World Network recently announced its latest upgrade to World ID, positioning it as a full-spectrum solution for cryptographic human proof verification in the age of AI. The renewed protocol, announced at the Lift Off event in San Francisco, securityPortability and usability for both businesses and individual users.
According to the update World NetworkThis marks World ID’s biggest evolution since its inception, providing private, verifiable confirmation that a real, unique person stands behind online actions without compromising anonymity or collecting personal data.
Key technical improvements include key rotation, account recovery, multi-key support authenticationand proper session management.
These features elevate the protocol to enterprise-level standards, making it reliable for high-volume deployments.
New disposable overriders further strengthen anonymity protection, while an open-source SDK allows any application to act as a World ID authenticator.
Complementing the protocol changes, World introduced the World ID app, a dedicated mobile experience now in public beta so users can seamlessly manage and apply human proof credentials across the web.
The upgrade comes alongside a wave of high-profile integrations that expand human proof into critical areas of the internet.
World partners with leading platforms to combat bots, enable secure agency workflows, and restore trust in everyday business. digital interactions.
The agent is on the network, where artificial intelligence agents Handling browsing, API calls, purchases, and increasingly complex tasks, World ID provides core primitives like agent delegation, human-in-the-loop verification, and brokered commerce.
The browser base integrates World ID, allowing agents to receive verified traffic status, bypass anti-bot measures, and reduce friction during web automation.
Exa offers free API Quotas for agents powered by World ID enable experimentation and protect against abuse.
Vercel’s open source Workflow SDK now supports the “human in the loop” step, allowing developers to add verifiable human approval (with full audit trails) to any delegate workflow via a simple npm package.
Okta is developing a new product, Human Principle, that allows API creators to validate a human policy behind an agent and enforce policies; World ID will serve as a key integration Common for privacy-preserving authentication.
Powered by World’s AgentKit, these tools enable agents to carry cryptographically authenticated human support for everything from transactions to sensitive actions.
The new World ID for business applications introduces “people continuity,” a missing piece in traditional applications. security systems that only authenticate devices or accounts.
Zoom and DocuSign are integrating technology to bring verifiable humanity to professional workflows.
For example, DocuSign will allow signers to cryptographically prove that they are real people (not bots) when approving identity attributes during document execution strengthen the overall trust model in an agent-driven environment.
On the consumer side, World Identity embeds human evidence into everyday platforms to foster real interactions.
Match Group’s Tinder now offers global World ID verification after successful implementation Japan Pilot offers users a privacy-first way to verify the real people behind the profiles and reduce fake accounts.
In gaming, Razer’s World ID-verified Razer ID sets a new standard against bots, while Mythical Games is applying it to player-owned economies where separating humans from automated players is vital.
Reddit is investigating World ID privacy-maintaining controls on suspicious bot accounts.
For live events, WorldThe new Concert Kit tool allows artists to book tickets only for verified people; Starting with the Bruno Mars World Tour (featuring DJ Pee Wee/Anderson .Paak), it provides VIP access at select stops and protects against scalper bots. By combining a production-ready protocol with these strategic partnerships, World Network focuses on accelerating the adoption of human evidence at scale.
