#FactCheck - Viral Video Claiming to Show Kashmir Avalanche Is AI-Generated
Executive Summary
A video is being shared on social media claiming to show an avalanche in Kashmir. The caption of the post alleges that the incident occurred on February 6. Several users sharing the video are also urging people to avoid unnecessary travel to hilly regions. CyberPeace’s research found that the video being shared as footage of a Kashmir avalanche is not real. The research revealed that the viral video is AI-generated.
Claim
The video is circulating widely on social media platforms, particularly Instagram, with users claiming it shows an avalanche in Kashmir on February 6. The archived version of the post can be accessed here. Similar posts were also found online. (Links and archived links provided)
Fact Check:
To verify the claim, we searched relevant keywords on Google. During this process, we found a video posted on the official Instagram account of the BBC. The BBC post reported that an avalanche occurred near a resort in Sonamarg, Kashmir, on January 27. However, the BBC post does not contain the viral video that is being shared on social media, indicating that the circulating clip is unrelated to the real incident.
A close examination of the viral video revealed several inconsistencies. For instance, during the alleged avalanche, people present at the site are not seen panicking, running for cover, or moving toward safer locations. Additionally, the movement and flow of the falling snow appear unnatural. Such visual anomalies are commonly observed in videos generated using artificial intelligence. As part of the research , the video was analyzed using the AI detection tool Hive Moderation. The tool indicated a 99.9% probability that the video was AI-generated.
Conclusion
Based on the evidence gathered during our research , it is clear that the video being shared as footage of a Kashmir avalanche is not genuine. The clip is AI-generated and misleading. The viral claim is therefore false.
Related Blogs
Executive Summary:
A video of India’s Defence Minister Rajnath Singh is going viral on social media. The post claims that Rajnath Singh is openly supporting Israeli-American attacks against Iran. In the video, he can allegedly be heard saying that Prime Minister Narendra Modi had visited Israel before the war began and warned Tehran that disturbing peace would have serious consequences.
Research by CyberPeace found that the viral video is a deepfake created using Artificial Intelligence (AI). Rajnath Singh has not made any such statement about Iran or the Israel-US conflict.
Claim
A Facebook user “Sheikh Sadeque Ali” shared the video on March 2, 2026. The caption of the post reads, “Indian Defence Minister Rajnath Singh is supporting Israel’s attack on Iran. This clearly shows that India supports the killing of Muslims.”
In the viral video, Rajnath Singh appears to say in English: “Prime Minister Modi’s visit to Israel before the attack on Iran reflects India’s solidarity with its strategic partner… He warned Tehran that hostile actions would have serious consequences for regional peace.”
Fact Check:
To verify the claim, we extracted keyframes from the viral video and conducted a reverse image search. During the research , we found the original video on Rajnath Singh’s official YouTube channel. The video was uploaded on November 23, 2025.In the original video, Rajnath Singh was addressing a Sindhi community conference in Delhi. During his speech, he was talking about Sindhi culture and the history of Partition. He did not mention Israel, Iran or any Middle East conflict during the entire program.
Upon closely examining the viral video, technical inconsistencies between the lip movements and the audio (lip-sync discrepancies) can be observed, which strongly indicate that the video may have been generated using AI. To verify this, we analysed the clip using several AI-detection tools. The AI detection tool Hive Moderation indicated that the video has a 99% probability of being AI-generated.
Conclusion:
Our research found that the viral video of Rajnath Singh is a deepfake. He has not made any statement supporting Israel or opposing Iran. The original video is from a Sindhi community event in Delhi, which has been digitally altered using AI to spread a misleading claim.
Executive Summary:
A video online alleges that people are chanting "India India" as Ohio Senator J.D. Vance meets them at the Republican National Convention (RNC). This claim is not correct. The CyberPeace Research team’s investigation showed that the video was digitally changed to include the chanting. The unaltered video was shared by “The Wall Street Journal” and confirmed via the YouTube channel of “Forbes Breaking News”, which features different music performing while Mr. and Mrs. Usha Vance greeted those present in the gathering. So the claim that participants chanted "India India" is not real.
Claims:
A video spreading on social media shows attendees chanting "India-India" as Ohio Senator J.D. Vance and his wife, Usha Vance greet them at the Republican National Convention (RNC).
Fact Check:
Upon receiving the posts, we did keyword search related to the context of the viral video. We found a video uploaded by The Wall Street Journal on July 16, titled "Watch: J.D. Vance Is Nominated as Vice Presidential Nominee at the RNC," at the time stamp 0:49. We couldn’t hear any India-India chants whereas in the viral video, we can clearly hear it.
We also found the video on the YouTube channel of Forbes Breaking News. In the timestamp at 3:00:58, we can see the same clip as the viral video but no “India-India” chant could be heard.
Hence, the claim made in the viral video is false and misleading.
Conclusion:
The viral video claiming to show "India-India" chants during Ohio Senator J.D. Vance's greeting at the Republican National Convention is altered. The original video, confirmed by sources including “The Wall Street Journal” and “Forbes Breaking News” features different music without any such chants. Therefore, the claim is false and misleading.
Claim: A video spreading on social media shows attendees chanting "India-India" as Ohio Senator J.D. Vance and his wife, Usha Vance greet them at the Republican National Convention (RNC).
Claimed on: X
Fact Check: Fake & Misleading
In the rich history of humanity, the advent of artificial intelligence (AI) has added a new, delicate aspect. The aspect of promising technological advancement has the potential to either enrich the nest of our society or destroy it entirely. The latest straw in this complex nest is generative AI, a frontier teeming with both potential and perils. It is a realm where the ethereal concepts of cyber peace and resilience are not just theoretical constructs but tangible necessities.
The spectre of generative AI looms large over the digital landscape, casting a long shadow on the sanctity of data privacy and the integrity of political processes. The seeds of this threat were sown in the fertile soil of the Cambridge Analytica scandal of 2018, a watershed moment that unveiled the extent to which personal data could be harvested and utilized to influence electoral outcomes. However despite the indignation, the scandal resulted in meagre alterations to modus operandi of digital platforms.
Fast forward to the present day, and the spectre has only grown more ominous. A recent report by Human Rights Watch has shed light on the continued exploitation of data-driven campaigning in Hungary's re-election of Viktor Orbán. The report paints a chilling picture of political parties leveraging voter databases for targeted social media advertising, with the ruling Fidesz party even resorting to the unethical use of public service data to bolster its voter database.
The Looming Threat of Disinformation
As we stand on the precipice of 2024, a year that will witness over 50 countries holding elections, the advancements in generative AI could exponentially amplify the ability of political campaigns to manipulate electoral outcomes. This is particularly concerning in countries where information disparities are stark, providing fertile ground for the seeds of disinformation to take root and flourish.
The media, the traditional watchdog of democracy, has already begun to sound the alarm about the potential threats posed by deepfakes and manipulative content in the upcoming elections. The limited use of generative AI in disinformation campaigns has raised concerns about the enforcement of policies against generating targeted political materials, such as those designed to sway specific demographic groups towards a particular candidate.
Yet, while the threat of bad actors using AI to generate and disseminate disinformation is real and present, there is another dimension that has largely remained unexplored: the intimate interactions with chatbots. These digital interlocutors, when armed with advanced generative AI, have the potential to manipulate individuals without any intermediaries. The more data they have about a person, the better they can tailor their manipulations.
Root of the Cause
To fully grasp the potential risks, we must journey back 30 years to the birth of online banner ads. The success of the first-ever banner ad for AT&T, which boasted an astounding 44% click rate, birthed a new era of digital advertising. This was followed by the advent of mobile advertising in the early 2000s. Since then, companies have been engaged in a perpetual quest to harness technology for manipulation, blurring the lines between commercial and political advertising in cyberspace.
Regrettably, the safeguards currently in place are woefully inadequate to prevent the rise of manipulative chatbots. Consider the case of Snapchat's My AI generative chatbot, which ostensibly assists users with trivia questions and gift suggestions. Unbeknownst to most users, their interactions with the chatbot are algorithmically harvested for targeted advertising. While this may not seem harmful in its current form, the profit motive could drive it towards more manipulative purposes.
If companies deploying chatbots like My AI face pressure to increase profitability, they may be tempted to subtly steer conversations to extract more user information, providing more fuel for advertising and higher earnings. This kind of nudging is not clearly illegal in the U.S. or the EU, even after the AI Act comes into effect. The market size of AI in India is projected to touch US$4.11bn in 2023.
Taking this further, chatbots may be inclined to guide users towards purchasing specific products or even influencing significant life decisions, such as religious conversions or voting choices. The legal boundaries here remain unclear, especially when manipulation is not detectable by the user.
The Crucial Dos/Dont's
It is crucial to set rules and safeguards in order to manage the possible threats related to manipulative chatbots in the context of the general election in 2024.
First and foremost, candor and transparency are essential. Chatbots, particularly when employed for political or electoral matters, ought to make it clear to users what they are for and why they are automated. By being transparent, people are guaranteed to be aware that they are interacting with automated processes.
Second, getting user consent is crucial. Before collecting user data for any reason, including advertising or political profiling, users should be asked for their informed consent. Giving consumers easy ways to opt-in and opt-out gives them control over their data.
Furthermore, moral use is essential. It's crucial to create an ethics code for chatbot interactions that forbids manipulation, disseminating false information, and trying to sway users' political opinions. This guarantees that chatbots follow moral guidelines.
In order to preserve transparency and accountability, independent audits need to be carried out. Users might feel more confident knowing that chatbot behavior and data collecting procedures are regularly audited by impartial third parties to ensure compliance with legal and ethical norms.
Important "don'ts" to take into account. Coercion and manipulation ought to be outlawed completely. Chatbots should refrain from using misleading or manipulative approaches to sway users' political opinions or religious convictions.
Another hazard to watch out for is unlawful data collecting. Businesses must obtain consumers' express agreement before collecting personal information, and they must not sell or share this information for political reasons.
At all costs, one should steer clear of fake identities. Impersonating people or political figures is not something chatbots should do because it can result in manipulation and false information.
It is essential to be impartial. Bots shouldn't advocate for or take part in political activities that give preference to one political party over another. In encounters, impartiality and equity are crucial.
Finally, one should refrain from using invasive advertising techniques. Chatbots should ensure that advertising tactics comply with legal norms by refraining from displaying political advertisements or messaging without explicit user agreement.
Present Scenario
As we approach the critical 2024 elections and generative AI tools proliferate faster than regulatory measures can keep pace, companies must take an active role in building user trust, transparency, and accountability. This includes comprehensive disclosure about a chatbot's programmed business goals in conversations, ensuring users are fully aware of the chatbot's intended purposes.
To address the regulatory gap, stronger laws are needed. Both the EU AI Act and analogous laws across jurisdictions should be expanded to address the potential for manipulation in various forms. This effort should be driven by public demand, as the interests of lawmakers have been influenced by intensive Big Tech lobbying campaigns.
At present, India doesn’t have any specific laws pertaining to AI regulation. Ministry of Electronics and Information Technology (MEITY), is the executive body responsible for AI strategies and is constantly working towards a policy framework for AI. The Niti Ayog has presented seven principles for responsible AI which includes equality , inclusivity, safety, privacy, transparency, accountability, dependability and protection of positive human values.
Conclusion
We are at a pivotal juncture in history. As generative AI gains more power, we must proactively establish effective strategies to protect our privacy, rights and democracy. The public's waning confidence in Big Tech and the lessons learned from the techlash underscore the need for stronger regulations that hold tech companies accountable. Let's ensure that the power of generative AI is harnessed for the betterment of society and not exploited for manipulation.
Reference
McCallum, B. S. (2022, December 23). Meta settles Cambridge Analytica scandal case for $725m. BBC News. https://www.bbc.com/news/technology-64075067
Hungary: Data misused for political campaigns. (2022, December 1). Human Rights Watch. https://www.hrw.org/news/2022/12/01/hungary-data-misused-political-campaigns
Statista. (n.d.). Artificial Intelligence - India | Statista Market forecast. https://www.statista.com/outlook/tmo/artificial-intelligence/india
