S&P Global Launches Adaptive Retrieval to Help AI Agents

S&P Global announced on July 21, 2026 the launch of Adaptive Retrieval, a service that allows AI agents and large language models to access and assemble S&P Global's licensed data through natural language queries, without the engineering overhead that has historically made connecting high-quality financial data into AI systems a significant undertaking.

The launch makes S&P Global what it describes as the first data provider to offer two complementary retrieval methods through a single portal, according to the press release. 

Deterministic Retrieval, built on the Kensho LLM-ready API available to customers since 2025, gives AI systems structured, API-driven access to specific data, suited for focused tasks like researching a single company or analyzing an earnings call transcript. 

Adaptive Retrieval handles the more demanding requirement: enabling AI agents to pull data from multiple sources simultaneously and automatically manage requests that span datasets, covering the kind of complex multi-step tasks that autonomous and multi-agent systems increasingly need to perform.

"The use of AI in financial services is rapidly accelerating and evolving, from tightly controlled workflows to fully autonomous, multi-agent systems," said Sally Moore, Chief Client Officer and Co-Head of Market Intelligence at S&P Global. "With Deterministic and Adaptive Retrieval now available together, we're ensuring that S&P Global's trusted data is accessible across that full range of workflows, so customers can access data the way they need it today and adapt as their architectures evolve."

The value S&P Global is positioning around its AI Data Portal is not the retrieval mechanism itself, it is the data that sits behind it. Connecting high-quality, licensed financial data into AI systems has historically required significant engineering work: finding and validating sources, building the retrieval logic to access them accurately, and maintaining domain expertise to know whether the output is reliable. 

For organizations moving from human-driven processes toward AI agents that act independently, that engineering burden has become a genuine barrier to deployment. S&P Global's argument is that its data removes that barrier. 

Customers can connect AI agents directly to the portal and begin using the data for workflows rather than spending engineering resources preparing it, according to the press release.

"Cited, verifiable S&P Global data provides the trusted foundation on which higher-value AI-native experiences can be built," said Bhavesh Dayalji, Head of Kensho Data and Intelligence at S&P Global. "Now, S&P Global data flows directly into the tools and platforms where customers work through financial skills and plugins, and MCP apps that allow customers to visualize, explore, and interact with S&P Global data inside AI applications."

The launch sits within a broader operational restructuring at S&P Global Market Intelligence, which has recently formed a Kensho Data Platforms vertical to consolidate its data, AI, software, and workflow capabilities, with the AI Data Portal positioned as the primary client-facing interface for AI-native access to its intelligence.