VAST Data Launched Confidential AI System DataEnclave
The new runtime environment secures AI processing by isolating sensitive data within hardware-based containers.
Updated on Sept. 26, 2026 in Artificial Intelligence
Live Poll
Do you trust that new security tools can sufficiently protect your sensitive data from AI model leaks?
VAST Data has introduced DataEnclave, a confidential AI runtime system designed to secure model processing and data handling. The platform uses Nvidia Confidential Computing to protect information by verifying the computing environment before decryption.
Why it matters
The system addresses critical security vulnerabilities that arise when sensitive AI models and data must be decrypted in GPU memory during inference. By providing hardware-isolated execution, it allows enterprises to maintain independent control over their security keys.
DataEnclave utilizes hardware-based attestation to log lifecycle actions and attestation events in a tamper-proof audit trail. The platform supports deployment across both connected and air-gapped infrastructure.
The players
VAST Data
This data platform company specializes in high-performance storage infrastructure for artificial intelligence and large-scale data applications.
Nvidia
This technology company is a leader in GPU manufacturing and provides the confidential computing hardware necessary for DataEnclave.
The details
DataEnclave integrates into the existing VAST AI Operating System to facilitate isolated environments for AI agents via AgentEngine. The system supports a broad partner ecosystem, including model builders like Nvidia, Cohere, and CrowdStrike, alongside hardware partners such as Cisco, Lenovo, and Supermicro.
Timeline
September 26, 2026: VAST Data officially launched the DataEnclave system.
The Tech Race
As businesses increasingly shift AI inference to the edge and cloud, the industry is moving away from legacy storage models toward hardware-verified security. This launch positions VAST Data to compete with other infrastructure providers that prioritize data sovereignty and air-gapped security.
Enterprises can now manage sensitive AI agents with greater control over data and model keys, reducing the risk of unauthorized access during inference. These capabilities enable organizations to deploy sophisticated models in secure environments while maintaining a transparent, tamper-proof audit log.
The takeaway
Securing AI at the infrastructure level is becoming a prerequisite for large-scale enterprise adoption of generative models. Organizations should prioritize runtime environments that support hardware-based attestation to ensure model integrity throughout the processing lifecycle.
Further reading
Learn more about the latest innovations in the Artificial Intelligence sector.
Source note: This article includes information reported by The Manila times.
Live Poll
Do you trust that new security tools can sufficiently protect your sensitive data from AI model leaks?
