DEVOPSdigest

Check Point® Software Technologies Ltd. is collaborating with NVIDIA to enhance the security of AI cloud infrastructure. Integrating NVIDIA BlueField DPUs, which feature a broad range of purpose-built, innovative security capabilities, the new Check Point AI Cloud Protect solution will help prevent threats at both the network and host levels.

“AI provides great benefits across healthcare, education, finance and more. At the same time, the rate and sophistication of cyber attacks is increasing, with threat actors increasingly looking at ways to disrupt AI workloads in the cloud,” said Gera Dorfman, Vice President of Network Security at Check Point Software Technologies. “We’re working with NVIDIA to deliver a new secure AI cloud solution with Check Point AI Cloud Protect that guards even the most sensitive and private AI workloads against cyber threats.”

The rapid proliferation of AI has brought about a revolution in workplace efficiency and innovation. However, this growth also creates additional attack vectors specifically targeting AI, such as backdooring AI models to control a model’s output or to gain unauthorized access to the environment, data exfiltration to expose intellectual property, and denial of service to degrade performance and reduce capacity. These threats compromise the integrity and security of AI systems and pose risks to business outcomes. They can also erode the foundational trust in AI operations, while potentially affecting other aspects of the data center. There is a critical need for a revamped security approach to protect not only the data in its traditional form but the AI models themselves, which are central to innovation and competitive edge.

Check Point aims to address these challenges with NVIDIA by integrating network and host-level security insights, offering a comprehensive solution that protects AI infrastructures from both conventional and novel cyber threats. This integrated approach helps ensure the security system is cognizant of network activities and host-level processes, which is crucial for safeguarding AI’s future.

“As AI becomes more pervasive, securing AI clouds becomes paramount,” said Yael Shenhav, Vice President of Networking Products at NVIDIA. “NVIDIA BlueField 3 enables innovators such as Check Point to offer robust cyber defense measures to secure AI cloud data centers, while also ensuring peak AI performance.”

In response to these emerging challenges, AI Cloud Protect emerges as a strategic solution, addressing the dynamic security requirements of the AI era. Designed for easy deployment and adaptability, it offers out-of-the-box security without impacting AI performance. Designed for effortless integration and scalability, the AI Cloud Protect provides a robust shield against sophisticated cyber threats. Engineered with the NVIDIA BlueField 3 DPU, which powers a new class of AI cloud data centers, and the NVIDIA DOCA software framework, AI Cloud Protect is designed to seamlessly integrate into NVIDIA’s AI ecosystems, providing:

- Robust Defense Against AI-Specific Threats: Empowers organizations to efficiently shield against model inversion, model theft and other attack vectors with unprecedented efficiency.

- Scalable, Seamless Integration: Facilitates easy deployment across diverse AI environments, ensuring security measures grow in tandem with organizational needs.

- Optimized Performance with Zero Compromise: Ensures AI operations continue unhindered, with security processes running discretely, leveraging NVIDIA’s technological infrastructure without impacting AI performance.