Fortinet Unveils End-to-End AI Infrastructure Security
▼ Summary
– Fortinet launched the Secure AI Data Center solution, an end-to-end framework designed to protect AI infrastructures with advanced threat defense and 69% lower power consumption on average.
– The solution includes the FortiGate 3800G firewall, which provides high performance, energy efficiency, and real-time GPU-cluster protection using NP7 and SP5 ASICs and 400 GbE connectivity.
– It secures the entire AI workflow, from network segmentation and encrypted traffic inspection to application and model-level defense, preventing data leakage and malicious inputs.
– The framework offers LLM security to guard against prompt injection and data misuse, and includes quantum-safe security features like Post-Quantum Cryptography for future threats.
– By unifying security under the Fortinet Security Fabric, it simplifies compliance, reduces operational complexity, and ensures uninterrupted uptime for mission-critical AI operations.
Fortinet has launched its Secure AI Data Center solution, a comprehensive framework engineered to protect artificial intelligence infrastructures from end to end. This system secures the entire AI stack, from foundational data center components to applications and large language models, while delivering advanced AI threat defense with minimal latency. A standout benefit is its significant reduction in energy use, cutting power consumption by an average of 69% compared to conventional security setups.
Included in this rollout is the new FortiGate 3800G, a high-performance data center firewall built to meet the rigorous demands of AI workloads. Equipped with NP7 and SP5 ASICs and 400 GbE connectivity, the 3800G provides the hyperscale throughput, energy efficiency, and real-time protection for GPU clusters required by the Secure AI Data Center framework.
Nirav Shah, Senior Vice President of Products and Solutions at Fortinet, emphasized the dual priorities for AI data centers: “They require both massive performance and deep protection. Our Secure AI Data Center solution brings these together, integrating ASIC-powered firewalls such as the FortiGate 3800G with sophisticated safeguards for data, applications, and LLMs. This lets organizations expand their AI operations without trading off security, speed, or efficiency.”
The Fortinet Secure AI Data Center framework extends protection across the complete AI workflow. It covers network segmentation, encrypted traffic inspection, and application and model-level security. By embedding security at every layer, the solution helps enterprises prevent data leaks, block harmful inputs, and satisfy evolving AI compliance mandates, all while preserving the performance essential for high-density GPU settings.
Key features of the solution include:
- ASIC-accelerated segmentation and performance: This enables hyperscale throughput while lowering energy costs per Gbps, supporting data center sustainability and helping CIOs manage AI infrastructure budgets without performance loss.
- Scalable AI infrastructure protection with Fortinet data center firewalls: Through ASIC-driven performance and 400 GbE connectivity, these firewalls provide real-time security for large GPU clusters and AI workloads, optimizing both protection and cost efficiency.
- LLM security and data protection: This guards large language models against prompt injection, data leakage, and misuse by overseeing all model traffic and enforcing input/output guardrails across local, hybrid, and public-cloud deployments.
- Operational efficiency via the Fortinet Security Fabric: By unifying firewall, application security, and AI runtime protections under a single management interface, the solution simplifies compliance, shortens audit times, and reduces operational overhead.
- Granular application security for AI models and APIs: Layered defenses inspect and sanitize all traffic before it reaches AI models, preventing code injection, tampering, and data theft.
- Quantum-safe security: Integration of Post-Quantum Cryptography and Quantum Key Distribution helps safeguard AI data confidentiality and compliance against future quantum computing threats.Together, these capabilities allow the Secure AI Data Center solution to offer a unified strategy for securing GPU clusters, AI workloads, and data pipelines. It enforces zero-trust segmentation, maintains alignment with global standards, and ensures continuous uptime for mission-critical operations.
Huy Ly, Head of Global IT Security at Monolithic Power Systems, shared his perspective: “AI is reshaping every part of our business, from product design to supply chain management, but it also brings new operational hurdles. Fortinet’s Secure AI Data Center solution gives us the visibility, performance, and protection necessary to confidently operate high-density GPU clusters. With this framework integrated, we can protect sensitive models and data while achieving hyperscale throughput with improved efficiency and cost performance.
(Source: HelpNet Security)
