Sarvam signs defence pact to build AI for Army, Navy and Air Force
By Anika SharmaAI & Innovations Editor, Peepals Tech
Representative image: geralt / Pixabay
Sarvam signed an MoU with the Defence Ministry's HQ IDS, announced October 27, 2026, to build AI tools for the Army, Navy and Air Force on Indian infrastructure.
What Happened
Bengaluru-based artificial intelligence company Sarvam has signed a memorandum of understanding (MoU) with Headquarters Integrated Defence Staff (HQ IDS), the joint planning body that works under the Ministry of Defence, to build and run AI applications for the Army, the Navy and the Air Force. The agreement was announced on Tuesday, October 27, 2026. Under it, Sarvam will design, develop and deploy AI tools for operational, training and decision-support needs across the three services and the joint organisations that sit under HQ IDS, and will provide technical support once those tools are in use. The work will be done by Chanakya, the company's separate, firewalled division for national security and defence.
Key Facts
- Parties: Sarvam and Headquarters Integrated Defence Staff, which functions under the Ministry of Defence.
- Announced: Tuesday, October 27, 2026. One report says HQ IDS first posted about the pact on its social media account a day earlier.
- Term: reported as a five-year umbrella agreement covering several AI projects rather than a single contract.
- Scope: AI applications for operations, training and decision support, plus follow-on technical support.
- Shared tools: the three services and joint bodies are to get access to the same foundation models, agentic infrastructure, and geospatial and perception systems.
- Delivery: through Chanakya, a firewalled Sarvam division focused on defence, with deployment on sovereign infrastructure controlled and run within Indian jurisdiction.
- No contract value has been disclosed.
Why It Matters
The pact is one of the clearest signs so far that India's defence establishment wants its military AI to be built at home. A foundation model is a large AI system trained on huge amounts of data that can then be adapted to many tasks, from summarising reports to reading maps. Agentic infrastructure refers to software that lets AI systems carry out multi-step tasks on their own within set limits. Geospatial and perception systems help machines interpret satellite images, terrain data and sensor feeds. By routing all of this through one umbrella agreement, the services can share the same base technology instead of buying separate, incompatible tools for each branch.
The phrase "sovereign infrastructure" matters as much as the software. It means the models and the data they handle stay on computers controlled within India, rather than on servers abroad run by foreign cloud companies. For military data, where leaks or outside control could be serious, this is a central concern. Reports say Chanakya will draw on Sarvam's own AI stack, including its 105-billion-parameter foundation model. Sarvam introduced its 30-billion and 105-billion-parameter models in February 2026, around the India AI Impact Summit, and it was selected in April 2025 to build a foundational model under the government's IndiaAI Mission.
Reports on the pact list possible military uses such as decision support, target acquisition and the fusing of data from many sensors so that commanders get one picture. One analysis stresses that the MoU is a framework for projects and not a move to fully autonomous combat systems, and that human oversight is to be kept over operational decisions. Those details will matter as projects take shape, because the use of AI in targeting is one of the most debated questions in military technology worldwide.
| Item | What the MoU provides |
|---|---|
| Who builds | Sarvam, through its defence division Chanakya |
| Who uses | Army, Navy, Air Force and joint bodies under HQ IDS |
| What is shared | Foundation models, agentic infrastructure, geospatial and perception systems |
| Where it runs | Sovereign infrastructure within Indian jurisdiction |
Impact
Short-term: Specific projects under the umbrella still have to be defined and approved. The first visible results are likely to be tools for training and staff work, such as analysing documents and data, before anything reaches front-line operations.
Long-term: If the arrangement works, it could set a model for how India's armed forces adopt AI: shared Indian-built models, one framework across all three services, and data kept within the country. It also gives an Indian AI company a long-term anchor customer in defence, which can help fund further model development.
Who is affected: Personnel of the Army, Navy and Air Force and joint commands, defence planners at HQ IDS, Sarvam and its engineers, other Indian AI start-ups that want defence work, and foreign technology suppliers who may face more competition from home-grown systems.
Key Takeaway
India's armed forces have chosen an Indian AI company, working through a firewalled defence unit on infrastructure kept within the country, to build shared AI tools for all three services under one framework.
Questions and Answers
What did Sarvam and HQ IDS agree on October 27, 2026?
They signed a memorandum of understanding under which Sarvam will design, develop and deploy AI applications for operations, training and decision support across the Army, Navy, Air Force and joint organisations, and will support those applications afterwards.
What is HQ IDS?
Headquarters Integrated Defence Staff is the joint body under the Ministry of Defence that works on planning and coordination across the three services. Joint and tri-service organisations come under it, which is why one agreement with it can cover all three forces.
What is Chanakya?
Chanakya is Sarvam's separate, firewalled division for national security and defence work. Keeping it apart from the company's commercial business is meant to protect sensitive military data and projects.
Will the AI make combat decisions on its own?
Nothing announced says so. Reported uses include decision support and combining sensor data, and one analysis notes that human oversight is to be kept over operational decisions. The MoU is a framework, and individual projects have not been described in detail.
Disclaimer: Prepared by the Peepals newsroom from publicly available sources with AI assistance. Information is accurate to the best of our knowledge at the time of publication and may change. Images may be representative. Not professional advice. Report an error via our contact page.
By Anika Sharma, AI & Innovations EditorCovers AI & Innovation and more for Peepals Tech. All stories by Anika Sharma
Get the day's stories in your inbox
Peepals Weekly: the week decoded, free. One email a week, unsubscribe in one click.
Anika Sharma is a pen name of the Peepals Global Editorial Team, which researches, fact-checks and publishes every story. The portrait is an illustration.
