DETECTING FALSE STATEMENT USING MULTIPLE MODALITIES

Patent №

US 9,047,253

Granted

2015-06-02

Filed 2013

Owner

CA, INC.

Lab

AI components

1

vision

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

13829454

Methods for preventing the transmission of sensitive information to locations outside of a secure network by a person who has legitimate access to the sensitive information are described. In some embodiments, in order for an end user of a computing device to establish a secure connection with a secure network and access data stored on the secure network, a client application running on the computing device may be required by the secure network. The client application may monitor visual cues (e.g., facial expressions and gestures) associated with the end user, detect suspicious activity performed by the end user based on the visual cues, and in response to detecting suspicious activity may perform mitigating actions to prevent the transmission of sensitive information such as alerting human resources personnel or requiring authorization prior to sending information to locations outside of the secure network.

VisionH04L 9/0866G06F 21/32G06V 40/20H04L 63/0861H04L 63/10H04L 63/102

AI classification

Vision1.00
Evolutionary computation0.44
Natural language0.15
Speech0.04
AI hardware0.02
Knowledge representation0.02
Machine learning0.01
Planning0.00

Ownership

CA, INC.

assignment · 300240187

Assignors

GATES, CARRIE E., SILBERMAN, GABRIEL M., VELEZ-ROJAS, MARIA C., MANKOVSKII, SERGUEI, GREENSPAN, STEVEN L.

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC