#United States #NVIDIA #San Carlos #Check Point #AI Cloud Protect

Check Point's Innovative AI Cloud Protect Solution

In a significant move for cybersecurity, Check Point Software Technologies Ltd. has announced the launch of its sophisticated security solution, AI Cloud Protect. Developed in collaboration with NVIDIA, this solution is tailored for enterprises deploying AI models and applications. It ensures that businesses can protect their AI-driven innovations without impacting performance.

The rapid growth of AI technologies in enterprises has brought about a formidable challenge in cybersecurity. With over half of corporate networks using AI tools, these systems have become prime targets for cyber attacks. According to Check Point's data, one in every eighty generative AI prompts has led to exposure of confidential data, underscoring the dire need for robust security measures. A recent Gartner report highlighted that 32% of organizations faced AI attacks involving prompt manipulation over the past year, while 29% encountered attacks on their generative AI infrastructure. As enterprises rush to integrate AI, safeguarding their development and operational pipelines is more critical than ever.

AI Cloud Protect addresses these issues, safeguarding the entire AI factory ecosystem, from model development to commercial deployment. Operational on the NVIDIA BlueField-3 DPU, AI Cloud Protect delivers comprehensive security without consuming CPU or GPU resources, a common drawback in traditional security solutions that can hinder performance.

Comprehensive Security Architecture

AI Cloud Protect enhances network-level protection to shield AI infrastructure from unauthorized access and to reduce risks associated with data corruption and model leakage. By leveraging NVIDIA's DOCAArgus technology, it provides visibility into all processes running on AI nodes, allowing Check Point to detect and prevent malicious processes, including those involving public large language models (LLMs).

Nataly Kremer, Check Point's Chief Product Officer, emphasized the importance of incorporating security into AI factories from the design phase. She stated, "In an environment where companies are vying to build AI-driven innovations, any oversight can be detrimental. Check Point collaborates with NVIDIA to create AI factories with embedded security, ensuring protection for models, data, and infrastructure without compromising innovation speed."

David Reber, NVIDIA's Chief Security Officer, echoed this sentiment, remarking, "Security is essential for next-generation AI infrastructure. Through our partnership with Check Point, we are integrating BlueField acceleration and the NVIDIA DOCA Argus runtime security framework into the AI Cloud Protect platform, enabling enterprises to confidently deploy AI."

Future-Ready Developments

Looking ahead, AI Cloud Protect is set to function on the forthcoming NVIDIA BlueField-4 DPU, which promises up to six times the computational power for AI inference and storage operations, thus facilitating scalable and secure AI operations at unprecedented speed.

Check Point offers a comprehensive portfolio of security products beyond AI Cloud Protect, enabling companies to protect their entire AI supply chain, including cloud infrastructure, enterprise applications, and end-users. The Check Point CloudGuard WAF (Web Application Firewall), integrated with Lakera functionality, prevents threats like prompt injection and jailbreaks, protecting AI enterprise applications. Infinity GenAI Protect enables secure deployment of generative AI applications within enterprises, applying security policies to detected generative AI applications to prevent data leaks and ensure compliance.

Together, these solutions provide end-to-end security across infrastructure, applications, and user layers, helping organizations navigate every stage of AI adoption confidently. Currently, Check Point is conducting trial operations for AI Cloud Protect in collaboration with clients from the financial services sector and strategic partners like World Wide Technology (WWT).

Chris Konrad, Vice President of WWT's Global Cyber Division, remarked, "With numerous enterprises rapidly building AI factories, the combination of Check Point's AI Cloud Protect and NVIDIA BlueField acceleration delivers enterprise-grade protection without compromising the performance demanded by modern AI applications from model training to inference."

Addressing Unique Security Challenges

AI factories present unique security challenges:

• - Extensive attack surfaces due to large-scale computation and data pipelines.
• - Emergent threats like prompt injection, jailbreaks, and model contamination.
• - Risks of data leaks from employee usage of AI outside corporate control.

AI Cloud Protect effectively addresses these concerns by offering:

• - Comprehensive protection across network, host, and application layers.
• - Accelerated security through NVIDIA BlueField, ensuring zero performance impact.
• - Centralized management and automation for security orchestration across thousands of AI nodes.

For further insights on AI Cloud Protect, please visit Check Point's official website.

About Check Point

Check Point Software Technologies stands as a global leader in cybersecurity, employing AI-driven solutions to protect over 100,000 organizations worldwide. The Infinity Platform provides a security-first approach, integrating management of on-premises, cloud, and workspace environments to deliver flexibility, simplicity, and scalability for businesses.