AI-Powered Defense: Leidos, VAST Data, and NVIDIA Team Up to Tackle Cyber Threats

WASHINGTON D.C. – October 28, 2025 – In a move signaling a shift toward proactive cybersecurity, Leidos, a leading government IT and defense contractor, has partnered with data infrastructure innovator VAST Data and technology giant NVIDIA to deliver a new, AI-driven cyber defense model. The collaboration aims to move beyond reactive security measures and tackle the ever-growing volume and sophistication of cyber threats facing federal agencies and critical infrastructure.

The Rise of Agentic Cybersecurity

The partnership focuses on what industry experts are calling “agentic cybersecurity” – a proactive approach leveraging artificial intelligence to autonomously detect, analyze, and respond to threats in real-time. This concept addresses a critical challenge in modern cybersecurity: the sheer volume of security alerts that overwhelm human analysts. “The amount of data security teams have to sift through is crippling,” notes one cybersecurity professional who requested anonymity. “You need AI to filter out the noise and prioritize what really matters.” The solution developed by Leidos, VAST Data, and NVIDIA intends to do exactly that.

VAST Data’s core contribution is its AI data platform, built on a unified architecture that integrates storage, database, and compute. This platform provides the foundation for processing massive datasets generated by network traffic, system logs, and other security sources. Leveraging NVIDIA’s AI computing power, particularly the Morpheus cybersecurity framework and BlueField DPUs, the system is designed to analyze this data at unprecedented speeds. “The combination of VAST Data’s data infrastructure and NVIDIA’s AI capabilities creates a powerful engine for threat detection,” explains an industry analyst familiar with the technology. “It allows you to identify anomalies and predict attacks before they can cause damage.”

A Shift to Proactive Defense

The traditional cybersecurity model is largely reactive, relying on identifying and responding to threats after they have breached defenses. This approach is increasingly inadequate in the face of sophisticated attackers and the growing complexity of IT environments. The Leidos, VAST Data, and NVIDIA solution aims to flip this model on its head by proactively identifying and neutralizing threats before they can cause harm.

“The key is to move beyond signature-based detection and embrace AI-driven anomaly detection,” explains a security architect involved in the project. “By learning the normal behavior of a system, you can identify deviations that may indicate malicious activity.” The system leverages NVIDIA’s Morpheus framework to accelerate the analysis of vast data streams, enabling real-time threat detection and automated response. BlueField DPUs play a crucial role by offloading security functions from the main CPU, improving performance, and enhancing security. This hardware-accelerated approach ensures that security processes do not become bottlenecks in the system.

Implications for Federal Agencies

The partnership carries significant implications for federal agencies responsible for protecting critical infrastructure and sensitive data. Government agencies are increasingly targeted by cyberattacks, ranging from ransomware to espionage. These attacks can disrupt essential services, compromise national security, and inflict significant economic damage. Leidos’ deep expertise in government contracting and systems integration is crucial for deploying and managing this complex solution within federal environments.

“Federal agencies are facing a perfect storm of challenges,” says one government IT professional. “They are under constant attack, facing a shortage of skilled cybersecurity professionals, and struggling to keep up with the rapid pace of technological change.” This new collaboration addresses these challenges by providing a scalable, automated, and AI-driven security solution that can augment existing security teams and improve overall security posture. The platform’s ability to analyze years of telemetry data in real-time is particularly valuable for identifying long-term threats and understanding attacker tactics. The potential cost savings from automating security operations are also a significant benefit for government agencies operating under tight budget constraints.

The collaboration is also a testament to the growing importance of public-private partnerships in addressing complex cybersecurity challenges. By combining the strengths of leading technology companies and government contractors, agencies can access cutting-edge innovation and accelerate the deployment of advanced security solutions. The system will be demonstrated at NVIDIA’s GTC DC event, providing a platform to showcase its capabilities to government officials and industry experts.