Centre invites proposals from startups, others to develop foundational AI models: Vaishnaw

New Delhi, Jan 30 (FN Agency) The government has invited proposals from startups, IT service providers, and engineering researchers to develop foundational AI models under the India AI mission, Union Minister for Electronics and IT Ashwini Vaishnaw announced on Thursday. “All proposals submitted by February 15 will be evaluated by a technical and legal committee, following a rigorous selection process similar to that used in the semiconductor program,” Vaishnaw told reporters. The India AI mission, approved by Prime Minister Narendra Modi in March 2024, has made rapid progress, with its first major milestone—the common compute facility—becoming operational in August 2024.

“The impanelment process, which started last August and concluded last week, has yielded excellent results. Against an initial target of 10,000 GPUs, we have deployed 18,693 high-end GPUs, including 15,000 very high-end GPUs,” he said. The mission is structured around four pillars: compute infrastructure, foundational model development, application deployment, and AI safety. As part of the applications segment, 18 AI-driven solutions have been approved in the fields of agriculture, learning disabilities, and climate change, with more in the pipeline. Additionally, the AI Safety Institute, designed as a virtual open school, has been tasked with ensuring algorithmic integrity. “Eight new AI safety projects have been approved, and further proposals are now open for submission,” the Minister said. He emphasized that the initiative reflects the Prime Minister’s commitment to democratizing AI technology and fostering innovation for large-scale problem-solving.