Cybersecurity company Palo Alto Networks has integrated its Prisma AIRS platform into the NVIDIA Enterprise AI Factory validated design to help enterprises secure artificial intelligence (AI) infrastructure while maintaining performance.
The integration places zero trust security directly within AI systems by running Palo Alto Networks Prisma AIRS Network Intercept on NVIDIA BlueField data processing units (DPUs). The setup also extends protection to cloud environments, allowing companies to manage security across the AI lifecycle while continuing AI development.
“The AI Factory is the new engine for value creation, and securing it is a board-level imperative,” said Rich Campagna, senior vice president of Product Management, Palo Alto Networks. “We are embedding trust directly into the infrastructure, giving leaders the confidence to safeguard their proprietary intelligence and deploy AI bravely.”
The design shifts security processing to a separate domain using the DPU’s hardware acceleration through NVIDIA DOCA. The system collects real-time workload data using DOCA Argus and sends it to Cortex XSIAM for AI-driven threat response through the Cortex XSOAR orchestration platform.
“AI is transforming every industry and security must evolve to protect AI factories,” said Kevin Deierling, senior vice president of Networking at NVIDIA. “To be scalable, security must be distributed and embedded within the AI infrastructure. This is achieved with NVIDIA BlueField running Palo Alto Networks Prisma AIRS to deliver robust, runtime security for the AI factory, with optimal AI performance and efficiency.”
The solution also uses centralized Hyperscale Security Firewall clusters positioned at entry and exit points of AI infrastructure. These firewalls help prevent cyber threats and stop attacks from spreading between workloads. Workloads can also route selected traffic through firewall clusters for deeper inspection.
By placing security functions on BlueField, companies can dedicate full host computing resources to AI workloads. The company said the design allows enterprises to move from general computing clusters to AI-focused infrastructure and prepare for future hardware upgrades, including NVIDIA BlueField-4.