Clearview AI has built and tested an unreleased “analyst assistant” that would take leads from a face-recognition search and automatically scrape the open web for employers, aliases, associates, social accounts, and other personal details, according to interface code and files reviewed by WIRED. The prototype, called InquiryIQ, sits on top of Clearview’s face-matching database — which the company now says holds well over 70 billion images and is used by more than 2,000 U.S. law enforcement agencies — and is designed to assemble what Clearview calls a “Candidate Graph” of possible identities and connections.
WIRED found InquiryIQ in files Clearview’s login page sends to browsers before sign-in, the same method it used last month to surface a Flock Safety AI search tool.
The interface text says supplying age, gender, and race can help the system make “smarter decisions” while it runs web and image searches, browses pages, and applies face recognition to photos it encounters. One of the models Clearview tested comes from xAI (now SpaceXAI after the February merger), the company behind Grok; Amazon Bedrock also appears as a model option.Clearview tells WIRED InquiryIQ has never been pitched or shipped to customers, has never been used by law enforcement, and is not currently planned for release. CEO Amos Kyler says the selector was for engineers comparing performance in testing, not for officers, and describes the tool as a way to automate searches detectives already perform — with humans responsible for verifying leads. Privacy scholars warn that collapsing days of digital investigation into minutes removes friction that once constrained mass profiling, and generative models can return different results from the same inputs.
This brief covers the September 10 WIRED exclusive and corroborating Biometric Update coverage; it does not allege that InquiryIQ has been fielded to agencies.