Snowflake brought together customers, partners and technology leaders at Snowflake World Tour Mumbai 2026 to discuss how organisations are moving from AI experimentation towards enterprise-wide deployment, with greater emphasis on trust, governance, data readiness and cost control.
The event followed Snowflake’s recent announcement of Dynamic Model Routing in Cortex AI Gateway, which is aimed at helping organisations optimise model selection, manage AI costs and improve the value generated from AI investments.
From AI experimentation to AI economics
The event focused on the foundations required to take enterprise AI beyond pilots and proofs of concept. Vijayant Rai, Managing Director – India, Snowflake, highlighted the importance of enterprise data quality, context, accessibility and governance in determining the effectiveness of AI deployments.
“AI is everywhere, but context isn’t. Context is what makes AI dependable, relevant, and capable of delivering business value at scale,” Rai said.
While organisations have spent the past year experimenting with models and developing AI proofs of concept, the focus is increasingly shifting towards deploying AI responsibly and economically at scale. Model costs, token consumption, governance and operational complexity are emerging as considerations alongside model performance.
Snowflake said the next phase of enterprise AI adoption will require organisations to balance innovation with trust, control and cost efficiency.
Building the control plane for the Agentic Enterprise
Snowflake also outlined its vision of becoming the control plane for the agentic enterprise, providing organisations with the ability to govern how AI agents interact with enterprise data, applications and models.
A key component is enterprise data and context. Snowflake highlighted Horizon Context, which brings business meaning and semantic understanding into enterprise data. The objective is to provide AI systems with trusted data and shared business context to support more consistent outcomes across teams and applications.
The company also emphasised model choice. Its model-agnostic approach allows customers to use frontier as well as open-source models while maintaining governance and flexibility. Rather than relying on the largest model for every workload, organisations can match models to specific tasks to balance performance and cost.
The third element is software and applications. Snowflake showcased capabilities that allow organisations to connect AI with business applications and workflows without moving or duplicating data. The company also highlighted Snowflake CoCo and CoWork, which are designed to help developers and business users build and interact with AI-powered applications within a governed environment.
Cortex AI Gateway was also highlighted as part of this control-plane approach. It provides visibility, governance and cost controls across AI agents, models and workloads, allowing organisations to monitor AI usage, manage access to data and applications and control AI-related spending as adoption expands.
Recognising India’s data leaders
The Mumbai event also featured the Snowflake Data Driver Awards, recognising organisations and technology leaders using data and AI to drive business outcomes.
Jubilant FoodWorks received the Data Driver of the Year award, while SMFG Credit was recognised for innovation in data analytics. Nishant Pradhan, Chief AI Officer at Mirae Asset Investment Managers (India), was named Data Executive of the Year.
The focus shifts to AI ROI
Rai concluded that the next phase of enterprise AI will depend on more than model sophistication. Organisations will need to combine trusted data, business context, model flexibility, governance and cost efficiency as AI deployments scale.
With enterprises across India increasing their AI investments, the discussions at Snowflake World Tour Mumbai 2026 highlighted a shift from AI experimentation towards governed, economically sustainable and enterprise-wide adoption.