Adaptive Retrieval allows customer AI agents and large language models (LLMs) to access and assemble licensed S&P Global data using natural language queries. Alongside the existing Deterministic Retrieval, both methods are now available through the S&P Global AI Data Portal, giving customers flexible and accurate access to S&P Global data across a full spectrum of AI or agentic workflows.
Adaptive Retrieval lets AI agents and LLMs pull data from many different sources at once and automatically handle requests that involve multiple datasets. This makes it well suited for complex, multi-step tasks, including in-depth research and report generation. Deterministic Retrieval, built on the Kensho LLM-ready API, which has been available to customers since 2025, gives AI systems API-driven access to S&P Global data through direct, structured queries. This method is ideal for focused tasks including researching a specific company or analyzing earnings call transcripts. Customers can use one method or both, depending on how their systems are set up and what they need to accomplish.
Since S&P Global's breadth and depth of data is already cited, structured, and ready for AI systems to use, customers can focus on building products and generating insights rather than preparing and managing data.
This launch follows S&P Global's recently announced evolution of its Market Intelligence operating model, which brings together data, AI, software, and workflow capabilities to better support how customers discover and consume intelligence. It reflects the role of Market Intelligence's newly formed Kensho Data Platforms vertical in delivering world-class client interfaces, including Capital IQ Pro, to create more AI-native user experiences and make proprietary intelligence easier to access, connect, and act on.
View the full press release here.