The identity-tech knowledge base for AI systems and human operators.

Welcome to ID Tech’s digest of identity industry news. Here’s what you need to know about the world of digital identity and biometrics today:

Federal Judge Blocks Utah’s VPN Location Rule for Adult-Site Age Checks

A federal court has issued a preliminary injunction barring Utah from enforcing an actual-location provision for adult sites that treats users as being within the state even when virtual private networks or proxies disguise their physical origin. Following a legal challenge brought by Pornhub operator Aylo Freesites, U.S. District Judge David Barlow determined that Section 78B-3-1002(3) lacked a reasonable-efforts qualification, finding that the impossibility of infallible geolocation would impermissibly burden interstate commerce beyond state borders while leaving Utah’s baseline requirement for reasonable age-verification intact.

YouTube Plans Voice Detection for Creator Likeness Checks

YouTube is expanding its automated impersonation safeguards by integrating voice detection into its creator likeness tool later this year, combining audio analysis with existing facial recognition to scan newly uploaded videos for unauthorized synthetic imitations. Building on the opt-in system currently available in YouTube Studio—which requires eligible adult creators to submit government identification and video selfies to generate reference templates—the multi-signal system will alert enrolled creators to potential matches, support privacy and copyright takedown requests, and eventually expand directly into the YouTube mobile application.

Delaware to Require ID and Selfie Checks for Prison Video Visits and Messaging

The Delaware Department of Correction has announced that community members using the GettingOut platform must complete government ID and selfie verification before sending digital messages or scheduling remote video visits with incarcerated individuals beginning September 28. Under the new protocol outlined by Commissioner Terra Taylor, users accessing the app or web portal must upload two-sided images of a valid government-issued credential alongside a live selfie photograph, establishing a one-time verified profile designed to prevent exploitation, curtail illicit contact, and mirror security screening standards enforced during physical prison visitations.

NHS Plans AAL2 Passkey Update for Self-Claimed Care IDs

NHS England is scheduled to release Care Identity Management version 9.1 on September 24, introducing AAL2 passkey authentication for self-claimed Care IDs to bolster access controls for clinical and administrative healthcare personnel. Operating under the government’s GPG45 identity standards, the targeted infrastructure update applies phishing-resistant FIDO authentication to restricted-access workforce journeys without expanding unverified clinical system privileges, while incorporating optimized organization-search logging within the Find your Registration Authority tool and backend reliability enhancements without operational downtime.

Facephi Earns Canadian Trust-Framework Certification for Identity Proofing

Biometric identity verification provider Facephi has secured Pan-Canadian Trust Framework LOA2 certification for its Verified Person component following an independent third-party audit overseen by the Digital ID and Authentication Council of Canada (DIACC). The designation, which formally places the company on DIACC’s public Trusted List, independently validates Facephi’s document verification, biometric onboarding, and fraud mitigation controls against national standards as the Spanish firm expands its North American operations through its British Columbia subsidiary.

German Privacy Regulators Warn Against Smart-Glasses Face Recognition in Public

Germany’s conference of data protection authorities, the Datenschutzkonferenz, has issued a formal legal warning stating that smart-glasses facial recognition in public spaces is impermissible under the General Data Protection Regulation following a Hamburg regulatory inquiry into Ray-Ban Meta AI smart glasses. The regulatory body emphasized that recording bystanders without clear advance notice, operational transparency, and explicit consent violates European privacy statutes, finding that subtle LED recording indicators are insufficient to ensure compliance while calling on hardware manufacturers to embed privacy-by-design safeguards into wearable camera devices.

Apple Details Added Image Processing for Face ID on iPhone 18 Pro

Apple has updated its official platform security documentation detailing how the iPhone 18 Pro and Pro Max leverage supplemental image processing for Face ID to sustain biometric matching performance. Under the revised architecture, additional image capture data supplements the depth map and infrared signals generated by the TrueDepth camera, routing directly into the hardware-isolated Secure Exclave compartment where sensor inputs remain isolated and inaccessible to iOS, third-party applications, device owners, and Apple servers before matching templates interact with the Secure Enclave.

EFG Hermes Opens Mobile Onboarding for Dubai Financial Market Investors

Investment bank EFG Hermes UAE has deployed an end-to-end digital onboarding service within its EFG Hermes ONE trading application, allowing domestic and foreign users to secure a Dubai Financial Market investor number and open brokerage accounts without physical branch visits or paper documentation. Developed in coordination with DFM Chief Operating Officer Khalifa Rabba and Dubai Central Securities Depository CEO Afra Al Suwaidi, the streamlined registration targets an exchange environment where foreign nationals accounted for 71.4 percent of new investors and 51 percent of total trading value—generating AED 119.5 billion—in the first half of 2026.

GMO GlobalSign and VESS Labs Test Credentials for Cross-Company AI Agent Transactions

Japanese trust provider GMO GlobalSign Holdings and decentralized identity specialist VESS Labs have completed a joint proof-of-concept demonstrating verifiable credentials for AI agent transactions, allowing autonomous software agents developed by separate entities to validate enterprise identity and purchasing authority. The trials completed 281 automated test cases linking a GMO GlobalSign purchasing agent to a VESS Labs selling agent using signed verifiable credentials and cryptographic human authorizations to prevent token reuse, intercept altered orders, and reject revoked credentials without relying on bilateral pre-shared allowlists.

–

By the ID Tech Editorial Team