Global summit introduces Agentic AI focus group for accountability

The International Telecommunication Union launched a specialised Agentic AI focus group on Thursday to establish safety standards for autonomous digital systems. The United Nations agency for digital technologies announced the creation of the ITU Focus Group on Trust and Identity for Humans and Agentic AI during the AI for Good Global Summit in Geneva, Switzerland. The international initiative aims to make independent artificial intelligence systems trustworthy and accountable as these advanced agents increasingly take on complex daily tasks once reserved structurally for humans.

An official announcement published on the website of the technology body detailed how these independent software agents differ fundamentally from earlier generative tools by acting on behalf of users without requiring direct human input. The systems independently manage diverse responsibilities ranging from personal scheduling and purchasing to intricate corporate workflows. However, this operational independence created significant structural concerns for the international regulatory community regarding potential impersonation and unauthorised decisions in sensitive areas like financial transactions or critical infrastructure.

The newly formed Agentic AI focus group intends to convene a diverse variety of global experts possessing technical, policy, and legal expertise. The collaborative panel will develop specific tracking mechanisms to maintain the identification, trustworthiness, and human-in-the-loop control of autonomous agents. Focus Group Co-Chair Debora Comparin explained that as virtual agents will soon negotiate, transact, and make critical choices for people, a common basis of international understanding is necessary to determine who these entities are and whether they deserve trust.

The expert project represents a major component of a larger accountability campaign introduced during the annual technology conference. Focus Group Co-Chair Amir Banifatemi summarised the initiative by noting that identity establishes exactly who the digital actor is, while trustworthiness defines how the actor behaves during operations. ITU Secretary General Doreen Bogdan-Martin confirmed that the work of the new investigative panel forms a key part of the core themes guiding the global summit this year.