United States, UK, and others Unveil International Agreement on Keeping AI "Secure by Design"

On November 27, 2023, the United States, the U.K. and 16 other countries unveiled what one U.S. official described as the “first detailed international agreement” on keeping AI safe from rogue actors. The agreement is nonbinding and includes general recommendations for designing, developing and deploying AI with security in mind, with recommendations including maintaining an “awareness of relevant security threats and failure modes,” developing “incident management procedures,” assessing the “appropriateness of your AI-specific design choices” and more. On the release of the guidelines, director of CISA Jen Easterly said, “This is the first time that we have seen an affirmation that these capabilities should not just be about cool features and how quickly we can get them to market or how we can compete to drive down costs . . . [the guidelines are] an agreement that the most important thing that needs to be done at the design phase is security.”

https://www.theguardian.com/technology/2023/nov/27/ai-safety-pact-us-uk