Sen. Mark Warner, D-Va., and Reps. Donald Beyer, D-Va., and Gerry Connolly, D-Va., joined MITRE this week at the grand opening of the company’s AI Assurance and Discovery Lab.
The new lab – which is based at MITRE’s McLean, Va., headquarters – will provide a space for government agencies to discover and mitigate critical risks in their AI-enabled systems.
“The emergence and evolution of new technologies like AI have led us to the brink of an exciting future,” said Rep. Connolly. “As we seek to embrace AI safely and with the proper guardrails in place, ventures like MITRE’s AI Assurance and Discovery Lab will be invaluable to our efforts across government.”
The new AI Assurance and Discovery Lab will evaluate AI-enabled systems intended for use in consequential applications including national security, healthcare, and transportation, MITRE said.
Government agencies can also use the lab to inform requirements development for new AI-enabled systems, create and evaluate proposed risk mitigation plans, and develop long-term AI assurance strategies for their organizations.
“Government use of AI will have consequential impacts on the nation and world, in areas such as efficient transportation, effective healthcare, and strengthened national security,” said Charles Clancy, MITRE senior vice president and chief technology officer. “However, in adopting these systems, we also need to mitigate risks. Providing an independent assessment of the security, safety, and efficacy of AI systems will play a critical role toward helping government and business integrate new technology in benefits processing, intelligence analysis, autonomous vehicles, and more.
“This lab will demonstrate a repeatable engineering approach and infrastructure that could serve as a blueprint for a national network of AI assurance facilities,” Clancy said.
MITRE said the lab features configurable space for risk discovery in simulated environments, AI red teaming, large language model evaluation, human-in-the-loop experimentation, and assurance plan development. The lab’s physical space can be tailored and instrumented for specific mission scenarios and workflows.
Federal agencies – and soon private companies – can bring AI-enabled systems to the lab to explore potential risks including whether they perform effectively, consistently, and safely in real-world contexts, the company said. MITRE will also use the lab to evaluate factors such as whether systems are secure and free from harmful bias and allow users to control how their information is used.
“For federal agencies and private companies, this new lab delivers an objective, independent analysis to complement and validate their own testing and evaluation of AI-enabled systems, helping establish confidence for end users,” said Cedric Sims, MITRE senior vice president. “Ensuring that AI provides transformational benefits for our government sponsors is a top priority as we solve problems for a safer world.”
