What Exactly Was Announced?
OpenAI has enabled both local data storage and local 'inference' for its API in India. 'Inference' is the technical term for the computational process where an AI model, like GPT-6 Luna, generates a response to a prompt. This means for eligible customers
using supported services, the entire lifecycle of an API request—from the initial prompt to the final AI-generated text—can now happen on servers located within India. Previously, OpenAI offered data residency for storage, but processing might have occurred elsewhere. This new development ensures both storage and processing stay in-country, a critical distinction for many organisations. The service initially supports the GPT-6 Luna model via the Responses and Chat Completions APIs.
Why Local Data Processing Matters
The primary driver behind this launch is data localisation. Many Indian businesses, especially in highly regulated sectors like financial services and healthcare, have been cautious about adopting AI for core operations. This is often due to regulatory mandates, like those from the Reserve Bank of India, or strict internal policies that require sensitive customer information to be processed and stored within India's borders. By offering in-country processing, OpenAI is removing a major barrier to adoption. As OpenAI India's Head of Enterprise, Nitin Bawankule, noted, this gives businesses more control over where their data runs and helps them integrate AI into important workflows like document summarisation and internal data analysis.
The Benefits for Indian Businesses
Beyond compliance, hosting the API locally brings tangible performance benefits. Processing data closer to the end-user reduces latency, meaning faster response times for applications built on the API. For developers and startups, this translates to a snappier, more reliable user experience. This move is part of OpenAI's broader 'OpenAI for India' initiative, signalling a deep commitment to the market. This initiative also includes a partnership with Tata Group to build out domestic AI data centre capacity, further strengthening the local infrastructure. The availability of local processing on other platforms, like Amazon Web Services' Bedrock, shows a clear industry trend toward meeting India's specific market needs.
Who is Eligible and How to Access It
This feature is not automatically enabled for all users; it is an enterprise-focused capability aimed at organisations with specific data-handling needs. According to OpenAI's documentation, eligibility is currently tied to organisations that have 'Zero Data Retention' (ZDR) controls active. This is a setting that prevents OpenAI from retaining any data from API requests. Accessing the service requires creating a new API project, specifically selecting India for both storage and processing, and then using a designated regional endpoint. Businesses that need this capability should review OpenAI's data control policies and contact their sales or account team to confirm eligibility and get started.
The Bigger Picture: Data Sovereignty
OpenAI's launch fits into a larger global conversation around data sovereignty and the legal framework governing digital information. While India's Digital Personal Data Protection (DPDP) Act does not universally mandate that all data must stay in India, it establishes a comprehensive framework for how personal data is handled and gives the government power to restrict transfers. Many enterprise contracts and sector-specific regulations already enforce strict data residency as a condition of business. By investing in local infrastructure, tech giants like OpenAI are acknowledging that to succeed in a key market like India, they must align their services with the country's regulatory landscape and the practical needs of its enterprises.
















