Kensho is S&P Global’s AI engine, built around enterprise data retrieval optimized for AI applications. Its core offerings include “Deterministic Retrieval” and “Adaptive Retrieval” – approaches designed to give large language models reliable, LLM-ready access to structured data rather than the inconsistent results generic retrieval-augmented setups often produce.
Deterministic retrieval specifically addresses one of the harder problems in enterprise AI: getting a language model to return the same correct answer every time it’s asked the same factual question against structured data, rather than a plausible-sounding but inconsistent one. Combined with direct access to S&P Global’s market intelligence data, that’s a meaningfully different bar than a general-purpose AI search layered on top of unstructured documents.
Kensho is built for enterprise deployment – there’s no self-serve signup or published pricing on the site. Expect an institutional sales process rather than a free trial, consistent with its target market of financial institutions and large enterprises already working with S&P Global data.
