On December 3, 2024, the Federal Trade Commission (“FTC”) announced that it reached a settlement with IntelliVision Technologies Corp. (“IntelliVision”) to resolve allegations that the company violated Section 5 of the FTC Act by making certain claims concerning its AI-powered facial recognition software. 

The FTC’s complaint alleged, among other things, that IntelliVision made certain inaccurate or insufficiently supported claims about its facial recognition software, including with regard to its accuracy as it pertains to gender, race, and ethnicity detection and bias.

To resolve these and other allegations, the FTC and Intellivision entered into a proposed consent order that places restrictions and obligations on IntelliVision with respect to its facial recognition technology.

Among other restrictions, the proposed consent order requires that IntelliVision not make misrepresentations about the accuracy or efficacy of its technology, including concerning “the comparative performance … with respect to individuals of different genders, ethnicities, and skin tones, or reducing or eliminating differential performance based on such factors” and detecting spoofing or determining “Liveness” (defined to mean “that a living subject is present at the point of capture”).

The proposed consent order also prohibits IntelliVision from making representations about the effectiveness, accuracy, or lack of bias of its facial recognition technology or its effectiveness at detecting spoofing unless the company “possesses and relies upon competent and reliable testing that substantiates the representation at the time the representation is made.”  The order provides documentation and testing requirements to satisfy this provision.

The proposed consent order is subject to a 30-day public comment period following publication in the Federal Register.