Google used I/O Connect India 2026 in Bengaluru to widen its India AI programme across enterprise deployment, education, healthcare, cybersecurity and Indic-language access, moving beyond model announcements into infrastructure and adoption pathways for developers, companies and public-sector organisations.
The most significant enterprise update is the availability of Gemini on Google Distributed Cloud from within Indian data centres. The setup is designed for regulated industries and public-sector organisations that need generative AI capabilities while keeping prompts, model weights and outputs inside their own perimeter. Google also said Gemini 3.5 Flash will be available to Indian enterprises and startups through its Gemini Enterprise Agent Platform and Gemini Enterprise app with strict in-country machine-learning processing commitments.
The education track includes Google DeepMind’s AI Research Foundations curriculum, a free 56-hour programme intended to help learners build and fine-tune large language models. IISc Bengaluru has begun incorporating the curriculum, while NASSCOM will make it available through FutureSkills Prime. Google said the programme has crossed 38,000 global enrolments. It is also launching ATL Saathi, a Gemini-powered assistant for teachers in Atal Tinkering Labs. The tool will start with 100 schools this year and is intended to scale toward 10,000 schools.
Healthcare was another major pillar. AIIMS researchers are expanding work with Google’s MedGemma open models to build India-specific models for leprosy and sexual and reproductive health. These models are expected to use image and text inputs and be made available to the Indian developer ecosystem. The company also said the National Health Authority recently used Gemma 4 and Google’s open-sourced Medical Data Toolkit to build Aarogya Setu 2.0.
On language access, Gemini Live is expanding to more than 25 Indian languages and dialects, including Sanskrit, Bhojpuri and Maithili. Google also highlighted Project Vaani, its collaboration with IISc Bengaluru, which has open-sourced speech and image datasets covering 109 Indic languages.
Google’s India AI push is being framed around enterprise-grade deployment and local controls rather than only consumer tools. The company is trying to combine cloud sovereignty, developer training, school-level experimentation, public-health models and Indic-language support into one adoption stack for the Indian market.
