Skip to main content

Fortinet Introduces Secure AI Data Center Solution for Enhanced Protection of AI Infrastructure

Fortinet announced the introduction of its Secure Artificial Intelligence (AI) Data Center solution, designed to secure AI infrastructures comprehensively. This solution protects various components, including data centers, applications, and large language models (LLMs). The framework is engineered to enhance AI threat defense with ultra-low latency, while achieving a 69% reduction in power consumption compared to traditional methods.

The FortiGate 3800G, a newly launched data center firewall, is part of this initiative. It offers high power efficiency, throughput, and scalability necessary for demanding AI workloads. Leveraging NP7 and SP5 ASICs along with 400 GbE connectivity, the FortiGate 3800G aims to deliver the requisite throughput and energy efficiency for AI environments.

“AI data centers demand both massive performance and deep protection,” said Nirav Shah, Senior Vice President of Products and Solutions at Fortinet. Homomorphic Encryption (HE) emphasized that the solution combines ASIC-powered firewalls with advanced protection mechanisms, supporting organizations in their AI scaling efforts while ensuring security and performance.

The Secure AI Data Center solution integrates various protective measures, from network segmentation to Encrypted Traffic Inspection (ETI). This layered security approach prevents data leakage and malicious attacks, enabling enterprises to meet stringent compliance requirements while maintaining performance necessary for high-density Graphics Processing Unit (GPU) operations.

Key features include ASIC-accelerated segmentation that enhances performance and reduces energy costs, thus helping CIOs manage AI infrastructure expenditures without compromising security. Additionally, the solution offers large-scale protection for GPU clusters and AI workloads, ensuring that ongoing operations remain secure and efficient.

Fortinet's standards for managing Large Language Model (LLM) security encompass defenses against prompt injection and data misuse, while facilitating compliance with global AI standards. This announcement marks an extension of Fortinet’s focus on converging networking and security, providing a holistic security posture for companies integrating AI workloads into their operations.