Stay up to date with free updates
Just sign up artificial intelligence myFT Digest – delivered straight to your inbox.
Global tech companies and local startups are looking to unlock lucrative new markets in India, leveraging artificial intelligence platforms that adapt to a wide range of languages and industries in the world's most populous country.
Microsoft, Google and Silicon Valley-backed startups such as Sarvam AI and Kultrim, founded by Indian mobility group Ola's Bhavish Agarwal, are all working on developing AI voice assistants and chatbots that work in languages such as Hindi and Tamil.These tools are targeting fast-growing Indian industries such as the country's large customer service and call center sectors.
India has 22 official languages, with Hindi being the most widely spoken, but researchers estimate there are thousands of other languages and dialects spoken by its 1.4 billion people. Google on Tuesday released its AI assistant, Gemini, in nine Indian languages.
Microsoft's Copilot AI assistant is available in 12 Indian languages, and the company is working on other projects tailored for India, such as building “tiny” language models at its research center in Bengaluru. These smaller alternatives to the expensive large language models that underpin generative AI can run on smartphones instead of the cloud, making them cheaper and potentially better suited for a country like India where connectivity may be limited.
Microsoft said: [AI] “We want it to be simple, easy to use and available to all our customers and partners,” Puneet Chandhok, Microsoft's president for India and South Asia, told the Financial Times, adding that this required “contextualizing it for the Indian context, making it more relevant and more accurate.”
Microsoft is also partnering with Sarvam AI, a Bengaluru-based company that was founded just last year to develop “full-stack” generative AI tools for Indian businesses. The startup has raised $41 million from investors including Peak XV, a former India arm of Sequoia, and Menlo Park-based Lightspeed Venture Partners.
Hemant Mohapatra, a partner at Lightspeed, said investing in local AI companies is becoming increasingly important as governments aim to develop “sovereign AI” that is trained and stored within their own countries.
“The AI supply chain is starting to fragment,” Mohapatra said. “If you're going to train a foundational model in India using Indian national data, voice, video, text, different languages, it has to be an Indian company, it has to be focused on Indian use cases, it has to be Indian-based, have Indian founders, etc.”
India's AI race doesn't include building an LLM from scratch to compete with leaders like Open AI, which investors argue would require too many resources and capital to make sense.
Instead, companies such as Sarvam AI are focusing on adapting existing LLMs to Indian languages and using voice data instead of text, making them more effective in a country where many people prefer to communicate through voice messages over written text.
“In a complex country like India, there's still a big gap between these foundational models and real-world use cases,” said Bejul Somaiah, partner at Lightspeed. “In a market like India, you're going to need a bit of an ecosystem to enable companies to leverage the capabilities of the foundational models.”
Tanuja Ganu, manager of Microsoft's research lab in Bengaluru, said an added benefit of testing new technologies and tools in a country of India's size and diversity is that they can be exported elsewhere.
“We are using India as a testbed to validate some of the Indian technology and see how we can scale it to other parts of the world,” she said.
