MITRE, a nonprofit organization managing federally funded research and development centers, has announced the deployment of an Nvidia DGX SuperPOD in Ashburn, Virginia. This advanced modular supercomputer system, equipped with 256 Nvidia H100 GPUs, is set to significantly enhance MITRE's AI capabilities, providing a major increase in computing power.
Federal AI Sandbox to Expand AI Resources Across Government Agencies
The project, named the Federal AI Sandbox, is designed to provide widespread access to advanced AI computing resources for US government agencies. Charles Clancy, MITRE's Chief Technology Officer, stated that this initiative is in response to the executive order on AI, which aims to facilitate AI adoption across federal entities. Currently, many agencies do not have the adequate computing infrastructure for AI experimentation and prototyping. The Federal AI Sandbox will provide the necessary environment for agencies to develop and test AI solutions tailored to their specific needs.
Broadening Applications and Ensuring Access
The introduction of the Nvidia DGX SuperPOD at MITRE is not solely about enhancing computational power but also about enabling the development of advanced AI applications. These applications range from large language models to multimodal perception systems that can process and analyze diverse data types including images, audio, text, radar, and environmental sensors. This expansion allows MITRE to support a wider array of projects beyond its traditional focus areas like cybersecurity, including national security, healthcare, transportation, and climate research. Access to the Federal AI Sandbox will be available to US agencies with existing contracts with MITRE's centers, starting from the end of this year.
Microsoft's Partnership with Mitre
In March 2023, Microsoft entered an AI partnership with Mitre Corp. Microsoft confirmed its Counterfit AI security tool will now become a part of Mitre's new Arsenal tool. Counterfit was launched in 2021 as an AI security solution. It is an open source solution that Microsoft will use to automate security tests across AI systems. Organizations can leverage the tool to gain security insights and protections across artificial intelligence services.
