With India’s AI market projected to hit $126 billion by 2030, tech giant Google has launched a new Market Access Program to help India’s AI startups from seed to scale to go global.
Alongside this, Google has introduced new updates to its specialized models like Medgemma 1.5 and FunctionGemma, reinforcing its push to make advanced AI tools more accessible.
“If you solve for India, you build for the world. Our focus now is accelerating how quickly Indian startups can scale, reach global markets, and deliver outcomes,” said Preeti Lobana, VP and Country Manager, India, Google, announcing the new program at the Google AI Startups Conclave in New Delhi Thursday.
“India’s AI startup story has entered a decisive new phase. We are moving from prototypes to products, and from early traction to sustainable businesses,” she noted in a Google blog post.
Google’s full-stack support to startups across the lifecycle, from capability building to real-world deployment and scale, focuses on removing challenges at every critical stage.
READ: Apple and Google forge unlikely alliance on AI: Siri to use Gemini, not ChatGPT (January 13, 2026)
The new program is a purpose-built engine for AI-first startups ready to scale responsibly, focused on three outcomes: enterprise readiness through global selling expertise, access to Google’s enterprise network, and global immersion in key international markets.
To ensure these startups have the “horsepower” to scale, Google also spotlighted its upcoming Global AI Hub in Visakhapatnam. This 1-gigawatt facility, powered by green energy, will provide the high-performance computing resources needed for startups in the program to refine their models at a global scale.
To further fuel deep-tech innovation by Indian startups, Google also announced new additions to its open Gemma model family, focused on two areas seeing rapid adoption in India: population-scale healthcare AI and action-oriented, on-device agents.
MedGemma 1.5 builds on the momentum of Google’s health-focused AI work. The new open 4-billion parameter model enables developers to build applications that support complex medical imaging workflows.
The release of MedGemma 1.5 follows Google’s recent collaboration with India’s flagship medical institute, All India Institute of Medical Sciences (AIIMS), which is leveraging MedGemma to build India’s Health Foundation Models, contributing to India’s Digital Public Infrastructure and the outcomes being made available to the ecosystem.
To support strong momentum toward agent-based systems, Google has introduced FunctionGemma, a specialized version of the Gemma 3 270M model tuned for function calling.
FunctionGemma helps translate natural language commands into executable actions, allowing startups to build on-device, low-latency applications with automated workflows.
The model supports the cost-effective and lightning-fast development of mobile solutions that respect user privacy and can work seamlessly even on low-end devices without a constant internet connection.
Together, these models expand the set of production-ready building blocks available to Indian founders, helping them move from experimentation to deployment across healthcare, enterprise, and consumer applications at scale, Lobana wrote.
READ: ‘AI will change every job’: Walmart, Google to launch new AI experience (January 13, 2026)
FunctionGemma is supported by popular tools across the entire workflow, including Hugging Face Transformers, Unsloth, Keras or NVIDIA NeMo, and can be deployed using LiteRT-LM, vLLM, MLX, Llama.cpp, Ollama, Vertex AI or LM Studio.
Alongside the Conclave, Inc42 released the “Bharat AI Startups Report 2026,” supported by Google. The findings confirm a decisive shift in the ecosystem.
India’s AI market is projected to hit $126 billion by 2030. Crucially, 47%of enterprises are already moving pilots into production, according to the report.
The innovation cost is falling, the report noted. Historically, high compute costs acted as a barrier on Indian ambition. With public rails lowering entry barriers, founder capital is shifting from infrastructure cost to funding product innovation.
India’s complexity—22 languages, patchy connectivity, and price sensitivity— is often viewed as a challenge – as the global ‘Stress Test.’ The report reframes it as India’s biggest asset. If an AI agent works reliably for a rural user in India, it is robust enough for the world. “Bharat-tested” is becoming the new gold standard for resilience.
The competitive edge has shifted to trust-by-design. Startups that embed safety, privacy, and security from day one, aren’t just “compliant”, they are the only ones winning the long-term enterprise contracts, according to the report.
Ultimately, the AI era will be defined by outcomes. This is seen in startups such as Cloudphysician reducing ICU mortality rates by 40% and Rocket Learning personalizing education for millions, among many others.
“By stitching together skilling, capital, infrastructure, and market access, we are clearing the path for founders,” Lobana wrote. “As we look to the AI Impact Summit in February, the signal is clear: The future of AI isn’t just being used in India; it is being built here.”