Eyeing China’s Growth, NIST Launches New Standards Initiative for AI Agents

Washington is seeking to place artificial intelligence (AI) agents at the center of American technology dominance. On February 17, the National Institute for Standards and Technology (NIST) launched the AI Agent Standards Initiative, a project intended to shape the development of AI agents. Unlike traditional AI tools that respond to questions, AI agents can independently plan, make decisions, and take real world actions, like writing code, booking a trip, or sending emails. Their ability to take autonomous action on behalf of their owners presents a far more significant safety challenge, and their rapid evolution necessitates constant reconsideration of access controls and the protection of sensitive information. At the same time, their versatility and scalability offer unparalleled opportunities for innovation.

The program marks Washington’s growing interest in guiding agentic AI use to spur domestic public adoption as Chinese firms aggressively roll out new agentic models in a bid to seize global market share.

NIST Looks To Secure American Innovative Advantage

The NIST initiative intends to develop industry-wide standards for AI agents and facilitate their adoption across a broad range of sectors while easing compliance costs for developers by supporting a nation-wide support network. NIST also seeks to pivot these standards to be adaptable for international technical bodies — forums Chinese entities often dominate — to allow the United States to shape the global development of AI agents.

Likewise, the new initiative builds on other existing NIST efforts to inform security guidelines for AI agents. In January, the agency introduced a platform for AI firms and security researchers to propose safeguards against threats from model manipulation and adversarial use of AI agents. China, Russia, and other adversaries have employed such techniques to degrade American AI models while refashioning them to serve their own objectives, such as spreading propaganda.

Chinese AI Firms Continue To Pivot to AI Agents

Chinese firms have increasingly pivoted toward launching AI agents to maintain pace with their American competitors. Over the past month, multiple Chinese AI firms, including Alibaba, ByteDance, and Zhipu AI, have launched models compatible with agentic capabilities to capture an emerging sector and cement their edge within a competitive domestic market. This surge has also influenced the American market, with Meta purchasing Chinese agentic AI firm Manus in December and Alibaba explicitly designing its latest products to integrate with Western-produced agents.

The Chinese military has also sought to procure AI agentic products to further its concept of “intelligentized warfare.” While Chinese officers have relied on large language models produced by firms such as DeepSeek to perform non-combat functions such as logistic tasks and hospital administration, Beijing remains interested in using agentic models to control swarms of autonomous weapons, such as those featured in its October 2025 amphibious landing exercises.  

Washington Should Prioritize Trust To Drive Public Adoption

To ensure Americans engage with agentic AI as productively as possible without endangering national security or undermining public trust, Washington must lead the creation of a framework that outlines best practices for their deployment. NIST should establish a tiered risk framework that gives all stakeholders a common language for deployment standards. The institute must prioritize transparency, auditability, and attribution to ensure that AI agents deliver real value without compromising national security. In so doing, NIST can position the United States to compete with China on its own terms.

Leah Siskind is a research fellow at the Center on Cyber and Technology Innovation (CCTI) at the Foundation for Defense of Democracies (FDD), where she focuses on artificial intelligence. Jack Burnham is a senior research analyst in the China Program at FDD. For more analysis from Leah, Jack, and FDD, please subscribe HERE. Follow Leah on X @Leahsiskind. Follow Jack on X @JackBurnham802. Follow FDD on X @FDD and @FDD_CCTI. FDD is a Washington, DC-based, nonpartisan research institute focusing on national security and foreign policy.