Adversarial Example Detection AI Security Patent

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 31st, 2026

Detected April 1st, 2026

Summary

USPTO granted NEC Corporation Patent US12592068B2 covering an adversarial example detection device, method, and program for AI security. The patent (8 claims) describes extracting features from input and comparative data using two different calculation methods to detect adversarial attacks on machine learning systems.

View original document View source feed page

What changed

NEC Corporation received granted Patent US12592068B2 for technology that detects adversarial examples—inputs crafted to deceive machine learning models. The invention uses a dual-feature extraction approach where first and second feature extraction units apply different calculation methods to input and comparative data, with a determination unit identifying whether either dataset represents an adversarial example.

Patent grants do not create compliance obligations for third parties. Technology companies developing or deploying AI/ML systems should consider this intellectual property when conducting freedom-to-operate analyses or evaluating potential licensing needs for adversarial robustness technologies.

Source document (simplified)

← USPTO Patent Grants

Adversarial example detection device, adversarial example detection method, and program

Grant US12592068B2 Kind: B2 Mar 31, 2026

Assignee

NEC CORPORATION

Inventors

Takuma Amada, Kazuya Kakizaki, Toshinori Araki

Abstract

An adversarial example detection device includes a first feature extraction unit configured to extract first features with respect to input data and comparative data in a first calculation method, a second feature extraction unit configured to extract second features with respect to the input data and the comparative data in a second calculation method different from the first calculation method, and a determination unit configured to determine whether or not at least one piece of the input data and the comparative data is an adversarial example through calculation using the first features and the second features.