On November 8, General Dynamics Information Technology (GDIT), a business unit of General Dynamics, announced that it has joined the National Institute of Standards and Technology's (NIST) Artificial Intelligence (AI) Safety Consortium.

As part of the consortium, GDIT will collaborate with government, industry and academic leaders to develop new standards that address the complexity and risks associated with AI systems. This group, housed under the U.S. AI Safety Institute, is tasked with implementing key components of the 2023 AI executive order, including risk management for generative AI, evaluating capabilities, and creating guidelines for safety and security.

GDIT, which has supported AI implementations across the U.S. government for over two decades, will provide technical expertise to support the development, deployment, transparency and evaluation of AI technologies. The company will leverage its Luna AI Digital Accelerator to assess large language models, analyze training data and identify risks that could undermine the accuracy and reliability of AI outputs.

"At GDIT, we are deeply committed to using AI to drive innovation while ensuring its safe, secure and trustworthy adoption," said Ben Gianni, GDIT's senior vice president and chief technology officer. "For more than two decades, we have harnessed this technology to advance agencies' missions responsibly. We look forward to drawing on this experience and partnering with NIST and other consortium members to advance its goals."

GDIT's participation in the NIST AI Safety Consortium builds on its decades of experience delivering AI solutions that enhance mission-critical operations across the federal government. The company has deployed machine learning models for fraud detection in healthcare, developed advanced data analytics for aviation safety, and implemented AI to advance national security and intelligence missions.

Source: GDIT