Artificial Intelligence
In support of the NIST AI program, the NCCoE is developing projects to produce practical guidance using a test platform like Dioptra to assist in managing AI risks.
Security Guidance
-
Cyber AI Profile
Guidance based on the NIST Cybersecurity Framework to address the cybersecurity risks related to AI development and use.Reviewing Comments -
Artificial Intelligence: Adversarial Machine Learning
Informing future standards and best practices for assessing and managing the security of machine learning componentsFinalized -
NCCoE Chatbot
The NCCoE seeks to understand the benefits and vulnerabilities of generative AI systems and opportunities to implement technologies to mitigate these vulnerabilities. The project is beginning to leverage generative AI to support the work of the NCCoE.Reviewing Comments -
Autonomous Vehicle Vision
The NCCoE is working on cyber assurance for autonomous vehicles by developing a public dataset and a testbed with difficult-to-handle and adversarial road/traffic conditions with the goal of improving autonomous vehicles and accelerating their safe deployPreparing Draft -
Dioptra
Dioptra is a software test platform for assessing the trustworthy characteristics of AI models.Preparing Draft -
Software and AI Agent Identity and Authorization
With the advancement of AI and software agents, systems have the capability for autonomous decision-making with limited human supervision to achieve complex goals. This increased scale and autonomy brings new opportunities as well as new risks.Reviewing Comments
Supplemental Resources
- Explore NIST's broader work in the area of Artificial Intelligence.
- Read and comment on the second draft of NIST's AI Risk Management Framework.