ITU Opens Standards Work on Identity and Trust for Humans and AI Agents

The International Telecommunication Union has created a group to develop frameworks for establishing who an AI agent is and whether its behaviour can be trusted, extending the machinery of digital identity standards to software that acts on people’s behalf.

The ITU, the United Nations agency for digital technologies, announced the Focus Group on Trust and Identity for Humans and Agentic AI at its AI for Good Global Summit. The premise is that as AI systems move beyond assistive tools into autonomous agents that plan, transact, and act with growing independence, they introduce risks ranging from agents impersonating people or organisations to taking unauthorised actions across interconnected systems. Identity systems establish who is acting, the ITU argues, while trustworthiness determines whether that actor is reliable, and both are needed before agents can safely handle tasks such as executing financial transactions or operating critical infrastructure under meaningful human control.

The group is open to technical experts along with specialists in policy, law, and regulation, and is chartered to produce common terminology, reference architectures for identity, trust, agent discovery, and interoperability, trust frameworks and lifecycle assurance models, interoperability mechanisms for digital identity and credentials, security criteria and benchmarks for continuously assessing AI agents, and a standardisation roadmap. It reports to ITU-T Study Group 17, the union’s expert group for security standards, and will hold its first meeting in Paris in November, with a second in Geneva in January 2027.

“AI agents will soon negotiate, transact and make decisions on our behalf,” said ITU Secretary-General Doreen Bogdan-Martin. “Before that future becomes reality, we need common international foundations that establish who these agents are, when they can be trusted, and how people will remain in control.”

The standards effort arrives as vendors improvise their own answers to the same question, with World extending proof-of-human checks into agentic workflows with Okta and Vercel and a wave of AI-agent identity tooling from CrowdStrike, SailPoint, Beyond Identity, Akamai, and Saviynt.

ITU focus groups are designed to accelerate the development of new global standards, and past groups have tackled AI in health, autonomous networks, and self-driving cars. A larger leadership team for the new group is being finalised following consultation, with membership open to any interested expert.

Sources: ITU

–

By the ID Tech Editorial Team