- World extends AgentKit to authenticated AI agents using a global ID.
- AI agents can perform online actions on behalf of authenticated users.
- The system aims to block bots while supporting reliable systems.
The world is expanding access AgentKitits a framework designed to create human-validated AI agents and allow people to interact with these agents National ID.
The system enables AI assistants to assist users online while verifying their identity through the global Internet.
The development comes as AI assistants increasingly perform online tasks such as shopping, booking, browsing websites, and interacting with digital services.
This growth has created difficulties for businesses in distinguishing between agents representing real users and bot networks.
AgentKit is a solution to this problem by connecting AI agents to World ID, allowing websites and apps to verify whether an agent represents a unique person.
The system was designed to facilitate the distribution of services while maintaining the security associated with authentication and user control.
How AgentKit integrates AI agents to verify their identity
To start using AgentKit, people need a valid World ID, access to the World App, and an AI agent, including tools like Claude Code, Codex, Cursor, Hermes, or OpenClaw.
Users plug in their personal credentials through World’s ToolRouter interface, generate an API key, and connect their AI assistant within minutes.
Once connected, the agent can connect to services that support AgentKit and perform tasks on behalf of the user.
The system is designed to allow people to provide digital services to AI assistants while maintaining control over what is associated with a verified identity.
According to the description of the framework, the system is designed to ensure that the work of AI remains in use by a real person and not by unknown or automated systems.
The demo shows real-world scenarios
The technology was recently featured in a limited edition of 500 “Human in the Loop” helmets available to ID holders worldwide.
In the demonstration, AI assistants achieved drop-off, proper authentication, check-in at the front of the store, and completed purchases on behalf of users while maintaining a one-item limit per person connected to verified credentials.
All 500 hats were claimed by authentic individuals in several countries, including the United States, Germany, Japan, and the United Kingdom.
The demonstration was used to demonstrate how AI agents can handle real-world situations while maintaining barriers designed to reduce violence.
The model shows how businesses can allow AI assistants to complete tasks on behalf of users while still preventing bot networks from being used.
Creating a reliable part of the sponsor’s economy
As more services integrate AgentKit, World aims to create what it describes as the most reliable part of the emerging economy.
The goal is to enable AI agents to interact and communicate online while being accountable to the people they represent.
The system is designed to support a growing number of applications where AI agents work autonomously but within the established knowledge and consent of users.
This includes commercial services and a wider range of digital services.
The Global Project was created by Sam Altman, Max Novendstern, and Alex Blania, and aims to provide evidence of personal, economic and social connections in the age of AI.
The company says AgentKit is part of its broader efforts to help verify identity in an environment where AI agents are increasingly autonomous on Internet platforms.
