US 12,284,212 B2
Defensive deepfake for detecting spoofed accounts
Kelly Renee-Drop Keiter, Waxhaw, NC (US); Michael Young, Davidson, NC (US); Kyle Mayers, Charlotte, NC (US); and Sanjay Lohar, Charlotte, NC (US)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on May 16, 2024, as Appl. No. 18/665,672.
Application 18/665,672 is a continuation of application No. 17/885,764, filed on Aug. 11, 2022, granted, now 12,063,249.
Prior Publication US 2024/0305666 A1, Sep. 12, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 9/40 (2022.01); G06Q 50/00 (2012.01); G06V 40/16 (2022.01); G06V 40/30 (2022.01); G06V 40/40 (2022.01); G06V 40/50 (2022.01)
CPC H04L 63/1483 (2013.01) [G06Q 50/01 (2013.01); G06V 40/171 (2022.01); G06V 40/172 (2022.01); G06V 40/33 (2022.01); G06V 40/40 (2022.01); G06V 40/50 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A defensive deepfake computer program product, the computer program product comprising executable instructions, the executable instructions stored on a non-transitory computer-readable storage medium, when the executable instructions executed by a processor on a computer system cause the processor to:
gather two or more verified images of a target individual;
analyze, through one or more artificial intelligence/machine learning (“AI/ML”) algorithms, the two or more verified images to determine one or more unique facial characteristics of the target individual;
create, through one or more AI/ML deepfake algorithms, one or more deepfake images of the target individual;
store the one or more unique facial characteristics, and the one or more deepfake images in a database;
search a network for one or more social media accounts that include the one or more unique facial characteristics; and
determine when the one or more social media accounts are illegitimate by comparing the one or more social media accounts with the one or more deepfake images.