India Weighs ₹20,000 Cr Anchor Fund for Frontier AI and Compute

The Indian government is considering an anchor investment of ₹15,000–20,000 crore in a proposed National Frontier AI & Compute Fund aimed at financing advanced AI companies and the infrastructure needed to build and run large models. The proposal remains under consultation, and its final corpus, structure, governance and government contribution have not yet been decided.

The proposed fund could provide long-term capital for frontier-model developers, specialised AI companies, GPU clusters and dedicated AI data centres. One structure under discussion would combine government and private capital on a 50:50 basis, potentially through a SEBI-regulated Category I alternative investment fund. Officials are also examining venture and growth equity, infrastructure investments, fund-of-funds commitments and direct co-investments.

A major question is how to finance GPUs without building expensive capacity that remains underused. Discussions have therefore included demand-linked financing, where AI companies and research institutions could commit to capacity or utilisation before infrastructure receives funding. That could give GPU and specialised cloud providers greater certainty while reducing the risk of stranded compute assets.

The proposal would complement the existing ₹10,371.92 crore IndiaAI Mission, which already subsidises access to computing infrastructure and supports indigenous AI models. Government data says the shared national compute pool had expanded to more than 45,000 GPUs by June 2026, while 237 projects had consumed 93.18 lakh GPU hours through subsidised access by August.

The proposed fund therefore tackles a different part of the AI financing problem: not simply making existing compute cheaper, but mobilising patient capital to build additional infrastructure and finance companies developing capital-intensive AI technologies. Its impact will ultimately depend on the final governance model, private-investor participation and whether funded GPU capacity is matched with sustainable demand. For now, the ₹15,000–20,000 crore figure should be viewed as a proposal under discussion rather than an approved government allocation.