Tech Giants Open Frontier AI Models to National Security Tes…

Source Domain: www.pymnts.com

Google DeepMind, Microsoft, and xAI have agreed to share innovative AI models with the Center for AI Standards and Innovation (CAISI) for national security testing before public release.
These companies signed agreements with CAISI, part of the Department of Commerce’s National Institute of Standards and Technology (NIST).
They join Anthropic and OpenAI, who previously signed similar agreements with the predecessor to CAISI, now known as the U.S. Artificial Intelligence Safety Institute.
CAISI facilitates testing, collaborative research, and develops best practices for the AI industry.
The agency has completed over 40 pre-deployment evaluations of AI models, including some that have not been publicly released.
The move comes amid growing concerns about AI posing threats to cybersecurity, underscored by a recent step change in the power of large language models.
Treasury Secretary Scott Bessent emphasized the necessity for the U.S. to stay ahead in AI technology, highlighting its potential cybersecurity risks and benefits.
Specifically, Anthropic recently allowed select partners early access to its model for defensive cybersecurity efforts to identify and mitigate potential vulnerabilities.
The World Economic Forum identified AI as a critical factor influencing cybersecurity strategies, expected to be highly influential in both defense and offensive capacities.

You may have missed