Loading page…
Zetaris Cloud pairs distributed data access with company permissions. Its claimed cost reduction of up to 67% rests on internal customer assessments.
Listen to this story
Zetaris is positioning its new cloud service as a way for AI agents to access company data across existing systems without first building a centralized copy. The approach could reduce data-preparation work and infrastructure duplication, but its headline savings claim—up to 67% lower total cost of ownership—comes from internal before-and-after assessments whose sample and methodology were not disclosed. Buyers will need to validate compatibility, governance and savings in their own environments.
The AI Data Harness supports structured, unstructured and streaming data sources.
Zetaris says the service works across cloud, data-platform and AI-model choices, without requiring organizations to replace existing systems.
The company says existing security policies and access controls apply when distributed data is queried.
Zetaris launched Zetaris Cloud and its AI Data Harness on October 6, 2026, offering AI systems a route to company data without first copying it into a central store. The company says the service combines in-place queries with security and access controls—a bid to connect agents to existing business systems without a separate data-centralization project.
The launch targets a bottleneck Zetaris describes between generating software and making it useful inside a business. In the company’s account, enterprise data remains scattered across cloud, on-premises and edge systems. Bringing it together can require weeks of work, leaving AI agents with partial or stale information while teams prepare a centralized dataset.
Its proposed alternative is a federated architecture: a system that queries data where it already resides rather than requiring another consolidated copy. The AI Data Harness supports structured, unstructured and streaming sources, according to Zetaris. That makes the release a data-access product for AI systems, rather than a new model that generates answers or writes code.
Zetaris Cloud delivers that architecture as a cloud-based service. The company says organizations can connect models and agents to their existing data environments without committing to a particular cloud, data platform or AI model. The claim concerns compatibility across those choices, not a requirement to replace the systems that already hold the organization’s information.
Permissions are part of the same pitch. Zetaris says security policies and access controls apply wherever data is queried. Its stated design therefore pairs access to distributed information with consistent rules about who can use it, rather than presenting faster access and governance as separate projects.
Every time enterprises copy and centralize data to build something new, they duplicate storage, compute, and networking.
Vinay Samuel, Zetaris CEO and co-founder
Zetaris claims the harness can lower total cost of ownership by up to 67%. A footnote in its launch announcement attributes that figure to internal measurements using before-and-after enterprise customer assessments. The release does not detail the customer sample or calculation method, so the headline percentage offers limited guidance about what another organization would save.
The infrastructure argument extends beyond that cost figure. By reducing copying and centralization, Zetaris aims to cut the storage, compute and energy infrastructure needed for enterprise AI. It also envisions more processing near where data is created, including across edge-device networks. That is a longer-term direction, not a completed outcome demonstrated by this launch.
Loading discussion...
Join the conversation
Explain how much weight you would give a vendor’s own customer assessments.
Be the first to share a perspective or an experience.
Reader comments
Newest comments first. Replies stay oldest first.