Enterprise AI is moving rapidly from experimentation to full-scale production, driven by purpose-built AI factories that combine accelerated compute, high-performance networking, and advanced software stacks. But as organizations operationalize AI at scale, the security stakes rise sharply. From protecting sensitive data and models to securing east-west traffic across GPU-dense environments, trust must be engineered into the AI factory itself.

Join Palo Alto Networks for an exclusive executive dinner and peer discussion focused on how organizations can securely deploy and operate enterprise AI factories built on NVIDIA’s accelerated computing platforms. This conversation will explore how security leaders can enable innovation without slowing it down by embedding protection across the AI lifecycle, from data ingestion and model training to inference and runtime operations.

Attendees will exchange practical insights on securing high-performance AI infrastructure, managing AI-specific attack surfaces, and aligning security architecture with the operational realities of large-scale AI systems. The discussion will also examine how enterprises can reduce complexity by converging networking and security controls while maintaining visibility, resilience, and governance across AI workloads.

Among the discussion topics:

• Embedding security into the AI factory architecture to support scale, performance, and trust from day one
• Protecting data, models, and GPU-accelerated infrastructure across training and inference environments
• Securing east-west traffic and high-speed networking in AI factories powered by advanced DPUs and accelerated networking
• Reducing risk and operational friction by integrating security directly into AI infrastructure rather than layering it on later