Has India made significant moves in AI or still continue to be at a nascent stage?

During Operation Sindoor, the Indian Army deployed an electronic intelligence coalition application developed by the Directorate General of Information Systems. The system was trained on nearly 26 years of data collected by Indian agencies and the armed forces. It included information on adversary sensors, their frequencies, movement patterns and the units they were attached to. 

Indian AI models (Large Language Models and foundation models) are trained to be locally relevant by prioritizing Indian languages, localized datasets, and compute-efficiency. This process relies on specific data integration, governmental subsidies, and resource-saving architectures.

AI models train to observe, analyze, and predict human behaviour and movements by processing massive, multimodal datasets through machine learning architectures. They identify spatial, temporal, and psychological patterns to forecast actions, movements, and target locations.

Artificial intelligence models do not possess the autonomous capability to destroy Pakistan, but they serve as powerful, disruptive tools in modern warfare, cybersecurity, and information control. AI-driven systems are utilized in modern defence networks for threat tracking, precision targeting, and the rapid deployment of autonomous drones. This automation has escalated the pace of border skirmishes, raising concerns about miscalculation in the region. AI is deployed both to strengthen national cyber defences and as an offensive vector to disrupt critical infrastructure (e.g., power grids and communication systems). Generative AI is actively used to fabricate synthetic media, such as deepfake audio and video, which can be weaponized during geopolitical tensions to manipulate public perception, stoke unrest, or spread panic. AI algorithms processing massive datasets (like mobile phone metadata) have been historically utilized in targeted military campaigns, occasionally leading to controversy over algorithm-based targeting accuracy. While these systems are highly disruptive and can cause severe localized damage, they are ultimately tools operated by human militaries and state actors rather than independent entities capable of large-scale annihilation on their own. 

India’s AI models are not capable of “destroying” Pakistan. While India has developed and integrated advanced artificial intelligence for military and defence applications—such as automated threat interception, electronic intelligence analysis, and smart border surveillance—these technologies act as strategic deterrence and precision defence tools rather than apocalyptic weapons. The destructive capacity between the two nations relies on nuclear arsenals, not AI. The integration of AI into military frameworks by both countries simply alters the speed of warfare, raising serious concerns about rapid escalation and potential miscalculation rather than total annihilation. 

Indian AI models do not physically disrupt Pakistan’s nuclear programme. Instead, they reshape deterrence by enabling autonomous, real-time conventional military actions. The integration of artificial intelligence by the Indian military creates complex strategic implications in several ways:

Accelerated Decision-Making: AI drastically shortens the timelines for detection and response to missile and drone threats. Indian automated defence systems, like the autonomous AI-driven combat cloud Akashir, rely on image recognition and real-time sensor data to identify and neutralize incoming threats in seconds, potentially altering offensive and defensive postures.

Enhanced Surveillance and Targeting: Machine learning algorithms trained on satellite imagery and real-time electronic signatures allow for precise, automated tracking of military infrastructure along the border. This level of precision intelligence enhances the capability for targeted operations against military assets.

Deterrence vs. Escalation Risk: While advanced AI enables more reliable early-warning architectures, it also increases crisis instability. The speed of AI-assisted operations may compress political decision-making windows during a crisis, heightening the risk of accidental escalation or misinterpretation between the nuclear-armed neighbours. 

India maintains advanced technical facilities and specialized agencies dedicated to equipping Indian spies and security forces with state-of-the-art technical intelligence (TECHINT) capabilities.

These capabilities include specialized satellite reconnaissance, signals intelligence, cyber espionage, and covert surveillance assets:

National Technical Research Organisation (NTRO): Established in 2004, this premier technical intelligence agency operates directly under the National Security Advisor in the Prime Minister’s Office. The NTRO serves as a “super-feeder” agency, providing vital communications monitoring, satellite imagery, and cybersecurity intelligence to both the Intelligence Bureau (IB) and the Research & Analysis Wing (R&AW).

National Institute of Cryptology Research and Development (NICRD): Managed by the NTRO, this is the first institute of its kind in Asia and develops cutting-edge cryptographic tools and secure communication methods for covert operations.

Space-Based Surveillance & Spy Satellites: India relies on an expanding indigenous constellation of military and civilian satellites—such as the EMISAT electronic intelligence satellite and various RISAT radar imaging satellites—to track enemy radars, communications, and border troop movements.

Airborne Reconnaissance: The Defence Research and Development Organisation (DRDO) continues to develop and integrate advanced surveillance aircraft, including customized ISTAR (Intelligence, Surveillance, Target Acquisition, and Reconnaissance) jets used as flying command centres to monitor threats across borders.

India’s AI surveillance architecture is highly pervasive, deeply integrated into both public spaces and digital governance. Technologies like facial recognition and predictive policing algorithms are being rapidly deployed across law enforcement and urban management, heavily outpacing the development of specific legal safeguards and regulatory oversight.

Key facets of this infrastructure include:

Urban Video Surveillance: Cities are utilizing AI-powered CCTV for real-time crowd management and anomaly detection. For instance, under the Safe City Project, the Greater Chennai Police have deployed thousands of AI-enabled cameras to detect loitering, overcrowding, and vandalism in real-time.

Biometric & Facial Recognition: The push for systems like the National Automated Facial Recognition System (NAFRS) aims to link vast biometric databases. This allows for widespread, automated tracking of individuals across state lines.

Digital Data Integration: Massive data-fication programs like the Crime and Criminal Tracking Network & Systems (CCTNS) feed into predictive policing models, while seamless public convenience systems (e.g., biometric air travel like DigiYatra) create pervasive tracking mechanisms.

Regulatory & Privacy Concerns: While the Digital Personal Data Protection Act (DPDPA) exists to manage data usage, broad exemptions for government agencies allow for unchecked state monitoring. Experts and civil rights organizations continue to point out that this ecosystem often lacks the legal proportionality and judicial oversight dictated by the Supreme Court’s landmark Puttaswamy privacy ruling.

To explore the specific guidelines for responsible practices and current regulatory debates, one can review policy updates from the NITI Aayog Frontier Tech Portal or analyze ongoing civil liberties discussions via the Internet Freedom Foundation.

Artificial Intelligence (AI) has significantly sped up India’s military targeting cycle, or “kill chain” (the sequence of finding, fixing, tracking, targeting, engaging, and assessing a threat). AI integrations have drastically reduced decision-making times while increasing precision and lethality.

The acceleration is driven by several interconnected advancements:

Shortening the Sensor-to-Shooter Loop: AI platforms integrate inputs from ground, aerial, and electronic sensors to create automated real-time situational heat maps. By doing the heavy lifting of data fusion, commanders can identify and track enemy assets in a fraction of the time.

Operation Sindoor: The Indian Army utilized traditional AI systems, including the Electronic Intelligence Collation and Analysis System (ECAS), during live operations. By parsing 26 years of historical electronic and frequency data, the model successfully located enemy gun and missile units with 94% accuracy, significantly slashing the time required to neutralize the threat.

Project SANJAY: This automated surveillance system fuses real-time intelligence feeds from diverse sources to generate a single common operational picture for the battlefield, effectively bypassing slow manual data analysis.

TRINETRA: This AI-driven battlefield surveillance platform aggregates multi-source intelligence to instantly pinpoint enemy sensors, aiding in faster counter-battery fire and electronic warfare. 

In the course of the conflict with Pakistan, AI-powered predictive targeting system model was able to predict and track these systems with a high degree of accuracy. “We achieved over 90 per cent accuracy,” Rajiv Kumar Sahni, who served as the Director General of Information Systems during the operation said.

This was one of the several indigenous AI tools used during the May clashes last year, to improve battlefield awareness and speed up decision-making. These systems are now being expanded, including in a military-specific large language model.  

As predictive intelligence complemented by AI advances each day, where does the buck stop?

“They’re going to continue integrating, and then that’s going to get accepted as a kind of normalization. What is emerging is not an exception, but a new baseline, where AI-mediated targeting, backed by private data ecosystems, becomes routine,” said Dr. Peter Asaro, a prominent philosopher, historian, and leading voice on the ethics, law, and policy surrounding Artificial Intelligence (AI) and autonomous weapons.

India actively conducts AI military operations and has systematically integrated artificial intelligence into its armed forces to enhance battlefield awareness, intelligence gathering, and precision targeting. 

India’s military AI and autonomous systems span several strategic and tactical applications: 

Combat and Targeting: The Indian Army has successfully utilized AI models—most notably during cross-border operations (Operation Sindoor)—to process decades of historical data, track enemy units, and significantly reduce the “kill chain” between target identification and strikes.

Border Surveillance: The military employs AI-powered smart fences and satellite imaging to monitor remote riverine gaps and detect troop movements in real time along the Line of Actual Control (LAC) and Line of Control (LoC).

Indigenous AI Infrastructure: The Ministry of Defence is collaborating with domestic tech labs (such as Sarvam AI and BharatGen) to establish a dedicated military AI centre, ensuring that operations, weapon systems, and battlefield intelligence run securely on sovereign Indian models.

Operational Platforms: Various “AI-as-a-Service” platforms are deployed, including the SAM-UN geospatial platform for mission planning and the Nabh Drishti system for real-time reporting.

Institutional oversight is provided by the Defence Artificial Intelligence Council and the Defence AI Project Agency, which guide the development and integration of these technologies across the Army, Navy, and Air Force.

Even while there are obvious ramifications for international law, this discovery creates a new pandora’s box that will be extremely challenging to close if the use of AI-enabled targeting is not governed by international frameworks and if adversaries do not follow the existing laws of warfare.

India does not have a single, codified domestic law dedicated to Artificial Intelligence in military operations. However, the use of military AI is governed by the broader framework of international treaties, emerging internal defence doctrines, and constitutional principles.

India’s approach is shaped by several key factors and principles:

No Blanket Ban on LAWS: India does not support an outright international ban on Lethal Autonomous Weapons Systems (LAWS). It maintains that AI-enabled autonomy provides a critical force multiplier, precision, and efficiency. Therefore, India has consistently voted against or abstained from UN resolutions seeking a blanket prohibition.

Compliance with International Humanitarian Law (IHL): Despite the lack of specific domestic legislation, India states that its military must comply with IHL and the Geneva Conventions. This mandates that AI deployment must observe the principles of distinction (differentiating civilians from combatants) and proportionality (avoiding excessive collateral damage).

Institutional Ethics and Policies: The Defence Research and Development Organisation (DRDO) outlines five pillars for trustworthy AI: reliability, safety, transparency, fairness, and privacy. Similarly, internal defence policies stipulate that the Acquisition Section of the military must ensure high-cost AI programs meet stringent technical and ethical standards.

Strategic Autonomy: India’s stance stems from a desire to preserve strategic autonomy and not relinquish operational advantages to neighbouring states. India focuses on the regulation of specific, harmful end-uses of weapons rather than restricting the underlying dual-use technology itself. 

Artificial intelligence in India is rapidly transforming both state governance and the operations of non-state actors. State actors use AI to automate administration and enhance national security, while political groups, corporate entities, and adversaries leverage AI algorithms for micro-targeting, surveillance, and asymmetric influence campaigns.

Influence on State Actors

Grassroots Governance: The Indian state is deploying localized AI tools (such as SabhaSaar) to transcribe and summarize Gram Sabha meetings in real-time, automating the state-citizen interface and increasing administrative transparency.

Predictive Policing & Surveillance: AI is now integrated into national criminal databases, allowing state actors to utilize massive datasets for predictive policing, as seen during major events in cities like Pune.

National Security & Defence: The Indian defence sector is shifting toward AI-integrated intelligence, surveillance, and reconnaissance (ISR) to secure borders, combat asymmetric threats, and process data faster than adversaries. 

Influence on Non-State Actors

Information Warfare: Generative AI is frequently weaponized by non-state actors and political campaigns to influence voters, spread disinformation, and create deepfakes. For example, regional elections have seen AI-generated fabricated audio and videos used to sway public opinion.

Asymmetric Capabilities: Extremist groups and non-state adversaries leverage openly available machine learning tools to enhance cyberattacks, automate phishing, and bypass defensive platforms.

Corporate & Private Sector Power: AI is blurring the line between private corporate data and state authority, as tech firms and surveillance companies secure lucrative state contracts to build proprietary public security and monitoring systems.

AI tools can theoretically be used to facilitate or streamline certain aspects of assassination attempts, and any advanced AI technology—regardless of its country of origin—can be misused for nefarious purposes like espionage, targeting, and remote disruption. 

While individual, everyday commercial AI tools are not specifically designed for violence, state-level intelligence agencies (including those in India and worldwide) increasingly employ advanced artificial intelligence for military and security operations. The intersections of these capabilities include:

Target Identification and Surveillance: AI-driven data fusion systems combine information from multiple sensors, satellite imagery, and human intelligence to precisely track, locate, and prioritize high-value targets across borders.

Reconnaissance: Advanced AI algorithms can process vast amounts of unstructured data, such as public social media posts, communication intercepts, and open-source intelligence (OSINT), to map out the daily routines, travel patterns, and security details of individuals.

Deepfakes and Deception: Generative AI tools and deepfakes can be used to generate realistic synthetic media or fake digital identities to deceive targets, bypass security protocols, or manipulate local populations into carrying out attacks.

Investigations by international agencies and independent journalists have previously documented covert, state-sponsored operations in the region involving the recruitment of local proxies and sleeper cells, though these are typically attributed to traditional intelligence methodologies rather than direct AI-driven attacks.

In India, Artificial Intelligence technology is instead actively used by law enforcement agencies (like Staqu’s Crime GPT and facial recognition software) to track down suspects, identify criminals, and solve murder cases. In the defence sector, AI is utilized for perimeter surveillance and air defence command systems rather than for targeted assassinations. 

Artificial Intelligence (AI) has significantly sped up and transformed India’s targeting cycle. By fusing multi-source data streams from satellites, drones, and radars, AI drastically compresses the “sensor-to-shooter” kill chain.

India’s military has actively integrated these systems into its operations and planning. Key technological developments include: 

Operational Success: During Operation Sindoor, the Indian Army successfully utilized AI tools that combined archival intelligence data (spanning over 25 years) with real-time drone and radar feeds. This system reportedly yielded 94% accuracy and reduced the sensor-to-shooter reaction time from days or minutes down to mere seconds.

Integrated Data Platforms: Systems like TRINETRA and Project SANJAY are being used to generate AI-powered heat maps, fusing inputs from ground and aerial sensors to create a unified, real-time operational picture for commanders.

Automated Threat Evaluation: The Electronic Intelligence Collation and Analysis System (ECAS) sift through historical frequency signatures to quickly identify, analyze, and prioritize adversary radars and surveillance systems, allowing forces to make quick and precise operational decisions.

AI is reshaping India’s national security. The adoption of AI represents a crucial paradigm shift for India, moving away from traditional net-centric warfare toward data-centric and intelligent, predictive combat.

India has built and rapidly expanded an indigenous, AI-powered target and air defence architecture following Operation Sindoor. The framework integrates homegrown sovereign AI models with multi-domain command-and-control grids to predict, track, and neutralize threats. 

Instead of fully autonomous targeting, India has deliberately focused its architecture on enhancing the speed and scope of military decision-making rather than fully uncrewed lethal operations. Key components include:

The Akashteer Command and Control System: An indigenous AI-powered air defence system developed by Bharat Electronics Limited (BEL). It integrates data from radars and airborne systems to detect threats and coordinate response units.

Intelligence and Analytics: The Indian Army’s AI stack generates heat maps, prioritizes resources, and assists with intelligence reviews rather than directly deciding individual combatant status.

Weather Integration: Apps like Anuman 2.0 utilize AI-driven weather forecasting to assist ground units in precise cross-border operations.

Institutional Structure

India’s military AI and targeting architecture operates on a distinct four-step institutional pipeline to ensure self-reliance: 

Defence AI Council: Handles strategy and resource allocation.

Defence AI Project Agency (DAIPA): Manages the execution and rollout of AI projects.

DRDO’s Centre for AI and Robotics (CAIR): Focuses on core research and development.

Service-Specific Centres: Individual branches of the military deploy and tailor AI tools for their specific domains. 

India has actively built and rapidly expanded its AI-powered military and air defence architecture following the success of Operation Sindoor. Rather than fully autonomous lethal systems, the upgrades focus on real-time intelligence synthesis, multi-domain sensor fusion, and joint operations.

The strategic shift is characterized by the following key programs and architectures: 

Unified Joint Operations Command

Air Defence Integration: The Indian Air Force’s Integrated Air Command and Control System (IACCS) is undergoing its third phase of expansion. It integrates land, sea, and air assets into a single “Common Operational Picture.”

Akashteer System: Developed by Bharat Electronics Limited (BEL), this indigenous command-and-control network uses AI to fuse data from AWACS and ground radars, automating threat evaluation and response allocation.

Upgraded AI and ISR (Intelligence, Surveillance, Reconnaissance)

Battlefield Analytics: AI is now heavily relied upon for intelligence review, generating heat maps, and generating real-time targeting analytics for equipment and troop positions.

Tactical Weather AI: Deployments of homegrown applications—such as the Anuman 2.0 AI weather forecasting tool—are being rolled out to assist field units with precision engagements.

Institutional AI Framework

The military utilizes a streamlined four-step technology pipeline connecting the Defence AI Council (strategy) with the Defence AI Project Agency (execution), supported by R&D from the Centre for Artificial Intelligence and Robotics (CAIR).

Individual service branches have established dedicated AI Centres of Excellence to continuously train algorithms on regional threat data. 

Secure Communications

Field communications have been reinforced with encrypted platforms (e.g., SAMBHAV smartphones).

These tactical communication grids are being upgraded alongside broader network developments, including indigenous quantum network breakthroughs.

Indian AI models process vast, hyper-diverse datasets using a localized tech stack and native digital infrastructure. By combining massive public datasets, shared government computing power, and indigenous foundation models, developers can efficiently scale AI to generate targeted solutions for agriculture, healthcare, and governance.

These models scale and generate actionable targets through several core mechanisms:

Digital Public Infrastructure (DPI)

India uses its expansive DPI and digitized public records—such as consent-based data sharing and Aadhaar—to feed vast, localized, and context-rich datasets into algorithms. This allows models to learn from massive real-world interactions rather than relying purely on western-centric data. 

Specialized Indigenous Models

Instead of using generalized, one-size-fits-all models, Indian developers utilize homegrown foundation models that natively support regional requirements and dialects:

BharatGen: India’s flagship government-funded multimodal AI model that integrates text, speech, and image understanding for 22 scheduled languages.

Sarvam AI: Builds speech-first models (like Bulbul for Text-to-Speech and Saaras for telephony audio) to seamlessly process and interact with code-mixed vernacular languages. 

Retrieval-Augmented Generation (RAG)

To avoid the intense costs of training massive models from scratch, Indian AI researchers rely on RAG and fine-tuning techniques. This approach connects lightweight models to specialized, domain-specific databases (e.g., local land records, healthcare logs, or agricultural databases) to generate highly accurate targets and responses on demand.

Aggressive Compute Infrastructure

To process this data at scale, India provides shared computing resources. Through the IndiaAI Mission, the government offers startups affordable access to a massive compute portal featuring thousands of GPUs, easing the hardware barrier for training complex models.

Indian defence AI systems prioritize real-time surveillance, decision-speed dominance, and autonomous operations. Their efficiency currently leans heavily on asymmetric advantages—deploying swarm drones and intelligent sensors for high-altitude border monitoring. While successful in specific regional tactics, full inter-service data integration and legacy tech modernization remain works in progress.

The efficiency of India’s present AI systems in defence spans across several key operational areas:

Border Surveillance and Intrusion Detection

Deployment: The Indian Army has actively deployed over 140 AI-based surveillance nodes along the Line of Control (LoC) and Line of Actual Control (LAC) to continuously monitor the rugged, remote terrain.

Deployment: The Indian Army has actively deployed over 140 AI-based surveillance nodes along the Line of Control (LoC) and Line of Actual Control (LAC) to continuously monitor the rugged, remote terrain.

Autonomous and Unmanned Systems

Efficiency in the Field: India deploys systems such as AI-driven loitering munitions (e.g., ALS50) and swarm drones, drastically reducing risks to human personnel while allowing for precision strikes on hostile targets.

Logistics & Maintenance: The Defence Research and Development Organisation (DRDO) and the armed forces utilize AI-driven predictive maintenance for combat platforms. This increases overall fleet readiness by proactively identifying maintenance needs before mechanical failures happen.

Cyber Warfare and Electronic Dominance

Efficiency: AI algorithms are highly proficient in cyber defence, helping to detect anomalous network activities and neutralize cyber intrusions in real-time. They adapt communication protocols to prevent jamming, which is crucial for secure tactical operations.

Current Challenges and Limitations

Despite notable advancements, the efficiency of Indian defence AI faces structural hurdles:

Interoperability: Achieving fully integrated data pipelines between the Army, Navy, and Air Force remains an ongoing challenge. Without unified data sharing, AI models cannot cross-reference the comprehensive operational picture required for large-scale, network-centric warfare.

Resource Scaling: India still lags behind global leaders like the US and China in absolute scale. To offset this, strategic doctrines prioritize niche, indigenous AI development over competing symmetrically across all domains.

AI models are rapidly becoming both more powerful and highly efficient. Rather than just copying massive global models, developers and institutions are focusing on Small Language Models (SLMs) and multilingual capabilities designed to run efficiently on limited computational resources.

Key initiatives and developments driving this trend include:

Sovereign & Multilingual AI: Projects like the BharatGen consortium and models from startups like Sarvam AI are building localized frameworks that accurately process numerous Indic languages. These models prioritize high accuracy in document understanding and localized translation while requiring much less compute.

Real-World Focus: The core strategy is AI diffusion—deploying lightweight AI in agriculture, healthcare, and governance. These application-focused models execute tasks much faster and more economically than huge, generalized Large Language Models (LLMs).

National Compute Infrastructure: Supported by the IndiaAI Mission, the country is expanding hardware capacity. Subsidized access to massive processing power combined with a massive tax holiday for data centre infrastructure is giving native developers the tools to optimize models without huge corporate budgets.

India is prioritizing efficiency to scale AI adoption smoothly across a linguistically fragmented and infrastructure-constrained market, turning out highly capable solutions tailored specifically to the country’s needs.

In India, a single drone conducts surveillance with exceptional power and efficiency. It drastically reduces response times by up to 50% compared to ground patrols. Operators must fly in permitted zones using the DigitalSky Platform and comply with DGCA Drone Rules for safe, legal operations.

How Powerful Are Single Drones?

A single commercial or tactical drone is a highly capable asset for real-time intelligence and security:

Payload Versatility: Standard surveillance drones can carry heavy-duty day/night optical zoom cameras, thermal sensors, and infrared lenses. This makes them ideal for 24-hour monitoring.

Advanced Features: High-end models (such as those by indigenous manufacturers like IdeaForge or Asteria Aerospace) offer AI-driven threat and motion detection, allowing them to track moving targets and identify anomalies autonomously.

Industrial & Military Utility: From tracking contraband at prison perimeters to monitoring crowd sizes at political rallies, single drones provide a continuous, high-definition “eye in the sky” without putting security personnel in harm’s way.

How Efficient Are They?

Drones offer high operational and economic efficiency: 

Cost-Effectiveness: While initial investments range from ₹1.5 lakh for standard models to over ₹10 lakh for high-altitude/military-grade drones, single drones save money long-term by reducing reliance on large security teams or manned helicopters.

Rapid Deployment: They can be launched in minutes, checking perimeters and identifying blind spots 30 to 50 times faster than traditional foot patrols.

Terrain Accessibility: Single drones can easily scan inaccessible or hazardous terrains like deep forests, mountainous borders, or disaster-hit zones

Key Regulations to Note

To maintain efficiency and stay legally compliant in India, operators must follow DGCA Guidelines:

No-Fly Zones: Drones cannot be flown within 5 km of major airports, 25 km of international borders, and over sensitive defence/strategic installations without special clearance.

Flight Ceilings: The maximum operating altitude for most civilian drones is capped at 400 feet (120) meters) above ground level.

Licensing: While nano drones (<250 g) do not require a remote pilot license, flying heavier micro and small drones for commercial surveillance purposes requires DGCA certification.

In India, a single drone conducts surveillance with exceptional power and efficiency. It drastically reduces response times by up to 50% compared to ground patrols. Operators must fly in permitted zones using the DigitalSky Platform and comply with DGCA Drone Rules for safe, legal operations.

How Powerful Are Single Drones?

A single commercial or tactical drone is a highly capable asset for real-time intelligence and security: 

Payload Versatility: Standard surveillance drones can carry heavy-duty day/night optical zoom cameras, thermal sensors, and infrared lenses. This makes them ideal for 24-hour monitoring.

Advanced Features: High-end models (such as those by indigenous manufacturers like IdeaForge or Asteria Aerospace) offer AI-driven threat and motion detection, allowing them to track moving targets and identify anomalies autonomously.

Industrial & Military Utility: From tracking contraband at prison perimeters to monitoring crowd sizes at political rallies, single drones provide a continuous, high-definition “eye in the sky” without putting security personnel in harm’s way. 

How Efficient Are They?

Drones offer high operational and economic efficiency: 

Cost-Effectiveness: While initial investments range from ₹1.5 lakh for standard models to over ₹10 lakh for high-altitude/military-grade drones, single drones save money long-term by reducing reliance on large security teams or manned helicopters.

Rapid Deployment: They can be launched in minutes, checking perimeters and identifying blind spots 30 to 50 times faster than traditional foot patrols.

Terrain Accessibility: Single drones can easily scan inaccessible or hazardous terrains like deep forests, mountainous borders, or disaster-hit zones.

Key Regulations to Note

To maintain efficiency and stay legally compliant in India, operators must follow DGCA Guidelines:

No-Fly Zones: Drones cannot be flown within 5 km of major airports, 25 km of international borders, and over sensitive defence/strategic installations without special clearance.

Flight Ceilings: The maximum operating altitude for most civilian drones is capped at 400 feet (120 meters) above ground level.

Licensing: While nano drones (<250 g) do not require a remote pilot license, flying heavier micro and small drones for commercial surveillance purposes requires DGCA certification. 

India’s tech ecosystem features multiple platforms and environments similar to Fire Factory. Primary examples include Activate VC (India’s native AI venture studio), Sarvam AI Startup Program for foundational model building, and the Yotta Shakti Cloud AI Factory for sovereign hardware compute.

Because “Fire Factory” can refer to different concepts, India’s AI landscape offers several direct equivalents depending on what one is looking for:

AI Venture Builders & Studios

If one is referring to a startup studio/venture builder that co-creates and funds AI ventures from scratch:

Activate VC: A Bengaluru-based native AI venture platform that partners with technical founders before incorporation, investing up to $3 million to co-build and scale ideas from 0 to 1. 

Together AI Studio: An initiative by the Together Fund that provides funding, mentorship, and cloud credits to help AI-first startups quickly transition from prototypes to enterprise-ready solutions

AI Compute & Sovereign Data Centres

If one is referring to an “AI Factory” in the context of infrastructure (NVIDIA’s term for large-scale GPU data centres built to produce AI):

Yotta Shakti Cloud AI Factory: Based in Navi Mumbai, this is India’s first sovereign platform purpose-built for large-scale AI development. Powered by NVIDIA’s Hopper GPUs, it allows local enterprises and startups to train models securely within the country.

Providers like Tata Communications and E2E Networks: These companies are also rolling out massive accelerated GPU clusters for Indian businesses to eliminate infrastructure complexity and host models.

Business Intelligence & Analytics

If one is looking for the Mumbai-based AI analytics platform Fire AI (sometimes associated with fire-related data/BI), which connects to hundreds of data sources to automate reporting for SMEs:

Fire AI: A CERT-IN certified Indian platform that enables users to interact with vast databases (like Tally, Oracle, or simple Excel sheets) using plain-English queries to generate instant, actionable business dashboards. 

India has developed and utilized AI platforms similar to “The Gospel” AI. The concept exists in India across two distinctly different contexts: military AI targeting systems and spiritual chatbots.

Military AI (Targeting and Defence)

In the context of military technology, “The Gospel” is an AI system used to rapidly identify and generate structural targets. India maintains advanced sovereign AI capabilities for similar defence, intelligence, and surveillance purposes. Through localized defence tech initiatives and platforms like BharatGPT, India is integrating agentic and conversational AI into its strategic and logistical operations to match global defence capabilities. 

Spiritual AI (Interpreting Sacred Texts)

If one is referring to the term “Gospel” in a religious or conversational sense, India has created similar, locally tailored AI platforms. Because of the country’s rich philosophical and religious heritage, several Indian AI models act as digital guides.

Religious Chatbots: India has rolled out specialized AI models and chatbots trained on sacred texts (such as the Bhagavad Gita) that answer spiritual questions, dispense dharma, and provide scripture-backed guidance.

General Indian AI Platforms: For broad conversational, consumer, and agency tasks, platforms like Ola Krutrim serve as India’s native AI assistants. 

Whether one is looking for sovereign military intelligence or localized spiritual chatbots, India’s robust developer community has built or integrated AI systems serving similar functions

During recent military conflicts involving Iran, artificial intelligence targeting systems like Israel’s “The Gospel” (Habsora) have faced intense criticism and allegations of misidentification. Critics have pointed to the “Police Park” (Park-e Police) incident in Tehran, where a public park was struck, suggesting the AI flagged the site based on the word “Police” in its name without adequate human verification. 

The Gospel, developed by the Israel Defence Forces (IDF) signals intelligence unit, is an AI-driven system designed to automate the extraction of intelligence and rapidly identify military targets. In complex conflicts, militaries often generate large datasets of potential targets to accelerate the operational tempo.

The “Police Park” strike in Tehran became a prominent example highlighted by analysts—such as Trita Parsi of the Quincy Institute for Responsible Statecraft—who argued that automated targeting systems and the rush to approve large volumes of targets led to insufficient oversight. Critics argue that relying on algorithms to process context-blind data can lead to strikes on non-military infrastructure or civilian sites if human review is rushed or bypassed. 

While the exact operational data and algorithms remain classified, the incident sparked widespread international debate regarding the risks of autonomous systems and the need for stricter ethical regulations on AI in warfare. Independent investigations into the full operational targeting mechanisms are heavily complicated by the fog of war and the closed nature of military intelligence.

In the Indian tech and policy ecosystem, the “AI Gospel” refers to the guiding philosophy of making Artificial Intelligence accessible, inclusive, and tailored to the country’s diverse, multilingual population. It centres on the concept of “AI for All,” a vision spearheaded by initiatives like the National AI Portal of India (INDIAai) and backed by sovereign AI models like Sarvam AI. 

While the term can mean different things across different domains, it primarily falls into three specific interpretations in India:

The Tech & Policy Vision (“AI for All”)

Championed by the Indian government and industry leaders, this gospel translates to “Sovereign AI.” Because over a billion Indians speak diverse languages, western models fall short. 

The Mission: Developing localized large language models (LLMs), like BharatGPT, that process regional languages, dialects, and local contexts.

The Focus: Utilizing AI for societal good—specifically in healthcare, agriculture, localized digital payments (e.g., UPI integrations), and rural education.

The Cultural Translation (Faith-Based Chatbots)

In a highly spiritual and religious nation, the “AI gospel” has also been taken literally. Many tech developers are bridging the gap between faith and technology, resulting in localized AI that provides daily spiritual guidance and scripture interpretations.

GitaGPT: A prominent example of an AI chatbot trained entirely on the 700 verses of the Bhagavad Gita. Users across India consult it to find answers to inner crises and gain ethical and moral guidance.

Just Like Me: A platform trending across India that allows users to converse with AI avatars of spiritual gurus and historical deities for a few rupees per minute

The Generative Music & Media Trend

If is one is referring to social media and pop culture, the “AI Gospel” points to the viral trend of using generative AI tools like Suno AI to create AI-generated gospel and worship songs. Indian creators are utilizing these tools to translate traditional Hindi and regional (e.g., Sadri) hymns into polished, AI-assisted gospel music channels.

The Military Targeting System

If one is asking about the military artificial intelligence system, “Where’s Daddy?” is an algorithmic tracking tool used by the Israeli military in Gaza.

What it does: The system uses mobile phone location tracking to monitor suspected militants. It is designed to alert operators exactly when the target enters their private family residence, facilitating airstrikes while they are at home.

Controversy: The system—alongside another AI programme named “Lavender”—has faced severe international scrutiny from experts and the UN for generating large-scale collateral damage and civilian casualties in residential areas. 

Artificial Intelligence (AI) has drastically enhanced Indian defence targeting capabilities by slashing response times, improving strike precision, and reducing human casualties. One may wish to explore broader defence tech initiatives on the Defence Research and Development Organisation (DRDO) and the Innovations for Defence Excellence (iDEX) portals. 

The strategic shift towards AI-enabled combat has profoundly upgraded how India detects and neutralizes threats: 

Reduced Kill Chains: By integrating AI models with historical and real-time data, the Indian military has achieved significant accuracy rates (e.g., up to 94% in operational testing) when identifying and neutralizing enemy artillery and missile units. This shrinks the time gap between detection and strikes to just minutes.

Enhanced Border Surveillance: Over 140 AI-based surveillance systems are actively deployed along the Line of Control (LoC) and Line of Actual Control (LAC). These systems use automated cameras, satellite imaging, and radar feeds to instantly detect unauthorized border intrusions and classify hostile targets.

Hypersonic & Missile Tracking: DRDO has developed cutting-edge L-band AI radar technologies specifically engineered to track high-speed, hypersonic missiles.

Autonomous Operations & Human-Machine Teaming: Deploying “silent sentry” AI-guided robots and AI-powered unmanned aerial vehicles (UAVs) enables round-the-clock reconnaissance and border patrolling without risking human lives in harsh environments like the Himalayas.

Network-Centric Warfare: AI acts as a central command system that fuses data from satellites, drones, and command networks, allowing the Army, Navy, and Air Force to coordinate precision strikes with unprecedented synergy.

Artificial intelligence has dramatically accelerated Indian defence capabilities by compressing the “kill chain” (the time between identifying and striking a target) from hours to minutes. By utilizing real-time AI battlefield analytics, the armed forces can make faster data-driven decisions. Key technological and strategic advancements include:

Reduced Kill Chain: AI processes decades of historical data, real-time satellite imagery, and reconnaissance feeds to deliver threat predictions and targeting data with up to 94% accuracy.

Indigenous AI Systems: Companies like Neuralix and Sarvam AI have developed military-grade Large Language Models (LLMs) that run locally without needing external cloud access, ensuring data sovereignty and secure inter-personnel communication.

Autonomous Drones & Surveillance: Over 140 AI-based surveillance systems are deployed along the borders to detect intrusions instantly. Autonomous combat drones use AI-driven target recognition to coordinate strikes with minimal human intervention.

Hypersonic Tracking: The Defence Research and Development Organisation (DRDO) has developed L-band AI radar technology specifically for tracking high-speed hypersonic missiles.

Several Indian companies are actively developing, securing, and hosting cloud systems for the Indian military to support the “Aatmanirbhar Bharat” (self-reliant India) initiative. These solutions range from private tactical clouds to sovereign AI infrastructure.

Notable companies include:

Tata Consultancy Services (TCS): Partnered with C-DAC to build sovereign cloud infrastructure and developed the SovereignSecure Cloud, specifically tailored to support mission-critical, classified operations for the government and public sector.

Esconet Technologies: Pioneered the development of a highly encrypted, private tactical cloud network for the Indian Army to replace decentralized physical servers, enhancing secure data backup and disaster recovery.

Bharat Electronics Limited (BEL): The state-owned defence electronics company partners with various Indian IT firms (like Softvan Limited) to co-develop secure on-premises defence cloud networks and cyber warfare solutions.

Esconet Technologies plays a foundational role in Indian defence AI by providing the mission-critical IT infrastructure, supercomputing hardware, and secure cloud ecosystems required for military intelligence.

Their specific contributions to India’s defence sector include:

Indigenous Supercomputing & AI Hardware: Through their proprietary brand, HexaData, Esconet manufactures high-performance servers and workstations integrated with advanced NVIDIA GPGPU technology. These “supercomputers in a box” provide the immense computational power needed to train complex machine learning algorithms, run simulations, and process real-time battlefield data.

Secure Cloud Infrastructure: Esconet pioneered the development of a highly encrypted private Cloud system for the Indian Army. Serving as a foundational base, this cloud network enables the armed forces to transition from decentralized, physical servers to secure, cloud-based data hosting that can be easily scaled for defence AI applications.

Data Protection & Cybersecurity: Modern warfare relies heavily on secure, AI-powered cyber vaulting and data recovery. Esconet launched advanced cyber vault solutions, such as HexaData ResQ, which ensure that mission-critical defence data remains resilient against ransomware, cyber-attacks, and unexpected operational disruptions.

Through their “Make in India” initiatives, Esconet delivers customized, localized IT and server solutions aimed specifically at government bodies and the defence sector, enabling the armed forces to securely harness the power of AI in modern warfare. 

RDP Technologies: Provides 100% “Make-in-India” AI servers and air-gapped infrastructure specifically designed to function in disconnected/classified military environments.

Sarvam AI & CoRover.ai: These AI deep-tech companies work directly with the Indian Army under the Innovations for Defence Excellence (iDEX) initiative to build on-premises Edge AI and agentic cloud systems for tactical, offline battlefield decisions.

Tata Consultancy Services (TCS) acts as a strategic digital transformation partner for the Indian military and aerospace sector. They drive defence AI by providing secure sovereign cloud infrastructure, real-time threat intelligence, and digital modernization services tailored for the Ministry of Defence and allied agencies.

TCS’s role in shaping and securing the Indian defence AI ecosystem focuses on a few strategic areas:

Sovereign Cloud & Secure Infrastructure: To meet strict domestic data localization and sovereignty mandates, TCS launched the [TCS SovereignSecure Cloud](url) and partnered with the Centre for Development of Advanced Computing (C-DAC). This provides the armed forces and public sector with localized environments to host critical workloads and classified data safely.

AI-Driven Cybersecurity: TCS provides the [TCS Cyber Defence Suite](url), a globally trusted, AI-powered platform for the Indian market. It provides the military and government sectors with 360-degree situational awareness, intelligent threat prediction, and automated responses against complex digital attacks.

Intelligence & Surveillance Support: TCS consults on projects like the Indian Navy’s Tactical Data Link (TDL) and Maritime Domain Awareness (MDA) systems. Their data analytics expertise helps translate raw, multi-source surveillance inputs into actionable, real-time intelligence for forces in the field.

Operational Modernization: TCS actively partners with aerospace and defence agencies to digitize product lifecycles and deploy “Digital Twins” and AI-driven predictive maintenance for military assets. This ensures higher readiness rates and reduces the lifecycle costs of strategic fleets.

Citizen-Centric Defence Services: Beyond direct combat AI, TCS maintains critical systems for the government, including the digital pensions administration network for defence personnel (SPARSH), ensuring the effective delivery of last-mile welfare services. 

By providing safety-critical engineering, deep contextual knowledge, and robust AI governance frameworks, TCS enables the Indian armed forces to safely harness cognitive computing for both operational dominance and data protection.

Bharat Electronics Limited (BEL), a Navratna Public Sector Undertaking, serves as the technology backbone for the Indian armed forces. Its role in defence Artificial Intelligence (AI) includes developing indigenous edge-AI platforms, autonomous systems, command-and-control networks, and smart surveillance capabilities to modernize the country’s defence infrastructure.

Key Pillars of BEL’s Defence AI Strategy

Centre of Excellence for AI: BEL has established a dedicated Centre of Excellence for AI (CoE-AI) to drive indigenous research and development, rapidly prototype military AI solutions, and collaborate with academia and defence startups.

AI-Based Predictive Maintenance: BEL developed Machine Learning (ML) models and edge-vision analytics to monitor the operational health of military hardware. This allows for the real-time prediction of component failures, anomaly flagging, and automated fleet management.

Intelligent Surveillance and C4I Systems: Integrating AI into Command, Control, Communications, Computers, and Intelligence (C4I) networks to enable automated target tracking, threat identification, and intrusion detection. This empowers systems like their Naval Anti-Drone Systems (NADS) with automated soft-kill and hard-kill capabilities.

Autonomous Systems and Robotics: Partnering with the Indian Navy and research agencies to develop emerging AI-driven robotics and unmanned solutions for complex maritime and land operations.

Edge AI Vision Platforms: Implementation of deep learning models at the edge (processing data directly on the sensor/camera) for facial/gesture recognition, forensic video analysis, and automated perimeter security.

Esconet Technologies acts as a crucial IT infrastructure and cloud computing partner for the Indian military. By leveraging high-end GPU servers, they provide the computational hardware and secure storage platforms required to process massive datasets, run predictive analytics, and deploy autonomous AI frameworks on the battlefield.

Key aspects of their involvement include:

Highly Encrypted Cloud Infrastructure: Esconet pioneered the development of secure, private Cloud Systems for the Indian Army. This setup enables centralized IT management, seamless data backups, and robust disaster recovery, replacing decentralized physical servers that are prone to logistical issues.

High-Performance GPU Computing: The company partners with NVIDIA to provide high-performance computing (HPC) and GPU servers. These servers form the backbone for running artificial intelligence, machine learning, and deep learning algorithms.

Make in India Initiative: Esconet manufactures supercomputers, workstations, and high-end servers under its proprietary brand HexaData, equipping defence and aerospace research teams with indigenous hardware.

Mission-Critical Support: They maintain ISO-certified processes to provide proactive IT infrastructure support, ensuring systems remain secure and operational in mission-critical defence environments.

India places immense weightage on maintaining human control and judgement in military AI, strictly adhering to a “human-in-the-loop” doctrine. Rather than allowing machines to make lethal decisions, the Indian armed forces use AI strictly for decision-support, intelligence synthesis, and rapid threat warning.

India’s integration of AI into defence is governed by concrete principles that explicitly prevent machines from overriding human command: 

The ETAI Framework

Launched by the Defence Research and Development Organisation (DRDO), the Evaluating Trustworthy Artificial Intelligence (ETAI) Framework ensures that military AI systems remain reliable, robust, transparent, and resilient to adversarial attacks. It prioritizes human-centric design, preventing blind automation and “automation bias”.

Strategic “Human-in-the-Loop” Mandate

The core consensus within the military command structure is that while AI can inform and predict faster than humans, only humans can bear moral and legal responsibility. Machines are integrated to shorten the “kill chain” (e.g., in radar target identification or drone swarm navigation), but the authorization of lethal force and overarching command judgment firmly remain with commanders.

Sovereign and Secure Algorithms

To protect its judgment systems from being compromised, India is building localized AI architectures. Initiatives like the Indian Army’s Ekam AI platform allow forces to analyze data and process documents securely in sensitive environments without relying on foreign software or external cloud networks.

Governance and Policy

Strategic direction for this balance is governed by the Defence Artificial Intelligence Council (DAIC) and the Defence AI Project Agency (DAIPA). These bodies ensure that AI adoption across the Army, Navy, and Air Force aligns with strict ethical boundaries, keeping combat operations accountable to both constitutional values and international laws on warfare.

Indian laws play a fundamental role in governing AI defence tools by balancing innovation with national security, data privacy, and ethical accountability. While India currently lacks a single, dedicated AI law, the government relies on a techno-legal framework using existing cyber, data privacy, and criminal statutes. 

Primary legislation governing AI defence tools includes:

Information Technology Act, 2000: Serves as the foundational law for addressing cyber offences, intermediary liability, and securing digital networks that power AI defence mechanisms.

Digital Personal Data Protection (DPDP) Act, 2023: Regulates how AI systems collect, process, and store personal data. It requires defence or intelligence tools to implement rigorous data-protection 

standards, even during algorithmic training.

Bharatiya Nyaya Sanhita (BNS), 2023: Addresses malicious AI use (such as deepfakes, impersonation, and AI-enabled cyber fraud) by holding individuals and organizations criminally liable.

India AI Governance Guidelines: These principle-based policies adopt a risk-based approach, ensuring that high-risk autonomous or AI-based defence projects (such as Lethal Autonomous Weapon Systems) undergo strict oversight rather than unrestricted deployment.

Strategic Goals of the Legal Framework

Liability and Accountability: The laws establish strict accountability for developers and operators of AI defence tools, ensuring human oversight over lethal autonomous operations.

Security and Ethical Deployment: Regulations prevent algorithmic bias, secure sensitive military datasets from breaches, and ensure interoperability without compromising operational secrecy.

Sectoral Oversight: Bodies like the Defence Research and Development Organisation (DRDO) and the Ministry of Defence implement additional classified guidelines to align indigenous AI tools with international humanitarian laws and national security protocols.

The evidence from the Gaza war—where algorithmic targeting displaced human scrutiny, leading to massive civilian casualties despite claims of “precision”—is highly relevant to India. India’s military increasingly integrates AI for targeting along its borders, prompting similar debates about efficiency, bias, and compliance with international law. 

India’s Military AI Landscape

Border Surveillance: The Indian Army has deployed over 140 AI-based surveillance systems and “smart fence” networks along the Line of Control (LoC) with Pakistan and the Line of Actual Control (LAC) with China. These analyze video feeds and radar to detect intrusions automatically.

Operational Success Claims: The Ministry of Defence has actively showcased indigenous software for target tracking. For instance, during the cross-border operation—Operation Sindoor—the Indian Army reported using historical and multi-sensor data refined by AI to pinpoint enemy positions and infrastructure. 

Indigenous Development: To avoid reliance on foreign systems, India is developing specialized, localized AI models through platforms like Ekam and collaborations with domestic tech firms like Sarvam AI.

Efficiency vs. Accuracy and Fairness

The cautionary lessons from conflicts like the one in Gaza apply to India due to several systemic factors:

Algorithmic Bias and Training Data: AI targeting models are only as good as the data they are trained on. Models trained on data from specific border disputes or counter-insurgency operations may inherit human biases, misidentifying civilians or non-combatants as threats based on erroneous pattern recognition.

The “Black Box” Problem: Deep learning systems often operate in ways that humans cannot fully explain. If an algorithm flags a target to accelerate the “kill chain,” the human operators validating that strike may not know why the AI selected it, increasing the risk of collateral damage.

Displacing Human Judgment: The Gaza war demonstrated how militaries can fall into the trap of letting systems outpace human capacity to verify targets. While the Indian Army officially mandates that final firing decisions remain with human operators (such as the TAIWS system deployed on the LoC, which requires a human in the bunker to pull the trigger), the operational drive to speed up reaction times naturally erodes the depth of that human verification.

India is acutely aware of these risks and emphasizes “Atmanirbhar” (self-reliance) so that AI models are trained on native datasets that respect Indian operational and ethical contexts. However, the fundamental tension between rapid computational “efficiency” and the legal/moral definitions of “fairness” and “proportionality” in warfare remains an ongoing challenge for all modern militaries using autonomous systems. 

Indian AI systems are not uniformly fair or accurate; performance varies widely depending on the application and dataset. While AI usage is booming for productivity, major gaps remain in algorithmic bias, causing systematic exclusion and discrimination for marginalized communities. 

For official public policies and ethical frameworks, one can review the IndiaAI portal or read the India AI Governance Guidelines published by the Ministry of Electronics and Information Technology (MeitY). 

Here is how the fairness and accuracy of Indian AI systems break down across different sectors:

Welfare and Public Services

Accuracy: Biometric and AI-driven welfare distribution systems often fail systematically when scanning weathered fingerprints, leading to wrongful denial of food rations and essential services.

Fairness: Marginalized communities (e.g., Dalit, Adivasi, and rural demographics) are disproportionately harmed by these algorithmic failures.

Healthcare and Education

Accuracy: AI tools for medical diagnostics (like eye disease screening) can suffer from high false-positive rates, necessitating strict human oversight.

Fairness: AI-driven exam-grading and essay evaluation systems have historically shown biases against students writing in regional Indian languages.

Policing and Surveillance

Accuracy: Facial recognition technologies used in cities perform poorly on darker-skinned individuals and lack disaggregated error reporting.

Fairness: Concerns are consistently raised about civil liberties, false identification, and disproportionate surveillance on minority and working-class populations.

Hiring and Finance

Accuracy: Credit-scoring models can generate unexpected or erroneous rejections due to a lack of complete data histories for lower-income groups.

Fairness: Studies show that many AI recruiting platforms in India harbour biases in favour of prestigious universities or specific caste groups

Why the Discrepancy?

Lack of Diverse Data: Most AI models globally are trained on Western datasets. In India, with its hundreds of dialects, regional contexts, and diverse identities, these systems frequently fail.

Regulatory Gaps: While India has enacted the Digital Personal Data Protection Act, it currently lacks a dedicated regulatory body specifically responsible for policing algorithmic fairness and guaranteeing user redress.

India does not have a policy of using AI to target innocent civilians; its military doctrine emphasizes meaningful human oversight and adherence to International Humanitarian Law (IHL), which mandates a clear distinction between combatants and civilians. 

While the Indian military has integrated AI into its operations to enhance precision and reduce civilian risk, its use remains governed by strict ethical and strategic frameworks: 

Precision Targeting: During Operation Sindoor (2025), the Indian Army utilized AI to identify enemy positions and terror infrastructure. Systems like TRINETRA and Project SANJAY are designed to fuse multi-sensor data to create accurate heat maps of enemy sensors and targets, theoretically reducing the likelihood of accidental civilian casualties.

Human-in-the-Loop Policy: India’s newly unveiled military AI policy (2026) officially backs Lethal Autonomous Weapon Systems (LAWS) and drone swarms but stresses that “meaningful human oversight” must be retained. The official stance is that critical life-or-death decisions must not be subordinated to machines.

Ethical Frameworks: In October 2024, the Indian government launched the Evaluating Trustworthy Artificial Intelligence (ETAI) Framework, which sets guidelines for AI to be reliable, transparent, and resilient. This includes adherence to the “Principle of Distinction” under the Geneva Convention to protect civilian populations.

Operational Roles: Currently, India primarily uses AI for non-lethal and support functions, including:

Border Surveillance: Over 140 AI-based systems (like AGNI-D) monitor the Line of Actual Control (LAC) and borders with Pakistan to detect intrusions.

Logistics & Intelligence: AI is used for weather forecasting (Anuman 2.0), satellite image analysis, and predictive maintenance.

International Advocacy: At global forums, India has argued that existing IHL is sufficient to regulate autonomous weapons, provided that human operators remain accountable for any lethal action.

India addresses bad information, hallucinations, and deepfakes in AI systems through strict amendments to the IT Rules, 2021 and the rollout of comprehensive national AI frameworks. Platforms face stringent takedown timelines, mandatory labelling for synthetic media, and strict legal liabilities for hosting unchecked AI-generated misinformation.

The country combats misleading AI outputs and synthetic media through several specific enforcement pillars and governance structures: 

Mandatory Content Labelling & Watermarking

To prevent users from mistaking synthetic media for reality, the government enforces strict guidelines on AI-generated content.

Watermarking & Metadata: Platforms and creators must embed permanent, non-removable metadata and digital watermarks to identify the source and creator of synthetically generated audio and video.

User Disclosure: Major social media and hosting platforms require users to declare whether the content they share is AI-generated, verified by technical detection tools.

Aggressive Takedown Deadlines (The 3-Hour Rule)

To stop the viral spread of fake news, deepfakes, and manipulated media, India has tightened intermediary guidelines. 

Strict Timeframes: Platforms must remove or block access to flagged content—such as sexually exploitative material, private images shared without consent, and deepfakes—within 3 hours of receiving a grievance or order.

Loss of Safe Harbour: Failure to moderate and remove illegal AI-generated content within this timeframe can result in the loss of legal immunity (“safe harbour”) for intermediaries, exposing them to criminal prosecution

Legal & Criminal Frameworks

Misinformation, impersonation, and fraud perpetrated through AI fall directly under India’s criminal laws and cybercrime framework.

Identity Theft & Cheating: Statutes criminalize the use of AI for identity theft and cheating by personation (e.g., creating voice clones or deepfake videos of individuals to deceive others).

Reporting Mechanisms: Citizens can report cybercrimes, including AI-driven frauds and deepfakes, to the National Cyber Crime Reporting Portal or via the central helpline (1930)

Broad Governance & Safety Frameworks

To foster a trustworthy ecosystem, the Ministry of Electronics and Information Technology (MeitY) and the IndiaAI Mission established the AI Governance Group (AIGEG) and the AI Safety Institute (AISI).

Risk Assessment: These bodies design risk classification systems and standards for safe AI deployment, aiming to mitigate harms like cascading hallucinations or deliberate manipulation.

Principle-Based Governance: Guidelines prioritize the “Seven Sutras” of AI—focusing on safety, reliability, inclusivity, and non-discrimination—ensuring AI innovation is trusted while protecting vulnerable citizens from algorithm failures.

India does not label any specific group of people or entities as “legitimate targets” under AI systems, as this concept violates core AI ethical principles. Instead, India’s national framework under the IndiaAI Mission promotes the principle of “AI for All” with a focus on democratization, inclusivity, and public good.

India’s AI governance and IT guidelines—such as the Digital Personal Data Protection Act (DPDP) 2023 and recent amendments to the IT Rules regarding synthetic content—focus on safeguarding citizens rather than targeting them.

Instead of targeting individuals, Indian regulations require AI platforms to label AI-generated content (deepfakes, synthetic media) to protect the general public from misinformation.

Key Safeguards and Directives:

“AI for All”: The core philosophy of India’s AI vision is to deploy artificial intelligence for inclusive growth, focusing on sectors like agriculture, healthcare, education, and climate action.

Content Labelling Mandates: Platforms are required to actively detect, label, and remove unlawful AI-generated or synthetically altered content. This protects users from deception rather than treating them as targets.

Risk Assessments: Developers and platforms are expected to conduct risk and bias assessments to ensure systems do not harm or unfairly discriminate against vulnerable communities. 

Indian AI defence systems—particularly those developed by the Defence Research and Development Organisation (DRDO) and deployed by the Indian Armed Forces—extensively utilize interconnected “associational” and network-centric models.

Rather than relying on completely isolated, single-node algorithms, India’s military AI primarily functions through highly integrated, interoperable, and communicative ecosystems. These work as follows:

Command-and-Control Integration: AI nodes are embedded within integrated networks like the Integrated Air Command and Control System (IACCS). These associational networks allow real-time sensor data from satellites, drones, and ground radar to cross-reference and paint a unified battlefield picture.

Cybersecurity & Threat Detection: The Defence Cyber Agency (DCA) uses AI tools to continuously analyze and detect anomalies across military networks. These systems associate normal traffic patterns with potential foreign intrusions to protect critical infrastructure.

Autonomous Swarms: India’s offensive swarm drones and loitering munitions use cooperative associational AI. Individual drones communicate with one another to share targeting data, divide combat roles autonomously, and adjust their flight paths based on the actions of adjacent drones.

Predictive Analysis: Advanced AI platforms analyze associations across historical border incursions, troop movements, and local terrain parameters to predict hostile activities and assist commanders in rapid decision-making.

While these interconnected systems greatly enhance situational awareness and response speed, the Ministry of Defence maintains a strict “Human-in-the-Loop” doctrine for combat operations, ensuring that the final authorization for lethal force remains with human personnel rather than fully autonomous associational systems.

Indian defence is transitioning from probabilistic warfare (relying on chance and numerical superiority) to definitive outcomes by leveraging AI-driven precision targeting, autonomous swarm warfare, and integrated theatre commands. This ensures absolute, assured overmatch against threats.

The push to replace probability with guaranteed, decisive outcomes is driven by the following strategic and technological overhauls: 

Assured Precision and Deterrence

Multi-Domain Dominance: India’s capability surge, featuring successful tests of Hypersonic platforms, MIRVs (Multiple Independently Targetable Re-entry Vehicles), and indigenous systems like the TARA range extension kits, guarantees precise, “zero-error” target neutralization from safe distances.

Integrated Air Defence: India aims to field indigenous long-range air defence with a single-shot kill probability exceeding 80-90%.

Autonomous and AI-Driven Warfare

Intelligent Battlefields: Through the Department of Defence Production (DDP) initiatives, India is deploying Artificial Intelligence (AI) for deep combat data management and autonomous weapon systems. This eliminates the margin of human error and probabilistic outcomes in surveillance and targeting.

Swarm and Drone Units: The raising of specialized lethal strike drone batteries (e.g., Divyastra) and specialized units like the Bhairav Battalion ensures that responses to adversary infiltrations are rapid, targeted, and conclusive.

Structural Reorganization (Theaterization)

The Ministry of Defence (MoD) has focused extensively on structural military reforms, establishing Integrated Theatre Commands. This “jointness” eliminates traditional inter-service turf wars, unifying land, air, and sea assets to guarantee a synchronized and rapid operational response. 

Self-Reliance and Continuous Modernization

Backed by record capital outlays (e.g., ₹6.81 Lakh Crore budgets), India has enforced Positive Indigenisation Lists to ensure the military is equipped with home-grown technology shielded from foreign supply-chain vulnerabilities. This drives rapid deployment of indigenous missiles and munitions directly to the field.

These integrated measures—moving from relying on conventional combat estimates to a guaranteed techno-centric dominance—are a core pillar of the Defence Forces Vision 2047.

There is no single official national aggregate for misidentifications by Indian AI systems. However, independent research and civil society audits show high error rates in specific domains:

Facial Recognition Technology (FRT): A study on Indian faces found that algorithms incorrectly identified gender up to 7% of the time for women, compared to less than 0.5% for men. In law enforcement trials, Delhi Police’s FRT registered system accuracy rates ranging from 1% to 2%.

Marginalized Communities: Research highlights systematic biases, with false match and misclassification rates clustering disproportionately around Dalit, Adivasi, and Muslim populations.

Accent Recognition: AI tools deployed for dialect and identity policing have shown error rates as high as 60%.

To review how civil society organizations are tracking and reporting these issues, one can refer to the findings from the Internet Freedom Foundation or the analysis on structural algorithmic bias published by the Criminal Justice and Police Accountability Project.

If a small percentage of errors accumulate in India’s AI systems, it can trigger a cascading “snowball effect.” Minor inaccuracies in automated decision-making feed into downstream processes, leading to systemic failures. This erodes public trust, risks financial losses, and impacts governance.

The specific risks of these accumulated errors spreading across India’s localized sectors include:

Compounding Systemic Effects

Model Drift and Degradation: AI relies on input from past decisions. If a model generates slightly incorrect data and that data is fed back into its training loop (model drift), the initial tiny error will magnify over time, completely degrading output reliability.

Workflow Bottlenecks: In automated supply chains or fintech platforms, undetected algorithmic errors cause cascading disruptions, amplifying inefficiencies and leading to widespread service delays. 

Deepening Bias and Inequality

Dataset Imbalances: India’s immense linguistic and cultural diversity can lead to dataset biases. If minority dialects, rural dialects, or specific socio-economic nuances are underrepresented, the AI will make slightly skewed predictions. Over time, this inherently discriminates against marginalized groups.

Exclusionary Policies: In credit scoring, loan approvals, or welfare benefit distribution, accumulated misclassifications might unfairly disqualify legitimate citizens from government schemes or banking services

Economic and Financial Spillovers

Financial Miscalculations: In India’s booming digital economy, minor percentage errors in algorithmic trading, crop yield forecasting, or logistics routing will add up to large financial losses.

Erosion of Brand Trust: E-commerce, digital payments, and local service platforms risk alienating consumers if AI-driven customer service or fraud-detection systems repeatedly fail, causing users to abandon the platforms.

Critical Services and Infrastructure Vulnerability

Healthcare Misdiagnosis: In telemedicine, minor misclassification of medical data could lead to incorrect triage, endangering patients and forcing a reliance on constant manual overrides.

Infrastructure Operations: In localized smart city grids or agricultural automation, minor errors can result in wasted resources, over-engineered safety mechanisms, or disrupted public services.

The Mitigation Strategy

To prevent this, tech leaders and the government emphasize strict adherence to ethical AI guidelines, continuous human validation, and zero-tolerance data governance frameworks. Implementing strict feedback-loop checks ensures that “drift” is detected and corrected before it compounds into a large-scale failure. 

An investigation by Israeli media outlets +972 Magazine and Local Call describes the use of “power targets,” buildings that are not strictly military in nature but are considered relevant by Israel for exerting pressure. AI systems can identify such targets, but their inclusion reflects a strategic choice rather than technical necessity. 

In Indian defence, “power targets” typically refers to identifying and neutralizing an adversary’s critical energy, utility, and electrical grids using precision-guided and cyber-electronic warfare. Disrupting power assets aims to cripple an enemy’s command centres, communications, and logistical supply chains without engaging in traditional, destructive frontline combat. 

India is heavily focusing on this concept through its modernization IndiaAI Mission and broader defence transformation Department of Defence Production strategies. 

Specific operational definitions and applications include:

Directed Energy Weapons (DEWs): India’s Defence Research and Development Organisation (DRDO) is developing high-power laser systems, like the 100-kilowatt class Dura-2. These weapons are used to rapidly melt steel, intercept incoming drones, and fry the electronic components and sensitive computer boards in enemy cruise missiles at a fraction of the cost of traditional interceptor missiles.

Grid Interdiction & Cyber-Electronic Warfare: Military operations increasingly use AI and machine learning to map and target hostile power infrastructure. By analyzing vast datasets in real time, AI can identify vulnerabilities in an enemy’s electrical grid, causing strategic blackouts that disable their logistical and communication networks.

The “Kill Chain” Compression: AI models—which are being utilized to rapidly reduce the time it takes to identify and strike targets—are now a core focus for the armed forces. By processing intelligence in real time, command networks can quickly designate specific enemy power sources or command and control (C2) nodes for automated strikes.

Hardening India’s Own Infrastructure: The flip side of offensive power targets is defence. Indian defence strategies use AI proactively to detect anomalies in domestic power grids and military bases, protecting critical national infrastructure from hostile cyberattacks or state-sponsored infiltration.

For India’s defence systems, data and human intelligence are the foundational pillars of sovereign AI. They transform raw battlefield information into actionable insights and strategic decisions. Key defence policy and strategic priorities are available through the Department of Defence Production and the Observer Research Foundation.

Data and human intelligence hold distinct, vital roles in India’s military AI framework:

The Importance of Data

Sovereign Big Data: Defence datasets (satellite imagery, electronic intelligence, and border surveillance) are highly sensitive. Using sovereign, indigenous data keeps these systems secure from foreign vulnerabilities.

Real-time C4ISR: AI relies on massive data ingestion to modernize Command, Control, Communications, Computers, Intelligence, Surveillance, and Reconnaissance (C4ISR) systems. Big Data analytics quickly identify hidden patterns, track naval movements, and detect anomalies in critical infrastructure.

Terrain Adaptability: Processing historical and real-time topographical data is crucial for operating in challenging environments like the Himalayas or the Indian Ocean without putting soldiers at risk.

The Necessity of Human Intelligence (HUMINT & Human Oversight)

Contextual Nuance: While data trains the machine, human intelligence provides the strategic context and intuition necessary to interpret enemy intent.

Ethical Guardrails & Accountability: Human oversight is essential in autonomous weapon systems and rules of engagement. Human commanders provide the moral and legal judgment required during conflicts.

Bridging the Tech-Field Gap: Continuous engagement between scientists (DRDO), policymakers, and war fighters ensure AI tools address real-world operational needs rather than just theoretical capabilities.

India’s defence strategy heavily relies on combining artificial intelligence (AI) with real-time data analysis to modernize its armed forces. The military utilizes AI for command, control, communication, intelligence, surveillance, and reconnaissance (C2ISR) by fusing vast amounts of data.

The convergence of data analysis and defence technologies is foundational to India’s military strategy for several key reasons: 

Real-Time Border & Maritime Security: The military aggregates data from edge sensors, satellites, and the Maritime Information Management and Analysis Centre (IMAC) to process localized activity. AI-powered systems can then identify ship movements or autonomous intrusions without constant human intervention.

Sovereign Dataset Protection: India’s defence framework emphasizes sovereign AI processing. Because national security applications rely on highly sensitive datasets, the government focuses on localized, indigenous data processing rather than relying on foreign platforms.

Operational Effectiveness: The Defence Research and Development Organisation (DRDO) and the Defence AI Council (DAIPA) are systematically shifting strategies to prioritize predictive analytics, autonomous combat systems, and intelligent threat detection to gain strategic advantages on modern digital battlefields.