The National Institutes of Standards and Technology is launching a new project around standards for artificial intelligence agents, with NIST positioning the project as key to advancing innovation in agentic AI.
NIST’s Center for AI Standards and Innovation (CAISI) announced the “AI Agent Standards Initiative” this week. The project aims to foster “industry-led technical standards and protocols that build public trust in AI agents, catalyze an ecosystem of interoperable agents, and disseminate their benefits to all Americans and across the world,” NIST said in a statement. release this week.
“AI agents can now work autonomously for hours, writing and debugging code, managing emails and calendars, and purchasing products, among other emerging use cases,” NIST added. “While the promise of productivity is enticing, the actual usefulness of agents is limited by their ability to interact with external systems and internal data. Without trust in the trustworthiness of AI agents and the interoperability between agents and digital assets, innovators risk facing a fragmented ecosystem and delayed adoption.”
While NIST’s press release positions the project around innovation, the initiative’s opening products are focused on security. Because AI agents can act autonomously, technology experts say they present significant safety and security concerns.
THE initiatives early results include a request for information on “AI agent safety.” The deadline to respond to the RFI is March 9.
“CAISI’s efforts to understand ecosystem perspectives on current threats, mitigations, measures, and other considerations related to agent security,” NIST wrote on its website.
NIST’s National Center for Cybersecurity Excellence also released a draft concept paper on “Identity and Authorization of Software and AI Agents.” » Comments on the document are due April 2.
The project’s goal is to explore how standards can “identify, manage, and authorize access to and actions taken by software agents, including AI agents, and provide practical guidelines for organizations to safely implement AI agents and benefit from their improved productivity, efficiency, and decision-making,” the NCCoE wrote.
The use and deployment of AI agent systems has also seen rapid growth over the past year. A recent Microsoft Cyber Pulse report found that more than 80% of Fortune 500 companies deploy active AI agents.
But as AI agents become more competent and autonomous, they can also “wreak havoc” when given free rein within tech companies, according to a recent study. paper published by the Georgetown Center for Security and Emerging Technologies.
CAISI’s work on agentic AI standards is expected to be crucial as the Trump administration and Congress consider a broader approach to AI regulation.
Michael Kratsios, director of the White House Office of Science and Technology Policy, previously said that CAISI was “a very important part of the broader AI agenda.” One of the top lawmakers in the House is develop legislation to codify the center into law.
“It is absolutely important that CAISI undertakes the work on AI standards, and that is what it is challenged to do,” Kratsios said during a House of Representatives hearing in January. “And that’s what they should be focusing on, because the high standards proposed by CAISI and NIST are the ones that will ultimately allow the proliferation of this technology across many industries.”
The NIST center is expected to play a key role in setting standards for “advanced model evaluation metrology,” Kratsios said.
“This is something that can be used across industries when they want to deploy these models,” he said. “You want to trust them so that when Americans are using them on a daily basis, whether it’s medical models or otherwise, they’re comfortable with the fact that they’ve been tested and evaluated.”
The Biden administration initially established CAISI as the “AI Security Institute.” The Trump administration renamed the center, arguing that it should focus on standards and innovation.
Copyright © 2026 Federal News Network. All rights reserved. This website is not intended for users located in the European Economic Area.