The rebuild makes readable what earlier technology could not capture – up to 75% of Sayari’s archive – and is expected to cut data infrastructure costs by more than 50%
WASHINGTON, Sept. 2, 2026 /PRNewswire/ — Sayari, the trusted AI platform for economic security and commercial risk, today announced it has selected Snowflake (NYSE: SNOW), the AI Data Cloud company, to rebuild the Commercial World Model, Sayari’s map of the world’s companies, the people behind them, and the trade between them. The rebuild turns more than a decade of primary-source data, 12 billion records from 715 sources across 250 jurisdictions, into a single AI-ready foundation for every Sayari product.
Sayari’s archive includes roughly 1 billion original documents including corporate filings, government gazettes, and shipping records in their original published form. For more than a decade, Sayari has built extraction pipelines that parse companies, people, and relationships from those documents with the precision its customers require. But deterministic extraction has limits. Sayari estimates that up to 75% of the useful information in its documents, the context and patterns surrounding those records, could not be captured that way. Advances in AI now put that information within reach, and rebuilding on Snowflake lets Sayari apply those techniques across the entire archive at once, not document by document.
Customers already rely on Sayari to identify who owns and controls companies and to detect risk across supply chains and business relationships. The rebuilt platform lets them go deeper, faster, and at greater scale: tracing ownership further, detecting risk earlier, and surfacing entirely new classes of risk intelligence from patterns no single record contains, with every finding traceable to the primary sources behind it.
The Commercial World Model brings this data together in one place, continuously updated, connected, and analyzed. Sayari collects it directly from official sources, including corporate registries, customs bureaus, and regulatory agencies, in more than 20 languages and scripts. Much of this is deep web data that has never been indexed including records invisible to search engines and to the general-purpose AI tools trained on the open web, and in many cases no longer available from their original publishers. Sayari’s archive is often the only place they exist in accessible form.
Sayari’s engineering team used Snowflake CoCo, Snowflake’s AI coding assistant, to accelerate the migration. The move is expected to reduce Sayari’s data infrastructure costs by more than 50%, according to the company’s projections.
“Fortune 100 companies and national intelligence agencies rely on Sayari for decisions that carry real consequences,” said Jessie Abell, chief product officer, Sayari. “Over the past decade, we built an archive of records nobody else has, but most of those records have been out of reach. Rebuilding the Commercial World Model on Snowflake lets us read all of it at global scale. Later this year, we will introduce products built on this foundation that deliver finished answers customers can trace, verify, and defend.”
“AI is only as powerful as the data behind it. Sayari has built something you can’t replicate with a model alone: a proprietary data asset that the world’s largest organizations trust for their most consequential decisions,” said Jennifer Chronis, vice president, US Public Sector, Snowflake. “Snowflake is proud to be the foundation it runs on, and to help Sayari bring AI to data the world hasn’t been able to access until now.”
Extracting information from complex documents is only part of the challenge. Sayari analysts evaluate which sources are reliable, how disclosure requirements vary across jurisdictions, and how sanctioned or other high-risk actors obscure ownership to evade detection. Sayari has built that tradecraft into its models, turning raw records into assessed intelligence: findings customers can source, verify, and act on with confidence.
Trade wars, sanctions, and geopolitical tensions are rapidly reshaping global supply chains and the risks facing companies and governments. Governments and adversarial actors increasingly use trade and financial networks to advance political, economic, and national security goals. Responding requires understanding the full web of ownership and control behind every entity an organization deals with. The rebuilt Commercial World Model is the foundation for that work, and for the products Sayari will introduce later this year.
About Sayari
Sayari is the judgment infrastructure for trustworthy, sovereign AI in economic security and commercial risk. Its Commercial World Model resolves more than 12 billion primary-source records from more than 250 jurisdictions, delivering AI that reasons like an expert analyst, shows its work, and traces every finding to a primary source customers can see and control. Trusted by regulators and the regulated alike, Sayari is used by U.S. Customs and Border Protection, U.K. HM Revenue & Customs, Fortune 500 enterprises, and thousands of professionals across more than 35 countries to secure supply chains, surface sanctions evasion and forced labor risks, and dismantle illicit networks.
SOURCE Sayari
