New Delhi, July 31 — IBM and homegrown AI startup Sarvam have partnered to accelerate the development and adoption of sovereign artificial intelligence (AI) technologies in India, with a focus on government agencies, public sector organisations and regulated enterprises, according to a statement on Friday.
Under the partnership, the two companies will jointly demonstrate and pilot sovereign AI technologies for use cases such as citizen services, grievance redressal, document processing and administrative workflows.
The collaboration combines IBM Sovereign Core, the company's sovereign-by-design AI software platform, with Sarvam's India-first sovereign AI stack, which includes reasoning models and multilingual language and voice AI developed and trained in India.
The combined offering is designed to help organisations deploy AI while maintaining greater control over data, governance, security and compliance in line with India's regulatory and operational requirements.
In addition, the initiative aims to accelerate sovereign AI adoption through innovation pilots, solution accelerators, technical advisory services and knowledge-sharing programmes.
The IBM GovTech AI Innovation Center in Lucknow will serve as a joint incubation and demonstration hub where government departments, public sector organisations and enterprises can evaluate practical sovereign AI applications and address technical, operational and governance requirements before scaling deployments.
"Sovereign AI is not simply about where AI runs. It is about giving organisations control over how AI is governed, deployed and operated," said Sriram Raghavan, General Manager, IBM Software, India and Software Innovation Lab.
He said IBM Sovereign Core provides an enterprise-grade platform designed to help governments and regulated enterprises scale AI while addressing governance, security and compliance requirements.
Pratyush Kumar, Co-Founder of Sarvam, said sovereign AI must work within the systems governments and enterprises already rely on while supporting large-scale operations.
"Our stack puts models, voice and language technologies on top of it, so a citizen can access a benefit or resolve a grievance in their own language, on a phone call," Kumar said.