US 12,244,553 B2
System for identifying and blocking suspect electronic communications based on Artificial Intelligence
Anna Kristen Pingel Berry, Indian Land, SC (US); Shweta Ambulkar, Plainsboro, NJ (US); Benjamin Daniel Hardman, Harrisburg, NC (US); Angela Ianni, Lincoln, RI (US); Olga Kocharyan, Matthews, NC (US); Luqman Sharief, Libertyville, IL (US); and Michael Wm. Whitaker, Fort Worth, TX (US)
Assigned to BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed by BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed on Jun. 13, 2022, as Appl. No. 17/838,978.
Prior Publication US 2023/0403247 A1, Dec. 14, 2023
Int. Cl. H04L 51/212 (2022.01); H04L 9/40 (2022.01); H04L 51/08 (2022.01); H04L 51/42 (2022.01)
CPC H04L 51/212 (2022.05) [H04L 51/08 (2013.01); H04L 51/42 (2022.05); H04L 63/1408 (2013.01); H04L 63/1441 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A system for electronic mail (email) management, the system comprising:
a plurality of email clients, each email client associated with one of a plurality of users and including an electronic mailbox configured to allow a corresponding one of the plurality of users to access email associated with one or more email accounts held by the corresponding user; and
a gateway apparatus having a memory and one or more processing devices in communication with the memory, wherein the memory stores an artificial intelligence-based email management tool that implements machine learning, is executable by at least one of the processing devices and configured to:
learn, over-time, malicious email patterns from (i) malicious emails previously identified by the email management tool, (ii) continuously crawling the World Wide Web to identify data relevant to learning malicious email patterns, and (iii) feeds from threat intelligence sources,
receive email addressed to the plurality of users,
prior to delivering the email to the electronic mailboxes of the email clients, analyze the received email by, at least, applying the malicious email patterns to identify malicious email, wherein malicious email is email that attempts to wrongfully obtain data, and
prevent the malicious email from being delivered to the electronic mailbox of the corresponding one of the plurality of users, and
wherein the plurality of email clients are configured with an email reporting add-on feature that is configured to allow for a single user input for reporting an email to the email management tool that the corresponding user identifies as a suspicious email, wherein the email was previously analyzed by the email management tool and was delivered to the electronic mailbox of the email client, wherein suspicious email is email that the email recipient perceives to be attempting to wrongfully obtain data from the email recipient,
wherein, upon receipt of the suspicious email, the email management tool is configured to:
implement machine learning techniques to further analyze the suspicious email to determine whether the suspicious email is a verified malicious email,
in response to determining that the suspicious email is a verified malicious email, implement machine learning techniques to determine one or more reasons why the verified malicious email was not identified by the email management tool as malicious prior to communicating the email to the electronic mailbox of the email client.