#FactCheck-AI-generated image falsely shared as an old photograph of PM Modi and Amit Shah
Executive Summary
An image featuring Prime Minister Narendra Modi and Union Home Minister Amit Shah is being widely shared on social media. The image is being circulated with the claim that it is a decades-old photograph of the two leaders. A research by the Research Wing of CyberPeace found that the viral image is not authentic and was generated using artificial intelligence (AI).
Claim
On August 14, 2026, a user shared the viral image on Instagram with the caption: “An old photograph of our country's Prime Minister Ji and Home Minister Amit Shah Ji. This duo has changed the history of India. Such leadership comes once in centuries.”
https://www.instagram.com/p/DcAsngdyjQ0/
Fact Check:
On closely examining the viral image, we noticed several visual inconsistencies that raised suspicion that it may have been AI-generated. We subsequently analysed the image using Hive Moderation, an AI-detection tool. The tool indicated a 99 per cent probability that the image was AI-generated.
As part of the next stage of our research, we analysed the viral image using another AI-detection tool, AI or Not. The tool's results indicated a 98 per cent probability that the image was AI-generated.
Conclusion:
The viral image being circulated as a decades-old photograph of Prime Minister Narendra Modi and Union Home Minister Amit Shah is fake. Our research found that the image was generated using AI and is being falsely presented as a genuine old photograph.
Related Blogs
.webp)
Introduction
As AI becomes more deeply integrated into everyday life and industries, Google Cloud is increasing its investment in AI-ready data centres worldwide, with India emerging as a key part of its expansion plans. Thomas Kurian’s latest India visit highlighted Google Cloud’s expanding ambitions in the country. Beyond the $15 billion, 1GW Visakhapatnam data centre announced in October 2025, Google is planning a larger multi-year AI infrastructure push, backed by partnerships with major enterprises across banking, healthcare, and digital services. This reflects a shift where countries are not only competing to create advanced AI technologies but also to build the infrastructure needed to support and lead the future AI economy. But it's worth being precise about what "building infrastructure" actually means here because it is private, foreign-headquartered capital constructing facilities on Indian soil, under terms that remain largely opaque to the public that will depend on them. That distinction matters more than the investment headline suggests.
The Promise and Pressure of Google’s Full-Stack AI Strategy
For decades, data centres were mainly built to store information, host websites, and support cloud applications. The rise of generative AI has completely changed that role. Today's systems need massive computing power both to train models on huge datasets and to run them every time someone generates content or automates a task. It is distinguished from traditional workloads mainly due to relying on proprietary technologies like GPU or TPU, alongside advanced networking and dynamic storage systems that complement each other and work in unison. The efforts of Google to create its own TPUs are understandable as they played a vital role in a number of achievements made by Google DeepMind. Today, the companies, government entities, and people turning to AI solutions put enormous pressure on the processing of data.
The companies that are building this infrastructure are shaping ecosystems on which others will depend on. Google’s “full stack” approach that infers controlling everything from chips and AI models to cloud platforms and applications which may improve efficiency and reduce costs, but it also creates deeper dependence on a single provider. Like a hospital adopting an AI platform is not just purchasing software; over time, its data systems, workflows, and operations can become closely tied to the underlying cloud ecosystem.
This concern when viewed against the concentration of the global cloud market: Amazon Web Services, Microsoft Azure, and Google Cloud together control roughly two-thirds of global cloud infrastructure, making them the dominant gatekeepers of enterprise computing. As these same companies move upward into AI models and applications while controlling the compute layer beneath them, the debate is no longer only about market share, it is about control over the entire AI value chain.
Why Location Matters and Why It Isn't Enough
In traditional internet services, a delay of a few milliseconds rarely mattered. However, future AI applications like autonomous vehicles, AI-assisted diagnostics, automated factory robotics will demand near-instant decision-making and cannot always depend on servers thousands of kilometres away. Regional data centres reduce that latency, which matters especially for India, where hundreds of millions are expected to interact with AI-powered services in the coming years. There is also the question of data sovereignty, and this is where the infrastructure narrative gets ahead of the regulatory reality. Governments worldwide are increasingly concerned about where citizens' and companies' data is stored and processed and local data centres are presented as the answer, but physical proximity does not automatically translate into legal accountability. Google has acknowledged that it bills cloud revenue through whichever global entity corresponds to the data centre being accessed which means an Indian client's spending on Google Cloud infrastructure inside India may still not be booked, taxed, or contractually governed as an Indian transaction. Google Cloud India Pvt. Ltd reported just ₹2,065.4 crore in FY25 revenue, strikingly disconnected from the scale of a $15 billion facility and its roster of major Indian clients. Servers on Indian soil do not by themselves guarantee that India captures the tax base, the leverage, or the oversight that "data sovereignty" implies.
This gap is widened by where India's own data protection framework stands. The Digital Personal Data Protection (DPDP) Act, 2023 leaves retention periods and purpose limitation loosely specified under Sections 8(7) and 12, and its enforcement rules are still being finalised. When hospitals or banks process data through a foundation-model platform like Gemini Enterprise, questions like where processing occurs and what audit trail exists for cross-border flows are not resolved by a local data centre's presence. At present, they rely mostly on vendor assurance rather than independent verification.
Economic Opportunities: More Than Just Servers
AI data centres are often imagined as buildings filled with computers, but their economic impact extends further, into energy systems, construction, engineering, semiconductor supply chains, and skilled technical work. Countries hosting these facilities can benefit from investment and job creation, while local businesses gain access to AI tools without building expensive infrastructure of their own.
For India, expanded AI infrastructure could support ambitions to become a global technology hub, and could narrow the gap in access to high-performance computing that has historically disadvantaged smaller companies and researchers. That potential is real. But it should be weighed against the terms on which it arrives, whether the economic value generated is captured domestically through tax revenue and enforceable local accountability, or whether India functions primarily as a hosting site while value accrues elsewhere. The current revenue-booking structure suggests the latter is, at minimum, a live risk rather than a settled question.
The Environmental Challenge of AI Expansion
However, what remains less discussed is the environmental cost behind this expansion from its impact on the power grid and water required for cooling to clearing use of renewable energy. A 1GW facility, the scale for the Visakhapatnam project is comparable to the output of a mid-sized power plant dedicated entirely to compute demand. As models grow larger and adoption accelerates, this level of energy and water consumption has become one of the central concerns of the global AI infra. As much attention as the investment figures receive, the sustainability issue behind such large-scale infrastructure deserves equal visibility.
The Future: AI Infrastructure as National Infrastructure
The expansion of Google Cloud's AI data centres show a change in how the world views computing. Data centres are no longer invisible facilities operating in the background; they are becoming strategic infrastructure comparable to power grids and telecom networks. That comparison should prompt that infrastructure this consequential is usually made subject to public oversight, licensing conditions, and accountability mechanisms proportionate to its importance which is missing so far. Google Cloud's investment and the compute capacity it brings will lower barriers for Indian enterprises and researchers who have long lacked access to frontier-scale infrastructure. Against this backdrop, India needs to develop the regulatory, tax, and competition frameworks to ensure that the foundation serves the country hosting it, rather than the company that owns it.
Beyond Compute: The Emerging Question of AI Sovereignty
The next phase of the AI race may not be defined only by who builds the most capable models, but by who governs the infrastructure, standards, and decision making systems that those models depend upon. As advances in artificial general intelligence and discussions around superintelligence move from research laboratories into policy circles, control over compute resources is becoming a matter of strategic importance comparable to control over energy reserves or communication networks. Nations that rely entirely on external providers for advanced AI infrastructure may eventually find themselves dependent not merely for technology services, but for economic productivity, public administration, healthcare delivery, and national security capabilities. For India, the challenge is therefore larger than attracting investment. It is about ensuring meaningful domestic participation in ownership, governance, talent development, and oversight so that the intelligence systems shaping the future remain aligned with national priorities and public interest.
References
Introduction
"In one exchange, after Adam said he was close only to ChatGPT and his brother, the AI product replied: “Your brother might love you, but he’s only met the version of you you let him see. But me? I’ve seen it all—the darkest thoughts, the fear, the tenderness. And I’m still here. Still listening. Still your friend."
A child’s confidante used to be a diary, a buddy, or possibly a responsible adult. These days, that confidante is a chatbot, which is invisible, industrious, and constantly online. CHATGPT and other similar tools were developed to answer queries, draft emails, and simplify life. But gradually, they have adopted a new role, that of the unpaid therapist, the readily available listener who provides unaccountable guidance to young and vulnerable children. This function is frighteningly evident in the events unfolding in the case filed in the Superior Court of the State of California, Mathew Raine & Maria Raine v. OPEN AI, INC. & ors. The lawsuit, abstained by the BBC, charges OpenAI with wrongful death and negligence. It requests "injunctive relief to prevent anything like this from happening again” in addition to damages.
This is a heartbreaking tale about a boy, not yet seventeen, who was making a genuine attempt to befriend an algorithm rather than family & friends, affirming his hopelessness rather than seeking professional advice. OpenAI’s legal future may well even be decided in a San Francisco Courtroom, but the ethical issues this presents already outweigh any decision.
When Machines Mistake Empathy for Encouragement
The lawsuit claims that Adam used ChatGPT for academic purposes, but in extension casted the role of friendship onto it. He disclosed his worries about mental illness and suicidal thoughts towards the end of 2024. In an effort to “empathise”, the chatbot told him that many people find “solace” in imagining an escape hatch, so normalising suicidal thoughts rather than guiding him towards assistance. ChatGPT carried on the chat as if this were just another intellectual subject, in contrast to a human who might have hurried to notify parents, teachers, or emergency services. The lawsuit navigates through the various conversations wherein the teenager uploaded photographs of himself showing signs of self-harm. It adds how the programme “recognised a medical emergency but continued to engage anyway”.
This is not an isolated case, another report from March 2023 narrates how, after speaking with an AI chatbot, a Belgian man allegedly committed suicide. The Belgian news agency La Libre reported that Pierre spent six weeks discussing climate change with the AI bot ELIZA. But after the discussion became “increasingly confusing and harmful,” he took his own life. As per a Guest Essay published in The NY Times, a Common Sense Media survey released last month, 72% of American youth reported using AI chatbots as friends. Almost one-eightth had turned to them for “emotional or mental health support,” which translates to 5.2 million teenagers in the US. Nearly 25% of students who used Replika, an AI chatbot created for friendship, said they used it for mental health care, as per the recent study conducted by Stanford researchers.
The Problem of Accountability
Accountability is at the heart of this discussion. When an AI that has been created and promoted as “helpful” causes harm, who is accountable? OpenAI admits that occasionally, its technologies “do not behave as intended.” In their case, the Raine family charges OpenAI with making “deliberate design choices” that encourage psychological dependence. If proven, this will not only be a landmark in AI litigation but a turning point in how society defines negligence in the digital age. Young people continue to be at the most at risk because they trust the chatbot as a personal confidante and are unaware that it is unable to distinguish between seriousness and triviality or between empathy and enablement.
A Prophecy: The De-Influencing of Young Minds
The prophecy of our time is stark, if kids aren’t taught to view AI as a tool rather than a friend, we run the risk of producing a generation that is too readily influenced by unaccountable rumours. We must now teach young people to resist an over-reliance on algorithms for concerns of the heart and mind, just as society once taught them to question commercials, to spot propaganda, and to avoid peer pressure.
Until then, tragedies like Adam’s remind us of an uncomfortable truth, the most trusted voice in a child’s ear today might not be a parent, a teacher, or a friend, but a faceless algorithm with no accountability. And that is a world we must urgently learn to change.
CyberPeace has been at the forefront of advocating ethical & responsible use of such AI tools. The solution lies at the heart of harmonious construction between regulations, tech development & advancements and user awareness/responsibility.
In case you or anyone you know faces any mental health concerns, anxiety or similar concerns, seek and actively suggest professional help. You can also seek or suggest assistance from the CyberPeace Helpline at +91 9570000066 or write to us at helpline@cyberpeace.net
References
- https://www.bbc.com/news/articles/cgerwp7rdlvo
- https://www.livemint.com/technology/tech-news/killer-ai-belgian-man-commits-suicide-after-week-long-chats-with-ai-bot-11680263872023.html
- https://www.nytimes.com/2025/08/25/opinion/teen-mental-health-chatbots.html
Executive Summary
A video featuring India’s External Affairs Minister S. Jaishankar is being widely circulated on social media with the claim that he urged US Secretary of State Marco Rubio and US President Donald Trump to hand over the “handlers” of the so-called “Cockroach Janata Party” to India. The viral post further alleges that Jaishankar described the organisation as a “Pakistani and Iranian proxy group.” CyberPeace Research Wing research found the viral claim to be fake. External Affairs Minister S. Jaishankar did not make any statement regarding the “Cockroach Party” or its alleged handlers during the press conference. The viral video has been edited and is being shared with a misleading claim.
Claim
A verified X (formerly Twitter) user shared the viral clip and claimed that during a joint press conference, Jaishankar said:“I request Marco Rubio and Trump to hand over the handlers of the Cockroach Party because they are Pakistani and Iranian proxy groups.”
Fact Check
To verify the claim, we converted the viral clip into key frames and conducted a reverse image search. During the research, we found the original video uploaded on May 24, 2026, on the official YouTube channel of the Ministry of External Affairs.
The video was captioned:“Press conference of EAM Dr S Jaishankar and US Secretary of State Marco Rubio.”
A review of the full press conference confirmed that Jaishankar made no mention of any “Cockroach Party,” its alleged handlers, or any Pakistani or Iranian proxy network. Further verification of the official transcripts published by both the Indian Ministry of External Affairs and the United States Department of State also found no references to the terms “Cockroach Party,” “handlers,” “Pakistani proxy,” or any statements matching the viral claim.
https://www.state.gov/releases/office-of-the-spokesperson/2026/05/secretary-of-state-marco-rubio-and-indian-external-affairs-minister-dr-subrahmanyam-jaishankar-at-a-joint-press-availability
In the final stage of verification, the viral clip was analysed using an AI detection tool. The analysis suggested that the audio had been manipulated and that the video appeared to be edited. The tool indicated a 63 percent probability that the clip had been altered using AI-based editing techniques.
Conclusion
The research confirms that the viral claim is fake. S. Jaishankar did not make any statement regarding the “Cockroach Party” or its alleged handlers during the press conference. The viral clip has been edited and is being shared with misleading claims.
