TrendAI brings NVIDIA Nemotron models into on-premises security platform

What TrendAI is launching

TrendAI says the integration extends AI-assisted security operations into environments where organizations need sovereignty, data residency and operational control. The capabilities named in the announcement include guided investigation, security reasoning and agentic SIEM.

TrendAI Vision One for Sovereign and Private Cloud (SPC) is available today. It lets organizations deploy TrendAI Vision One in customer-controlled private cloud, sovereign cloud and on-premises environments. The Nemotron integration itself is described as planned: TrendAI invites visitors to its booth at NVIDIA GTC Berlin (October 21-22, 2026) to preview these capabilities, and one partner is working on it ahead of general availability.

The release builds on an existing collaboration with NVIDIA. TrendAI also points to joint work on OpenShell, which it says adds a further level of security for agents and gives organizations control and visibility of agentic activities.

Why open models matter for sovereignty

The company argues that highly regulated organizations need AI-driven security that fits their data-control requirements. In its words, Nemotron open models can support this by bringing the model to the data rather than the data to the model.

Customers decide where the model runs, which version is deployed and who can access it. The AI-powered security operations run on NVIDIA accelerated computing in the customers' own data centers, sovereign clouds and restricted-connectivity environments.

Rachel Jin, CEO of TrendAI, said: "Sovereignty extends beyond where data resides. Enterprises also need control over where the intelligence acting on that data operates, and how it is governed." John Fanelli, Vice President, Enterprise Software at NVIDIA, said open models give organizations "the transparency and control to run on their own terms, which is critical for sovereign environments."

Supported capabilities

According to the announcement, Vision One SPC with the Nemotron integration supports the following:

• Customer-controlled AI: Nemotron open models run on customer-controlled infrastructure, with no dependency on external inference services.
• AI-assisted threat analysis: reasoning across security signals to identify complex threat patterns, attack chains and AI-native attack paths.
• Policy-based threat response: protective actions through organization-defined policies and security-team controls.
• Guided security operations: help for security teams to investigate threats and prioritize response activities.
• Agentic SIEM capabilities: correlating telemetry and assisting with security reasoning while maintaining team oversight.

Partner views and what it means for customers

TITECH, which delivers sovereign security operations to customers in Oman, is a design partner on the Nemotron integration. CEO Mohammed Hassan said Vision One SPC has given local customers a foundation for keeping security operations under customer control while complying with country data residency regulations. He added that Nemotron-powered capabilities will let TITECH offer advanced threat analysis and AI-assisted security operations while maintaining data sovereignty.

Mathias Schick, Business Manager of Security at Bechtle AG, called integrating AI-powered security into sovereign and private-cloud environments "an important step toward enabling secure AI adoption at scale," particularly for organizations with stringent security and compliance requirements.

For organizations that must keep security data and AI processing within their own boundaries, the offer is aimed at removing a trade-off: using AI in security operations without sending data to external services. The Nemotron-based features are still to be previewed, so customers should treat them as planned capabilities, while the SPC platform itself can be deployed now.