Metadata standardization unlocks AI potential

As enterprises race to adopt artificial intelligence and automate decision-making, metadata standardization has become a defining barrier to progress. Without a shared framework for describing, organizing and governing data, organizations risk building AI on a fragmented and unreliable foundation.

Despite advances in cloud storage and compute, metadata remains disconnected, inconsistent and often manual. Industry experts argue it’s not about speed or cost anymore — it’s about shared meaning. From adaptive compute to agentic AI, the message is clear: Without unified metadata, even the most powerful platforms will fall short, says Sanjeev Mohan, principal at data advisory firm SanjMo, according to Sanjeev Mohan (pictured), principal at data trend advisory company SanjMo.

SanjMo’s Sanjeev Mohan talks about metadata standardization.

“We should be spending our time not on [the] Polaris of the world. We should be spending our time on [the] Horizons of the world,” Mohan said. “The difference is that Polaris is a technical metadata store and Horizon is where the business knowledge is being captured. The business semantics, the knowledge graph, that is what will fix AI.”

Mohan spoke with theCUBE’s Dave Vellante and Rebecca Knight as a part of the day two keynote analysis at Snowflake Summit, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how metadata standardization is essential for unlocking the full value of AI and modern data platforms, emphasizing that without a unified approach to metadata, even the most advanced technologies will fall short of their potential. (* Disclosure below.)

Why metadata standardization matters for platform evolution

While Snowflake Inc. is making strides with transactional and analytical integration through its Crunchy Data Solutions Inc. acquisition, the bigger opportunity lies in harmonizing how metadata is defined, stored and accessed, Mohan explained. That fragmentation across platforms and vendors continues to introduce unnecessary friction, especially in multicloud and multi-vendor environments.

“There is no metadata standard. We need a TCP IP for metadata basically,” he said. “There are open standards. There’s something called OpenMetadata — it’s been around for four years. I don’t think it has gone anywhere.”

The conversation also explored how Snowflake’s new Gen2 Standard Warehouse aims to reduce operational waste by offering elastic, usage-based scaling. The traditional “T-shirt sizing” model of provisioning compute is not aligned with modern workload variability, Mohan noted. By moving toward adaptive compute, Snowflake is addressing both cost optimization and developer efficiency.

“Customers are saying we want minimal overhead, what AWS would call undifferentiated heavy lifting,” he added. “Why? We pay our IT folks to help deliver business goals, not to tune the database and tune the cluster and size it.”

But solving infrastructure inefficiency is only part of the picture. The rise of open table formats such as Iceberg has introduced new complexity in managing metadata at scale. Mohan expressed concern that too much attention is being paid to technical implementation details rather than outcomes that drive AI progress.

“We are spending inordinate amount of time on Iceberg,” he said. “Iceberg should be invisible. We should never have to think about it.”

Tools such as Snowflake’s Horizon Catalog represent the kind of metadata layer enterprises need, one that captures business context and supports AI-driven use cases, Mohan added. Without that layer of semantic understanding, even the most advanced platforms will struggle to meet enterprise expectations for insight and automation.

“There’s a lot of standardizations going on at [the] agent level. There is standardization communication with MCP and A2A,” Mohan said. “The only place there’s no standardization is metadata. There is no metadata standard.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of Snowflake Summit:

(* Disclosure: TheCUBE is a paid media partner for Snowflake Summit. Neither Snowflake, the primary sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE