Enterprise data teams may finally have a way to tap into vast stores of unstructured data without spending months moving files around.
Komprise today announced Transparent File Tables, a new capability that exposes enterprise file data as queryable Apache Iceberg tables for AI, analytics, and business intelligence platforms, including Snowflake and Databricks.
The launch addresses a longstanding challenge facing organizations pursuing AI initiatives: most enterprise data exists in unstructured formats such as documents, images, videos, and research files, yet only a tiny fraction is accessible for analytics or AI workflows.
Komprise brings unstructured data into AI workflows
According to Komprise, unstructured data accounts for more than 80% of enterprise data footprints, but less than 1% is currently used in AI applications.
The company argues that traditional ingestion approaches often require organizations to copy large volumes of data into analytics environments, creating high cost, complexity, and governance hurdles.
Komprise Transparent File Tables takes a different approach. Rather than physically relocating data, the technology provides a structured view of files stored across data centers and cloud environments while preserving references to their original locations.
“The reason 99% of enterprise unstructured data has been dark to AI and analytics is because discovering and generating its schema and moving it is inherently complex and costly,” said Kumar K. Goswami, CEO and co-founder of Komprise, in the announcement.
“Komprise brings to light the huge petabytes of enterprise unstructured data in a form that data teams can access easily and transparently for analytics. Komprise Transparent File Tables opens a whole new world to AI,” he added.
Metadata strategy supports data governance
The new offering builds on technology foundations Komprise has been developing as enterprises search for ways to make AI projects more effective.
In an interview with Channel Insider last year, Komprise co-founder and COO Krishna Subramanian described the company’s broader strategy around creating a comprehensive metadata layer across enterprise file environments.
“We’re building a metadata base, if you will,” Subramanian told Channel Insider. “This is a powerful way to find the right unstructured data for what you need.”
The company’s platform indexes enterprise data into what it calls a Global Metadatabase, allowing organizations to catalog files across hybrid environments. Users can enrich those datasets with metadata tags, content analysis, header scanning, and sensitive data classification before exposing them to analytics and AI tools.
By exporting Transparent File Tables from that metadata repository, organizations can query unstructured data through familiar lakehouse and analytics interfaces without requiring direct interaction with the underlying Komprise platform.
“Komprise leverages the industry-standard Iceberg format to allow data analysts and data engineers to directly query unstructured data from tools of their choice without moving any of the raw data,” Subramanian told Channel Insider via email ahead of this launch.
“Komprise always leverages open formats such as NFS, SMB/CIFS, and S3/Object, and Iceberg is a natural choice for queries.”
Enterprise AI projects face a data quality challenge
The announcement comes as many organizations continue to struggle with preparing data for AI deployments.
Subramanian told Channel Insider that unstructured data remains one of the biggest obstacles to achieving reliable AI outcomes, particularly as enterprises increasingly adopt commercial AI models rather than building their own from scratch.
“AI has a finite amount of compute. If you clutter that up with junk, you will get junk answers,” said Subramanian. She also noted that while awareness of the challenge is growing, many organizations still lack a clear path forward.
“There is greater recognition now that unstructured data is a problem, but I think a lot of enterprises still don’t necessarily know the solution to fix that problem,” Subramanian said.
Komprise positions Transparent File Tables as one potential answer by allowing enterprises to curate, classify, and govern data before it reaches analytics platforms or AI pipelines.
Combining structured and unstructured data
Subramanian highlighted several use cases for the new technology.
In pharmaceutical research environments, analysts could combine project files generated by laboratory instruments with financial data and research management systems inside platforms such as Snowflake or Databricks.
Media organizations, meanwhile, could use structured project information to identify and selectively access relevant archives for AI-driven content analysis and summarization workflows.
If full file access is required, Komprise said its existing Intelligent AI Ingest technology can transfer only the necessary data rather than entire repositories.
Early access begins
Komprise said the Transparent File Tables offering is available immediately through an early-access program.
The launch expands the company’s effort to help enterprises prepare unstructured data for AI adoption while reducing the cost and operational burden associated with large-scale data movement.
As organizations continue to search for ways to unlock value from years of accumulated file data, technologies that make those datasets more accessible without requiring massive migrations could attract growing attention.