NIST launches AI Agent Standards Initiative as autonomous AI moves into production – News

The U.S. National Institute of Standards and Technology has launched the AI Agent Standards Initiative, a new program aimed at developing technical standards and guidance for autonomous artificial intelligence agents as their use accelerates across enterprise and government environments.

The initiative, led by NIST’s Center for AI Standards and Innovation, is designed to address emerging interoperability, identity and security challenges associated with AI agents.

Issues around AI agents such as trust, authentication and safe integration with existing infrastructure have become more pressing as organizations experiment with agent-based systems for coding, workflow automation, research and task execution. The initiative is focused on enabling industry-led standards development while coordinating with federal agencies and international bodies, with a goal to reduce fragmentation in how AI agents communicate with external systems and with one another.

“The AI Agent Standards Initiative ensures that the next generation of AI — agents capable of autonomous actions — is widely adopted with confidence,” explains the Center for AI Standards and Innovation. “By fostering industry-led technical standards and open protocols, CAISI aims to catalyze an ecosystem where agents function securely on behalf of users and interoperate smoothly across the digital landscape while cementing U.S. dominance at the technological frontier.”

A key area of focus for the initiative is identity and authorization. Because AI agents may operate continuously, trigger downstream actions and access multiple systems in sequence, defining how such agents are authenticated, how permissions are scoped and how activity is logged and audited presents new architectural considerations.

NIST is planning to explore technical approaches that extend existing cybersecurity frameworks to agent-based systems while also examining whether new models are required.

The initiative will also encourage open protocol development to support interoperability across platforms. NIST is aiming to foster broader participation from private sector developers, research institutions and standards organizations by promoting a community-driven standards process.

The agency is working in coordination with partners including the National Science Foundation and other federal stakeholders. As part of the initiative, NIST has issued a request for information seeking public input on agent security risks, identity models and deployment considerations. The agency is also developing concept papers that outline potential technical frameworks for securing and governing autonomous AI systems.

“The industry should welcome NIST’s push for industry-led standards, but standards alone will not prevent abuse,” Gunter Ollmann, chief technology officer at offensive security services company Cobalt Labs Inc., told News via email. “Security validation, continuous testing, and adversarial simulation must evolve in parallel so organizations can understand how agents behave under attack conditions before those weaknesses are exploited in the wild.”

Photo: NIST