Snowflake: AI Agents to Unlock Financial Data Value

AI Data Cloud company Snowflake has introduced Cortex AI for Financial Services, a suite of AI capabilities to help financial firms unify their data and deploy AI agents.

A key feature is the managed Model Context Protocol (MCP) Server, in public preview. It enables organisations to connect proprietary data with third-party sources from partners like FactSet and MSCI.

The MCP Server also connects customer data to AI platforms such as Anthropic, CrewAI and Salesforce’s Agentforce. This allows firms to build context-rich AI agents while maintaining security and compliance.

Baris Gultekin, VP of AI at Snowflake, notes that while the financial industry leads in tech adoption, it faces challenges with fragmented data, compliance and security.

He explains that Snowflake addresses this by bringing AI directly to the data and enabling secure interoperability with remote agents. This approach makes it easier for regulated industries to power critical use cases within a unified ecosystem.

Cortex AI for market analysis and fraud detection

Cortex AI for Financial Services is designed to accelerate tasks including market analysis, quantitative research, fraud detection and claims management.

The ecosystem includes data from a range of providers, while unstructured data from publishers like FactSet and The Associated Press is available through Cortex Knowledge Extensions.

Its Data Science Agent automates data cleaning, model prototyping and validation, allowing teams to quickly build models for quantitative research and fraud detection.

Snowflake Cortex AI SQL also provides functions for AI-powered extraction and transcription, enabling users to process documents, audio and images at scale.

MCP Server connecting AI agents and enterprise systems

The Snowflake MCP Server addresses the challenge of connecting AI agents to enterprise systems, which often required slow, customised solutions.

MCP provides a standardised way for large language models to integrate with data, APIs and services.

It connects Cortex Analyst and Cortex Search to external AI agents via a standards-based interface, unifying structured and unstructured data retrieval. This removes the need for custom integrations, accelerating the delivery of context-rich AI applications.

"Enterprises are moving from AI pilots to production, but securely connecting AI to proprietary data has been a critical barrier,” says Jonathan Pelosi, Head of Industry, Financial Services at Anthropic.

He adds that the partnership with Snowflake helps solve this by using MCP to connect an organisation's governed data directly to their model, Claude.

"Customers can now use Claude's advanced reasoning on both structured analytics and unstructured documents... while maintaining enterprise security standards."

From multi-agent architecture to business insight

Gary Lerhaupt, VP, Product Architecture, Agentforce at Salesforce, says Salesforce is excited to "deliver agent interoperability through protocols like MCP.”

He notes this will enable deeper cross-platform connectivity and accelerate the development of AI agents within Agentforce.

João Moura, Co-Founder and CEO of CrewAI, believes the next wave of enterprise AI requires orchestrating specialised agents. “For these agentic workflows to succeed... they must be grounded in secure, high-quality data," he explains.

João adds that Snowflake’s MCP Server "provides the essential, secure pipeline for our agent crews to access, analyse and act upon governed data."

Financial services firms are already using these capabilities. Ian Macomber, Head of Analytics at Ramp, explains how Ramp processes customer feedback.

“With Snowflake Cortex AI, we can securely tap into and analyse our unstructured customer data, allowing teams across Ramp to ask questions in plain English and get instant answers.”

He adds, "Snowflake makes it easy to democratise data-driven decision-making, which enables us to innovate faster."