Snowflake and Salesforce initiated the Open Semantic Interchange project in collaboration with industry partners to solve the problem of the fragmentation of enterprise system data definitions. There is the BlackRock, dbt Labs, and RelationalAI collaboration that is developing a universal semantic framework that normalizes business measures to apply AI.
Big data Industry giants come together to address a trillion-dollar problem
Salesforce, Snowflake, Tableau, and over a dozen other technology firms have all built a universal standard of how business data can be defined and shared across platforms in hopes of fixing what executives have described as the biggest bottleneck of AI, according to Salesforce Ben. Known as the Open Semantic Interchange (OSI), this project will unite the competing parties to assist them in distributing data to bring about differences in the definitions of data between systems in enterprises.
Snowflake declared a radical plan to become the leader in Open Semantic Interchange (OSI), an amateurish open source project. In the age of AI, the problem of mixed tool and platform semantics of data has become one of the largest bottlenecks both to human and AI analysis. This project satisfies that need by proposing a standard, vendor-neutral semantic model specification that standardizes the definition and sharing of semantic metadata.
A comprehensive framework eradicates the clashing definitions of business metrics
The imminent introduction of agentic AI is exerting greater pressure on businesses to structure their data and present agents with explicit business indicators. VentureBeat gives the case of determining who an active customer is – when a sales platform might define it as someone who has purchased within the past 90 days, a marketing team might define it as someone who has interacted with content within the past month.
Semantic standardization speeds up AI adoption through platforms
With the changing role of AI in the ways business uses data, the semantics of consistency have never been more required. Currently, any tool views business measures and metadata variably – making them perplexing, slowing adoption, and undermining trust in AI-based insights. The OSI attempts to solve this problem and offer a single, vendor-neutral standard of business, domain, and industry semantics.
Among the objectives are improved interoperability between tools and platforms, faster adoption of AI and BI, and improved operations. In the absence of a common semantic specification, data and AI teams can occasionally waste weeks sorting out conflicting definitions or redoing work on different platforms. OSI removes this difficulty by providing a standard specification that minimizes overhead and liberates teams to work on innovation rather than on troubleshooting.
Open source collaboration drives vendor-neutral industry standard
At Snowflake, we have always thought that interoperability and open standards are the keys to the full potential of AI with your data, said Christian Kleinerman, EVP of Product, Snowflake. With the Open Semantic Interchange project, we are delighted to be at the forefront together with our partners to address a foundational problem of AI, which is the absence of a standard semantic standard.
Southard Jones, Chief Product Officer, Tableau, added that the future of AI relies on trust–and trust begins with reliable data that must be consistent. By co-leading the Open Semantic Interchange, together with Snowflake and our partners, we are establishing the backbone that any AI agent and BI application requires: a shared semantic framework, which holds meaning cross-platform. It is the Rosetta Stone of business data.
This is an industry-wide project that focuses on the basic issues that have obstructed the implementation of AI, laying the groundwork for more dependable artificial intelligence applications. Through the creation of shared semantic standards, the involved firms are aligning themselves to drive innovation faster and minimize the complexity of operations that will eventually help businesses by utilizing AI at full potential in their data ecosystems.