- Engaging with the Cybersecurity Executive Order Revision: We are excited to share that we are proactively engaging in the upcoming revision of the Cybersecurity Executive Order and anticipate its release in the next few months.
- AI Red Teaming White Paper: Following a series of collaborative meetings with our ecosystem at the White House in July, we developed a comprehensive paper on AI red teaming, which we are pleased to share as a valuable resource. This paper has been shared with government agencies and NIST, and we are eager to continue engaging on the development of relevant AI red teaming guidelines.The paper explores AI red-teaming practices, focusing on definitions, objectives, and reach. It discusses the importance of red teaming in evaluating AI models' resistance to attacks through realistic adversarial simulations, especially for generative AI deployments. The paper highlights the necessity of combining automated tools with human expertise for comprehensive security assessments. It also emphasizes the need for up-to-date intelligence on threat actor behavior and the ethical considerations in red teaming, particularly concerning sensitive areas like child safety. We remain available for any questions and look forward to further discussions on this important topic.
- Engaging on NIST SP 1800-35: Implementing a Zero Trust Architecture: We are actively engaging and commenting on the NIST Special Publication 1800-35, focusing on implementing a Zero Trust Architecture, with feedback due on September 30th. We welcome any feedback or comments from our partners they would like us to address in our submission.
.png)
