Check Point Unveils AI Factory Security Blueprint to Strengthen Protection of AI Infrastructure from GPU Servers to LLM Prompts
DUBAI, UAE – Check Point® Software Technologies Ltd. (NASDAQ: CHKP), a prominent leader in cybersecurity, has launched the AI Factory Security Architecture Blueprint. This framework is designed as a vendor-tested reference architecture to secure private AI infrastructures from the hardware layer to the application layer. By utilizing Check Point’s advanced firewall and AI security technologies, in conjunction with NVIDIA’s BlueField data processing capabilities, the blueprint integrates security at every level of the AI factory and data center.
Nataly Kremer, Chief Product Officer at Check Point, highlighted the critical nature of AI infrastructure, noting its status as one of the most valuable yet vulnerable assets within organizations. The AI Factory Security Blueprint aims to safeguard these investments from the ground up, making security a foundational aspect rather than an afterthought.
The Strategic Importance of AI Data Centers
AI data centers have become essential components of enterprise infrastructure due to their strategic value and inherent vulnerabilities. Organizations are increasingly establishing private AI environments to protect intellectual property, adhere to sovereignty requirements, and reduce public cloud expenses. This rapid evolution has resulted in significant investments in GPU clusters, training pipelines, inference workloads, and proprietary models. However, the development of security architecture has not kept pace with these advancements.
Unlike traditional data centers, AI computing environments incorporate high-performance GPU clusters, distributed training pipelines, large-scale data lakes, and real-time inference APIs. This complexity creates attack surfaces that conventional security tools are ill-equipped to manage. Threats include training data poisoning, model theft, lateral movement within Kubernetes namespaces, prompt injection targeting inference APIs, and supply chain vulnerabilities arising from open-source dependencies.
Layered Protection Across Four Levels
The Check Point AI Factory Security Blueprint employs a multi-layered protection strategy across four distinct levels:
- Perimeter Layer: The Check Point Maestro Hyperscale Firewall provides Zero Trust Network Access (ZTNA), virtual security group segmentation, and scalable policy enforcement at the entry point to the AI fabric. This layer effectively manages north-south traffic from external users, internet sites, and enterprise networks.
- Application and LLM Layer: Check Point AI Agent Security safeguards inference APIs and LLM endpoints against threats such as prompt injection, data exfiltration, adversarial queries, and API abuse. This layer extends beyond traditional web application firewalls, with Check Point AI Agent Security integrated into Check Point Firewalls across various deployment models, including cloud, virtual, and appliance forms.
- AI Infrastructure Layer: In partnership with NVIDIA, Check Point has integrated its firewall and threat prevention capabilities directly into NVIDIA BlueField data processing units (DPUs) via the NVIDIA DOCA software platform. This integration provides hardware-accelerated, inline security at the infrastructure level, ensuring high-performance AI prompt defense and traffic inspection without consuming CPU or GPU resources.
- Workload and Container Layer: Check Point collaborates with third-party microsegmentation solutions to facilitate micro-segmentation and east-west traffic control within Kubernetes clusters. This approach prevents lateral movement between inference namespaces and isolates compromised containers before they can propagate.
Aligning with Security Principles and Regulatory Frameworks
The AI Factory Security Blueprint aligns with the Cybersecurity and Infrastructure Security Agency (CISA)’s principle that AI must be Secure by Design. This principle advocates for security to be embedded from the outset—in the fabric, hardware, and orchestration layers—rather than added as an afterthought. Check Point’s architecture enforces a Zero Trust model at every interaction, ensuring that every user, API call, and service request is authenticated, authorized, and continuously validated.
Additionally, the blueprint is designed to comply with AI governance frameworks, including the NIST AI Risk Management Framework and Gartner AI TRiSM. It provides the necessary traceability, auditability, and policy enforcement to meet emerging regulations such as the EU AI Act, GDPR, HIPAA, PCI-DSS, and ISO 42001.
As reported by cyberwarriorsmiddleeast.com, the AI Factory Security Blueprint represents a significant step forward in securing AI infrastructures.
About Check Point Software Technologies Ltd.
Check Point Software Technologies Ltd. is a global leader in cybersecurity, protecting over 100,000 organizations worldwide. The company is dedicated to securing enterprises’ AI transformation through a prevention-first approach and an open ecosystem architecture. Check Point enables organizations to block advanced threats, prioritize vulnerabilities, and automate security operations across complex digital environments. Its unified architecture simplifies protection across hybrid networks, multi-cloud environments, digital workspaces, and AI systems. Structured around four strategic pillars—Hybrid Mesh Network Security, Workspace Security, Exposure Management, and AI Security—Check Point delivers consistent protection and visibility across multivendor environments, allowing organizations to reduce risk, enhance efficiency, and accelerate innovation without increasing complexity.
Follow the latest developments and breaking updates in the Latest News section.
Published on 2026-03-26 11:58:00 • By Editorial Desk
