US 12,316,654 B2
Using artificial intelligence to detect malicious upload activity
Syed Ali Bilgrami, Gilbert, AZ (US)
Assigned to Sequoia Benefits And Insurance Services, LLC, San Mateo, CA (US)
Filed by Sequoia Benefits and Insurance Services, LLC, San Mateo, CA (US)
Filed on Feb. 7, 2024, as Appl. No. 18/436,002.
Application 18/436,002 is a continuation of application No. 18/109,772, filed on Feb. 14, 2023, granted, now 11,936,670.
Application 18/109,772 is a continuation of application No. 16/916,572, filed on Jun. 30, 2020, granted, now 11,588,830, issued on Feb. 21, 2023.
Prior Publication US 2024/0179157 A1, May 30, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 9/40 (2022.01); G06F 18/23 (2023.01); G06F 18/24 (2023.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01)
CPC H04L 63/1416 (2013.01) [G06F 18/23 (2023.01); G06F 18/24765 (2023.01); G06N 3/08 (2013.01); G06N 20/00 (2019.01); H04L 63/145 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A method for training an artificial intelligence (AI) model using information pertaining to characteristics of upload activity performed at a client device, the method comprising:
generating, by a processing device, training data to train the AI model, wherein generating the training data comprises:
generating first training input, the first training input comprising (i) information identifying, for each of a plurality of application categories, a frequency of upload activity that corresponds to uploading first amounts of data during a specified time interval, wherein each of the plurality of application categories comprise one or more applications that are installed at the client device and that upload the first amounts of data; and
generating a first target output for the first training input, wherein the first target output indicates whether the frequency of upload activity that corresponds to uploading the first amounts of data corresponds to malicious or non-malicious upload activity; and
providing the training data to train the AI model on (i) a set of training inputs comprising the first training input, and (ii) a set of target outputs comprising the first target output,
wherein the trained AI model is configured to generate one or more outputs that cause a performance of a remedial action with respect to a new malicious upload activity associated with a particular application category.