World, the iris-scanning identity project co-founded by OpenAI CEO Sam Altman, introduced new World ID tools aimed at banks and financial firms that need to confirm a real person is on the other end of a video call. The company is pitching the system as a defense against deepfake fraud and against AI agents that act on behalf of verified users.
The announcement points to a 2024 case at engineering firm Arup, where an employee transferred $25.6 million after joining a video call in which every other participant turned out to be a deepfake.
How the call verification works
Zoom participants can now confirm they are the same Orb-verified human tied to their account. Verified users show up with a small blue badge, and companies decide when to require that badge on a call. The check is meant to give counterparties a way to tell a live human from a synthetic face.
World ID for agents takes a different angle. It lets AI agents carry proofs delegated by people. Chief Product Officer Tiago Sada compared the arrangement to a power of attorney, with each user able to assign credentials to only one trusted agent at a time.
Bain, Pantera and VanEck join as design partners
Initial design partners include Bain Capital, OpenExchange, Outtake, Pantera Capital and VanEck Funds. A company spokesperson said financial firms will keep Orb devices in their offices for employees who have not yet had their irises scanned.
The Orb is football-sized. It encrypts each scan so World cannot read it, and the company says no central database of individuals exists. World ID comes in three tiers: the lowest relies on government documents, while the top tier, Proof of Human, matches a selfie against the Orb record.
The privacy question Sada keeps fielding
Iris scanning tends to draw a strong reaction, and Sada has heard the skepticism before. He compared the wariness to early reactions to Apple's Touch ID and facial verification. "People understand you don't have to give Sam or any of us your photo," Sada said. "Everything is open source and auditable."
That argument will get tested as World pushes into regulated financial firms, where biometric data is already subject to strict handling rules. The company's claim that it cannot read the encrypted scans is central to its pitch.
The bot-blocking alternative World wants to displace
Some large companies have gone the other way. Amazon and JPMorgan have started blocking bots outright. Sada said that tactic would limit companies in agentic commerce, where software agents are expected to transact on behalf of people rather than just browse.
Cloudflare announced an agent ID tool and a wallet for agents in August. Its Chief Strategy Officer Stephanie Cohen tied the launch to bots being treated as a business opportunity rather than a nuisance to be filtered out.
World's native token WLD was trading at around $0.49 at the time of writing, down 0.7% over 24 hours and more than 11% over seven days, though up over 14% over 30 days. It remains nearly 96% below its all-time high of $11.74. In June, WLD fell more than 10% to $0.34 even after a $52.5 million funding round meant to expand World ID infrastructure.
The new tools are available to the initial design partners first. How quickly other banks and financial firms adopt the blue badge on Zoom calls, and whether customers accept an Orb scan as the price of proving they're human, are the questions World now has to answer in production.
