Relational AIa startup that applies artificial intelligence on to relational data, today announced the overall availability of its Knowledge graph coprocessor on the Snowflake Data Cloud. The offering, which was first announced last yr (which we covered intimately here), enables Snowflake customers to construct knowledge graphs and leverage advanced AI and analytics capabilities without having to maneuver data out of their Snowflake environment.
“This is a big win for the client,” said Molham Aref, co-founder and CEO of RelationalAI, in an interview with VentureBeat. “Typically, to do the things we do – graph evaluation, rules-based reasoning, prescriptive evaluation, predictive evaluation – you will have to drag the info back out of Snowflake and put it into some extent solution. If you're a chief data officer and also you've spent the last two or three years moving all of your data into Snowflake, the very last thing you would like to do is pull it back out.”
The RelationalAI Knowledge Graph Coprocessor is uniquely designed to use AI on to structured enterprise data in relational databases like Snowflake. This is in contrast to most machine learning and AI approaches that concentrate on unstructured data like text, images, and audio.
Utilizing machine learning to unlock the potential of structured enterprise data
“We are the one system, the one solution that applies AI to relational data as comprehensively and extensively as we do,” Aref said. “The most respected data (within the enterprise) is definitely often stored in a structured form. And historically, machine learning and AI haven’t worked natively or directly on that data.”
RelationalAI reports strong initial momentum and “never-ending demand” for its offering, particularly within the financial services, telecommunications, retail and consumer goods industries. Customers comparable to AT&T, Block, Ritchie Bros and Blue Yonder use the platform to create knowledge graphs that add a semantic layer to their complex enterprise data.
Enabling generative AI for firms with knowledge graphs
The launch on Snowflake is a serious milestone for the 50-person startup, which reportedly raised $122 million in funding at a $569 million valuation. Pitch BookSnowflake, which has a market capitalization of over $40 billion, highlighted its partnership with RelationalAI for instance of its growing data cloud ecosystem in recent quarterly earnings calls.
With large language models like GPT-4o capturing the general public's imagination, Aref believes knowledge graphs play a key role in applying generative AI to the enterprise.
“All of these items need something like a knowledge graph as an interface to have the opportunity to check with the underlying structured data, to speak with you and to grasp and use the business logic of the applications,” he said.
With generally available Snowflake integration, RelationalAI is well-positioned to assist enterprises align the ability of AI with the precious structured data they have already got within the cloud. With growing interest in generative AI and knowledge graphs, the startup's technology fills a critical missing piece in enterprise AI stacks.
