-experts/66969/1

‍

Executive Summary

‍
A Pakistani propaganda account on social media has claimed that Russian President Vladimir Putin ignored Indian Prime Minister Narendra Modi and stood alongside the Pakistani Prime Minister. Research by CyberPeace’s research wing found that PM Modi and President Putin held a bilateral meeting on the sidelines of the SCO Summit in Bishkek, Kyrgyzstan. During the meeting, the two leaders exchanged warm greetings and discussed several issues, including the India-Russia strategic partnership.

‍

Claim: 

‍

A social media post claims that Russian President Vladimir Putin ignored Prime Minister Narendra Modi and stood alongside the Pakistani Prime Minister.

https://x.com/TacticalTribun/status/2094808040973172786 

‍

Fact Check:

‍

To verify the truth behind the viral claim, we searched Google using relevant keywords related to Prime Minister Narendra Modi’s official visit to Kyrgyzstan. During the search, we found a video posted on ANI’s X handle on September 1, 2026. According to the video, Prime Minister Narendra Modi met Chinese President Xi Jinping and Russian President Vladimir Putin in Bishkek during the 26th Shanghai Cooperation Organisation (SCO) Summit.

https://x.com/ANI/status/2094675518897692896 

‍

In the next step of our research, we scanned Prime Minister Narendra Modi’s X account, where we found a post dated August 31, 2026. In the post, PM Modi wrote, “Had extensive discussions with President Putin. Discussed the full range of the India-Russia Special and Privileged Strategic Partnership. It is gladdening to see significant progress across various sectors in recent years. Focus areas for the India-Russia friendship remain energy, innovation, space, fertilisers and people-to-people linkages.”

https://x.com/narendramodi/status/2094433881714032725?s=20 

‍

‍

At the end of our research, we found a report published on the Aaj Tak website on September 1, 2026. According to the report, Prime Minister Narendra Modi was in Bishkek, the capital of Kyrgyzstan, where he attended the Shanghai Cooperation Organisation (SCO) Summit. Several leaders of the organisation, including Chinese President Xi Jinping, Russian President Vladimir Putin and Pakistani Prime Minister Shehbaz Sharif, also attended the summit. During the summit, a moment was captured in which PM Modi warmly greeted Chinese President Xi Jinping and Russian President Vladimir Putin together, while Shehbaz Sharif was seen quietly walking past them.

https://www.aajtak.in/world/story/pm-modi-xi-jinping-vladimir-putin-handshake-shahbaz-shareef-sco-summit-ntc-mnrd-smsr-dskc-2633823-2026-09-01

‍

Conclusion:

‍

Our research established that the claim that Russian President Vladimir Putin ignored Prime Minister Narendra Modi and gave preference to the Pakistani Prime Minister is misleading.  The viral post presents video footage from different occasions out of context. In reality, Prime Minister Narendra Modi and President Putin held a bilateral meeting, during which the two leaders discussed several key issues, including the India-Russia strategic partnership.

‍

Introduction

In the digital landscape, there is a rapid advancement of technologies such as generative AI(Artificial Intelligence), deepfake, machine learning, etc. Such technologies offer convenience to users in performing several tasks and are capable of assisting individuals and business entities. Certain regulatory mechanisms are also established for the ethical and reasonable use of such advanced technologies. However, these technologies are easily accessible; hence, cyber-criminals leverage AI tools and technologies for malicious activities or for committing various cyber frauds. By such misuse of advanced technologies, new cyber threats have emerged. 

Deepfake Scams 

Deepfake is an AI-based technology. Deepfake is capable of creating realistic images or videos which in actuality are created by machine algorithms. Deepfake technology, since easily accessible, is misused by fraudsters to commit various cyber crimes or deceive and scam people through fake images or videos that look realistic. By using the Deepfake technology, cybercriminals manipulate audio and video content, which looks very realistic but, in actuality, is fake. 

Voice cloning 

To create a voice clone of anyone's, audio can be deepfaked too, which closely resembles a real one but, in actuality, is a fake voice created through deepfake technology. Recently, in Kerala, a man fell victim to an AI-based video call on WhatsApp. He received a video call from a person claiming to be his former colleague. The scammer, using AI deepfake technology, impersonated the face of his former colleague and asked for financial help of 40,000. 

Uttarakhand Police issues warning admitting the rising trend of AI-based scams

Recently, Uttarakhand police’s Special Task Force (STF) has issued a warning admitting the widespread of AI technology-based scams such as deepfake or voice cloning scams targeting innocent people. Police expressed concern that several incidents have been reported where innocent people are lured by cybercriminals. Cybercriminals exploit advanced technologies and manipulate innocent people to believe that they are talking to their close ones or friends, but in actuality, they are fake voice clones or deepfake video calls. In this way, cybercriminals ask for immediate financial help, which ultimately leads to financial losses for victims of such scams. 

Tamil Nadu Police Issues advisory on deepfake scams

To deceive people, cyber criminals misuse deepfake technologies and target them for financial gain. Recently, Tamilnadu Police Cyberwing have issued an advisory on rising deepfake scams. Fraudsters are creating highly convincing images, videos or voice clones to defraud innocent people and make them victims of financial fraud. The advisory states that you limit the personal data you share you share online and adjust privacy settings. Advisory says to promptly report any suspicious activity or cyber crimes to 1930 or the National Cyber Crime Reporting portal.

Best practices 

1. Pay attention if you notice compromised video quality because deepfake videos often have compromised or poor video quality and unusual blur resolution, which poses a question to its genuineness. Deepfake videos often loop or unusually freeze, which indicates that the video content might be fabricated.

1. Whenever you receive requests for any immediate financial help, act responsively and verify the situation by directly contacting the person on his primary contact number.
2. You need to be vigilant and cautious, since scammers often possess a sense of urgency, leading to giving no time for the victim to think about it and deceiving them by making a quick decision. Scammers pose sudden emergencies and demand financial support on an urgent basis. 
3. Be aware of the recent scams and follow the best practices to stay protected from rising cyber frauds. 
4. Verify the identity of unknown callers.
5. Utilise privacy settings on your social media.
6. Pay attention if you notice any suspicious nature, and avoid sharing voice notes with unknown users because scammers might use them as voice samples and create your voice clone. 
7. If you fall victim to such frauds, one powerful resource available is the National Cyber Crime Reporting Portal (www.cybercrime.gov.in) and the 1930 toll-free helpline number where you can report cyber fraud, including any financial crimes.

‍

Conclusion

AI-powered technologies are leveraged by cybercriminals to commit cyber crimes such as deepfake scams, voice clone scams, etc. Where innocent people are lured by scammers. Hence there is a need for awareness and caution among the people.  We should be vigilant and aware of the growing incidents of AI-based cyber scams. Must follow the best practices to stay protected.

References: 

• https://www.the420.in/ai-voice-cloning-cyber-crime-alert-uttarakhand-police/
• https://www.trendmicro.com/vinfo/us/security/news/cybercrime-and-digital-threats/exploiting-ai-how-cybercriminals-misuse-abuse-ai-and-ml#:~:text=AI%20and%20ML%20Misuses%20and%20Abuses%20in%20the%20Future&text=Through%20the%20use%20of%20AI,and%20business%20processes%20are%20compromised.
• https://www.ndtv.com/india-news/kerala-man-loses-rs-40-000-to-ai-based-deepfake-scam-heres-what-it-is-4217841
• https://news.bharattimes.co.in/t-n-cybercrime-police-issue-advisory-on-deepfake-scams/

‍