(Photo by RawPixel via RawPixel/CC0 1.0 DEED)
In the spring of 2025, war broke out in South Asia—but not in the way most imagined. There were fighter jets in the sky, yes. Drones, missiles, and military movements played their part. But before the first Rafale jet took off or a single border was crossed, the battle had already begun—in the palms of over two billion people. It arrived as trending hashtags, deepfake confessions, fake NGO reports, viral war videos, and AI-generated rumors sent via WhatsApp, X and voice notes.
South Asia, home to more than 2.08 billion people, is one of the most densely populated and politically tense regions on Earth. At the heart of its tensions lie two rival nuclear-armed neighbours: India and Pakistan. Since their independence from British colonial rule in 1947, these nations have fought four full-scale wars, three of which were over the disputed region of Kashmir.
Kashmir is not just a piece of land; it’s a strategic goldmine. The region controls critical water sources, including the Indus River system, which fuels hydroelectric power and irrigation for millions in both countries. It also feeds parts of Afghanistan and China, making its control even more significant for regional stability.
The Indus Waters Treaty of 1960, brokered by the World Bank, was designed to ensure the peaceful sharing of water. Under this treaty, India controls the eastern rivers while Pakistan controls the western rivers. This treaty survived wars and global tensions until April 2025, when India unilaterally suspended its participation following a militant attack in Kashmir. Using water as a geopolitical lever, as India halted the treaty with stated intentions to manage water flow more assertively, Pakistan declared it an “act of war”.
Despite ceasefire agreements and international mediation efforts, tensions have persisted for decades, often flaring into cross-border skirmishes and militant attacks. Both governments maintain large armies, with India ranked the fourth strongest military in the world and Pakistan ranked twelfth in terms of active personnel.
The New Battleground
In South Asia, the nature of conflict has changed. Wars are no longer just about tanks and troops but about data, influence, and the digital mind-space. This new era—called Fifth-Generation Warfare (5GW)—relies on artificial intelligence, fake narratives, cyberattacks, and psychological manipulation rather than traditional weapons.
At the centre of 5GW is Artificial Intelligence (AI): technology that can scan millions of messages, generate deepfakes, and spread propaganda at speeds no human can match. Armed with bots and fake news, governments and non-state actors now battle for control over what people think, feel, and believe, right from their smartphones.
This digital arms race became tragically real on April 22, 2025, when a deadly attack in Pahalgam, Kashmir, killed 26 civilians. India quickly blamed Pakistan-based militants, pointing to encrypted messages traced to Balochistan, a Pakistani province. In response, India’s media—driven by bot networks and AI messaging systems—fueled a nationwide call for retaliation.
On May 7, India launched Operation Sindoor: a joint cyber and aerial offensive targeting areas in Pakistan-controlled Kashmir and Punjab. Pakistan hit back with Operation Banyanul Marsus, a counterstrike blending air defence and AI-guided disinformation. This wasn’t just a military exchange. It marked a turning point: a clash where artificial intelligence shaped battlefield strategy, media narratives, and public emotions.
Welcome to South Asia’s age of algorithmic warfare.
The Road to Pahalgam: Algorithmic Shadows Before the Fire
Between April 20 and 22, 2025, Indian intelligence detected tourists being tracked via cellphone and messaging app locations. Following the April 22 attack, India blamed the Resistance Front, a splinter group of a UN-sanctioned terrorist outfit linked to Pakistan. However, Pakistan denied the claim and proposed a joint investigation, but India responded by closing borders, airspace, and visas, and even withdrawing from the Indus Waters Treaty.
Drone footage of the attack surfaced on Telegram, and botnets on X amplified the reach. Digital forensics experts warn that altered metadata via image editors, manipulated file timestamps, and deepfake video tricks are now common in disinformation, blurring the line between real and staged events.
The Digital Battlefield: From Narwan to Shadow Gate
In the hours following the attack, both India and Pakistan activated AI-integrated war protocols—codenamed Narwan and Shadow Gate, respectively.
Operation Narwan (India) – Pre-Combat AI Fusion
Narwan was India’s conflict prediction system, developed jointly with Israeli tech firms. It worked by pulling together many types of information (videos, audio, maps, and expert tips) and combining them to help military leaders make quick, informed decisions. The Indian Space Research Organisation layered satellite data with information gathered from social media, which analysts scanned in real time for local posts in Urdu, Kashmiri, Punjabi, and regional dialects. This process enabled Narwan to predict potential riot zones and detect signs of unusual gatherings before they occurred.
Narwan integrated advanced AI to craft believable counter-messages, which trusted influencer accounts actively spread, especially on Instagram and WhatsApp groups in Kashmir and other parts of Pakistan.
Narwan’s neural engine aimed to predict enemy moves 6–12 hours in advance, but it failed to foresee the Pahalgam attack, highlighting the dangers of over-relying on AI and sidelining human intelligence.
Operation Shadow Gate (Pakistan) – Cyber-Defensive Reflex and Offensive Jammer
Shadow Gate was Pakistan’s answer, designed in partnership with Chinese cyber-defence units. Unlike Narwan, it emphasized disruption over prediction. It deployed AI-driven tools to disrupt Indian satellite and drone feeds, scrambling the original message, a technique known as signal obfuscation. Shadow Gate also launched fake servers that pretended to be part of India’s military network and spread false information, creating confusion in the command and control structures of the Indian Army. On top of that, it sent malware into India’s supply chain and logistics platforms.
Analysts identify Shadow Gate’s key feature as an AI-powered disinformation engine that adapts tone, slang, and religious context to craft hyper-localised posts convincing enough to fool even regional journalists.
Operation Sindoor and the Opening Gambit
Launched on the night of May 7, 2025, Operation Sindoor saw the mobilization of 5th-generation Rafale fighter jets, cyberwarfare units, and AI-enhanced reconnaissance tools. The offensive targeted alleged launch pads in Pakistan-controlled Kashmir and parts of the South and Central parts of Pakistan. This combination of physical and AI-driven attacks did create an impact, as Pakistan reported civilian casualties.
Pakistan’s Response: Banyanul Marsus and Operation Shadow Gate
Pakistan’s counter-response came swiftly. Within 24 hours of India’s first aerial attack, Pakistan initiated Operation Banyanul Marsus, deploying 4.5th-generation J-10C fighter jets acquired from China in recent years. Backed by Operation Shadow Gate, the counteroffensive started matching the Indian cyber advances in no time.
On May 11, a direct air-to-air engagement occurred between Indian and Pakistani fighter jets clashing over the Kashmir mountain ranges. India said its Narwan system helped target and block enemy signals, but Pakistan claimed Shadow Gate jammed Indian communications through its AI-powered tools. Satellite images show at least two Indian drones and one relay plane lost connection during the skirmish.
A clash between Narwan and Shadowgate shows modern war is both digital and physical. It also proves a point that human intelligence is still an essential component of decision-making. India’s AI-led Narwan missed key signs in Pahalgam due to sidelining human intel, while Pakistan’s Shadowgate blended AI with real-time tactics more effectively. The lesson: success lies in smartly combining AI and human insight—hybrid intelligence wins modern battles.
What Algorithmic War Looked Like on the Ground (and Phones)
In this conflict, the real front wasn’t only at the line of control (existing border), bunkers or cockpit radars. It was in push notifications, trending hashtags, and WhatsApp voice notes, all making rounds through mobile phones. Before the first combat jet took off, millions had already picked a side based on AI-shaped feeds. India trended #JusticeForPahalgam while Pakistan responded with #FalseFlagIndia. Each had its army—not of traditional soldiers, but of computational bots.
Resultantly, while jets clashed in the skies, a no-less aggressive war unfolded online. Indian and Pakistani information units flooded social media with manipulated content. A deepfake video of a Pakistani officer allegedly confessing to orchestrating the Pahalgam attack went viral on X, before being debunked by the Press Information Bureau (PIB) Fact Check unit and the video was labelled FAKE.
Pro-Pakistan Telegram channels circulated clips claiming Indian warplanes were shot down. Fact-checkers later identified the video as fake, confirming it came from a 2019 training exercise and that editors had used AI to alter military insignia and timestamps.
According to open-source investigations from Oxford’s Computational Propaganda Project (COMPROP), the Indian and Pakistani information ecosystems have repeatedly engaged in bot-driven and hashtag-coordinated campaigns during geopolitical flashpoints. Ongoing probes into May 2025 disinformation draw on 2019 Indian surgical strikes within Pakistani territory findings, showing military-led messaging, and digital cross-border clashes.
After Trump tweeted the ceasefire, AI-generated warfare continued silently. Actors rebranded old war footage and geo-tagged it to current conflict zones. They fabricated fake NGO reports using human rights language to provoke international outrage. At the same time, cyber attackers targeted payment and banking systems by deploying social engineering malware disguised as aid donation platforms. This wasn’t just a war about truth—it was a war against truth.
Cognitive Warfare and Civilian Trauma
Over four intense days, the India‑Pakistan clash spilled from battlefields into everyday life in unprecedented ways. In Punjab (India), telecom blackouts sparked fears of nuclear escalation. In Pakistan, panic buying of essential food items surged after AI‑generated fake government advisories spread on social media.
The strategy at play was Cognitive Warfare, a form of 5GW assault that targets people’s perceptions, trust, and memories. Media outlets, wittingly or not, became vehicles in this narrative battle: one side trying to reassure its public, the other seeking to sow chaos and mistrust.
A torrent of conflicting reports, via videos on messaging apps, social media, and TV, eroded trust. Researchers note that in today’s media environment, algorithm-driven disinformation and synthetic media often make it nearly impossible for people to verify facts.
As one resident in Kashmir told the BBC’s Conflict Disinformation Monitoring Report: “What I see on my phone and what I hear on the news feel like two different realities.”
Strategic Takeaways
By May 13, militarily, Pakistan was celebrating the outcome, while India was still trying to analyze why rivals outperformed its superior 5GW arsenal. But technologically, both sides validated the next generation of AI-enabled 5GW phenomenon in a real combat situation.
| Aspect | India | Pakistan |
| Air Combat | Rafale fighter jets guided by Narwan’s surveillance | J-10C fighter jets shielded via Shadow Gate’s countermeasures |
| Cyber Operations | AI-generated deepfakes, predictive target selection | Neural network-driven jamming and AI monitoring |
| Information Warfare | Narrative control via GPT-powered propaganda units on social media | Botnet amplification of global South narratives on social media |
| Outcome | Improved real-time integration, but data vulnerabilities are concerning | Enhanced digital defence, deeper future reliance on Chinese tech |
What Lies Ahead
The four-day India-Pakistan confrontation in May 2025 was not merely a border skirmish; it was a live field trial for 21st-century hybrid warfare. Both New Delhi and Islamabad, the respective capitals of India and Pakistan, are now recalibrating their doctrines to incorporate perception engineering, machine-speed decision-making, and cognitive disruption into their national security frameworks.
Security analysts at Carnegie’s International Peace argue that the next Indo-Pak conflict may not involve bullets or bombs at all, but instead focus on satellite spoofing, AI-guided stock market destabilization, and electromagnetic pulse (EMP) tactics, powerful bursts of energy that can disable electronics, communication systems, and even power grids without any physical explosion.
Unless international law catches up with the ethical, legal, and strategic ambiguities of AI warfare, South Asia risks becoming a perpetual battleground, not of soldiers and tanks, but of AI codes, clicks, and collapses.
Edited by Light Naing
