Bank of America Corporation

Robert Riley Zink

Jack Lawson Bishop, Iii

Timothy Andrew Wright

Eric Joseph DePree

Jared D Wilson

Matthew E Kelly

Stephanie Margaret Pirman

Michael Edward Hunter Dunten

Systems, computer program products, and methods are described herein for dynamically performing linked security tests. The present invention may be configured to determine a fingerprint of an application, perform, in an order based on the fingerprint of the application, security test sequences on the application, parse responses of the application to the security test sequences to generate results of the security test sequences, and label, with the fingerprint, the results. The present invention may be further configured to provide, to one or more machine learning models, the labeled results to determine probabilities of applications having a same fingerprint as the fingerprint of the application failing the security test sequences, update, based on the probabilities of the applications failing the security test sequences, the order, and store, in a temporary persistent storage device and based on the updated order, the security test sequences.

Electronic system for dynamically performing linked evaluation tests to identify faulty code and vulnerabilities in software programs based on efficacy

A device configured to receive a data sample about a configuration for one or more network devices in a public network. The device is further configured to compare one or more threat indicators to the data sample where each threat indicator is associated with a configuration setting. The device is further configured to identify a first network device in the public network that comprises a configuration that matches a threat indicator and to generate a bad actor profile for the first network device. The device is further configured to receive data traffic for a second network device in a private network and to block data communications between the second network device in the private network and the first network device in the public network in response to determining that the first network device is associated with the bad actor profile.

Preemptive threat detection for an information system

Polymorphic non-attributable processes and architectures to monitor threat domains (e.g., pharming or phishing websites) are disclosed. Obfuscated requests may be generated by control servers to be blended in with normal traffic sent over cloud networks with randomized exit nodes or with normal traffic sent through an anonymization network. Requests may be sent at randomized intervals or time periods determined algorithmically. The requests are obfuscated in order to mask the origination information and location so that the threat actor does not detect that the website is being monitored. User agents may be spoofed and requests may present as if they originated from residential IP addresses. Automatic real-time monitoring can be provided to determine when sites resolve and are addressable. Fingerprint information, screenshots, security certificate, and other threat domain data can be captured. Request responses can be scanned for threat indicia.

Polymorphic non-attributable website monitor

Systems, computer program products, and methods are described herein for dynamically generating linked security tests. The present invention may be configured to perform security tests on an application, generate, based on the results of the security tests, security test sequences that include at least one security test that the application failed, perform the security test sequences on the application, and, iteratively and until the application passes each security test sequence in an iteration, generate additional security test sequences. The present invention may be further configured to provide results of the security tests and security test sequences to one or more machine learning models to generate supplementary security test sequences and determine probabilities of the application failing the supplementary security test sequences.

Electronic system for identifying faulty code and vulnerabilities in software programs using linked evaluation tools

Embodiments of the invention are directed to systems, methods, and computer program products for creating an effective baseline for controlling software and programs running on a user device or virtual machine environment of an enterprise network. Historical data may be used to train one or more iterative machine learning processes for intelligently identifying relevant program characteristics and aiding or automating a determination as to a software or program's need, usage, redundancy, or security. Determinations regarding software or program classifications may be used for further review or automated control of software or program execution within a network environment.

System and methods for automated software analysis and classification

Systems, computer program products, and methods are described herein for testing an application with dynamically linked security tests. The present invention may be configured to perform, using a request engine, based on the first data, and based on test protocols stored in a first data structure, a first security test on an application. The present invention may be further configured to determine, based on determining that the application failed the first security test and based on a second data structure, whether the first security test is linked to one or more other security tests, where the second data structure includes security test sequences linking security tests and/or data to transmit from the first security test to the one or more other security tests, and provide, to the first queue, one or more other security tests to which the first security test is linked.

Growing community of inventors

Chicago, IL, United States of America

Robert Riley Zink