World is launching a developer toolkit designed to help websites verify that AI agents are acting on behalf of real people.
Launching in beta today, AgentKit integrates World’s World ID identity system with x402, an open protocol started by Coinbase and Cloudflare.
The system allows AI agents to carry cryptographic proof that they represent a unique human as they interact with websites, APIs and online services.
The release aims to address a persistent problem: AI agents are taking over tasks previously handled directly by users, including booking reservations, purchasing concert tickets, accessing APIs, and comparing prices.
“Right now there are a lot of services where agents can spam them – social platforms or things like ticketing,” DC Builder, a research engineer at the World Foundation, told me. Declutter. “Think of Ticketmaster: if you give an agent the ability to book tickets, you can generate about 100,000 tickets. “Even if they have the money to pay, it’s not a great user experience for people competing with bots.”
Platforms are increasingly struggling to distinguish legitimate automated activity from large networks of bots.
Earlier this month, a federal judge issued an injunction against AI developer Perplexity, blocking the company’s Comet browser from making purchases on Amazon on behalf of users.
AgentKit allows people who have verified their identity using World’s orb device to delegate their World ID to an AI agent.
Originally known as Worldcoin, World launched by 2023, with the aim of providing a global “proof of personhood” digital ID and cryptocurrency to help distinguish humans from bots and expand access to the digital economy.
According to World, the network includes nearly 18 million verified individuals in more than 160 countries. Although the sphere rewards users with crypto in the form of Worldcoin (WLD) tokens, according to Builder, tokens are not necessary to use the AgentKit.
Once the AI agent is linked to a user’s World ID, it can cryptographically prove that it represents a human without revealing the person’s identity.
By expanding to the x402 protocol, sites can request proof in addition to or instead of micropayments before granting access to services or APIs.
“What this allows you to do is program against knowing whether the request is coming from a human or an agent – or an agent associated with a human,” Erik Reppel, head of engineering for Coinbase’s developer platform, told me. Declutter. “As a seller, you can just say, ‘There’s no evidence of human presence attached to this, so I’m going to decline the payment.’”
Platforms can then enforce limits or allocate resources based on the number of unique people interacting with a service, rather than the number of automated agents involved.
“With a proof of humanity, you at least know that the account is controlled by one person and that there are not thousands of accounts all trying to buy something,” Builder said. “But it is not necessary for the identity part to contain information about the individual themselves – we are purely anonymous in the proof-of-human protocol.”
“I think there’s a big part of the internet where it doesn’t really matter whether it’s a human or an agent, and a big part of the internet where it really, really matters a lot whether it’s a human or an agent,” Reppel added. “What we need are robust, open ways to understand what’s what: to be able to tell when you’re talking to an AI, a human, or the AI of a specific human.”
In February, Coinbase introduced a wallet for AI agents on its Base network, designed to let automated software handle payments while keeping private keys isolated in trusted execution environments.
In addition to transactions, Builder said the technology could help maintain human interaction online as automated agents spread across the Internet.
“People go to social networks to have that human connection,” Builder says. “If they want to communicate with an agent, they go to ChatGPT, Claude or another service.”