Bank of America Corporation

Elena Kvochko

Maria Carolina Barraza Enciso

Michael Ogrinz

Albena Noveva Fairchild

Katherine Dintenfass

Monika Kapur

Crystal M Sundaramoorthy

Jinna Zevulun Kim

Dipika Jain

George Albero

Manu Jacob Kurian

Stephen Thomas Shannon

William August Stahlhut

Lalit Dhawan

John Christopher Checco

Daniel Joseph Serna

Richard Scot

Christine D Black

Jesse Strickland Newsom, Iii

Luis Fernando Villeda

Aspects of the disclosure relate to using machine learning to modify account privacy settings. A computing platform may identify initial account settings for an individual of a plurality of individuals. Using a settings optimization model, the computing platform may identify account data and third party data for the individual. Using the settings optimization model, the computing platform may identify discrepancies between the initial account settings for the first individual and account settings for a subset of the plurality of individuals having common characteristics with the individual. Based on the discrepancies, the computing platform may identify settings modifications for the individual, and may determine that a modification of the settings modifications applies to a detected interaction of the individual. The computing platform may direct an enterprise data source to perform the modification, which may cause the enterprise data source to modify the initial account settings based on the modification.

Settings optimization engine using artificial intelligence to enhance client privacy

Aspects of the disclosure relate to using secure authentication tokens to grant power of attorney. A computing platform may receive user input indicating a task and requesting that a power of attorney be granted to an individual. The computing platform may generate an authority token granting the individual the power of attorney to perform the task, and may send the authority token, along with task information indicating the task, to the individual. The computing platform may receive a validation request including the authority token and the task information. Based on successful validation, the computing platform may send validation information to an event processing system. After receiving confirmation that the task has been performed by the individual, the computing platform may update dashboard information to indicate that the task has been completed, which may cause the user device to display a dashboard indicating completion of the task.

Coordination platform for generating and managing authority tokens

A system that provides an exchange platform for resource interaction processors that qualify under a modelling process. The platform is continuously updated with information regarding the resource interaction processors with both private and public information. The platform includes regulatory and other rules dictating interaction parameters associated with the resource interaction processors. The platform provides a viability metric of the resource interaction processors for providing resource interaction processor services to the resource interaction processors. Authorized entities may access and view the merchant exchange platform to gather information about the resource interaction processors for use in determining whether to provide services. Furthermore, the platform provides information regarding regulatory oversight for the specific resource interaction processors, allows for the establishment of a dialogue between resource interaction processors and service providers, provides ongoing review of resource interaction processors in real-time data upgrading in an encrypted environment.

Multi-dimensional modeling of resources for interaction systems

Aspects of the disclosure relate to using machine-learning models to authenticate users and protect enterprise-managed information and resources. In some embodiments, a computing platform may receive user interaction data from enterprise computing infrastructure and may train one or more authentication models based on this data. Subsequently, the computing platform may receive, from a first application server, a request to authenticate a first user to a first user account in a first usage session hosted by the first application server. In response to receiving this request, the computing platform may identify whether session-specific interaction data for the first usage session is valid based on the one or more authentication models. If the interaction data is identified as being valid, the computing platform may generate and send one or more commands directing the first application server to allow the first user to access the first user account in the first usage session.

Using machine-learning models to authenticate users and protect enterprise-managed information and resources

Systems for using cognitive automation techniques to detect unauthorized events are provided. In some examples, a request for a product or service (e.g., event processing) may be received. The request may be received electronically, via a telephone communication, in person, or the like. The request may be analyzed (e.g., using a cognitive automation model) to determine whether it is potentially unauthorized. If so, one or more requests for additional information may be generated or identified. For instance, requests for authentication information, responses to questions or a series of questions, or the like, may be generated. In some examples, the additional information requested may be identified using the cognitive automation model. Responses to the request for additional information may be analyzed (e.g., using the cognitive automation model) to determine whether the request for the product or service is unauthorized. If so, processing the request may be prevented.

Cognitive automation platform for dynamic unauthorized event detection and processing

Systems and methods for implementing sentiment analysis of computer code are provided. Developers who write source code may include comments or other natural language artifacts in the source code. These artifacts may be illustrative of current or legacy cybersecurity threats. Systems and methods may mine comments, and/or other code artifacts, for the dual purposes of cybersecurity threat detection and mitigation. Advanced code analytics may be leveraged for a deeper understanding of the sentiments expressed by the artifacts. Such sentiment may include negative sentiments expressed in error code selection and/or descriptions. All information retrieved is preferably human identity agnostic in line with personal data regulation compliance.

Sentiment analysis for securing computer code

Aspects of the disclosure relate to a system that provides enhanced security of information transmitted by connected devices. The system may be used to request resources from members of a donee's virtual group. Donated funds or line of credit may be loaded to the donee's digital wallet with no link to any of the donee's personal accounts. The request for resources may utilize secure correspondence between the donee and the prospective donors. The secure correspondence may be locked and prevented from being forwarded or captured via a screenshot to ensure confidentiality. If an attempt is made to forward or copy the secure correspondence, the secure correspondence may be deleted. Donated funds/line of credit may be allocated for specific use, and may not be owned by the donee until used.

System for secure distribution of peer requests for resources

A system for quantum key synchronization within a server-cluster is provided. The system may include a plurality of silicon-based servers encapsulated in quantum cases. Each quantum case may include a quantum tunneling transmitter module, a quantum random number generator and a quantum entanglement module. The quantum cases may communicate with each other via the quantum tunneling transmitter module or any other suitable manner. The quantum cases may only communicate with cases with which they are entangled. Therefore, in the event of a compromise on one of the servers, the quantum entanglement module, included in the case that encapsulates the compromised server, may become disentangled, and therefore not be able to communicate with the other servers included in the cluster using an internal communications protocol.

Quantum key synchronization within a server-cluster

A blockchain-based message transmission is provided. The system may include a plurality of silicon-based devices encapsulated in quantum cases. Each quantum case may include a quantum random number generator and a public key. The quantum random number generator may generate quantum-resilient random numbers to be used as private keys. The system may include a private network. The private network may include a subset of system's devices. A first device, included in the private network, may transmit a message to a second device included in the private network. A first quantum case that encapsulates the first device may intercept the message, generate a private key, encrypt the message using the private key, generate a data transaction block that includes message metadata, upload the data transaction block to a system blockchain and transmit the message to the recipient upon receipt of an approval from a majority of devices.

Blockchain system for hardening quantum computing security

Apparatus and methods for providing protection from economic exploitation. The apparatus and methods may include a platform for tagging a social security number and/or other suitable identification number. The system may allow a person associated with a social security number to automatically flag and deny any product or service request associated with the social security number. The alarms are 'silent,' because the alarm may initiate an institution internal process in which the customer interaction with the institution follows a normal diligence process until an intervention can be effectively executed on behalf of the customer. This reduces the likelihood of a reaction from the customer's associate prior to completing the diligence process. A system may provide the customer with an opportunity to electronically activate an alarm by typing keystrokes on a card reader keypad or interacting with a physical sensor on the reader, on the card, or the like.

Native-feature silent coercion alarm

Aspects of the disclosure relate to a cognitive automation based compliance management system. A computing platform may receive, via a network, an alert notification indicative of a change to a legal framework applicable to an enterprise organization. Subsequently, the computing platform may, in response to the alert notification, detect a task that may be impacted by the change. Then, the computing platform may, upon detecting the task, identify at least one task parameter associated with the task. Then, the computing platform may compare the at least one task parameter with the change to determine whether the task is impacted by the change. Subsequently, the computing platform may, upon a determination that the task is impacted by the change, recommend, via the network, one or more actions associated with the task.

Cognitive automation based compliance management system

A method for filtering internet traffic is provided. The method may include using a private network for receiving a request message from an electronic device within the private network and identifying the type of the electronic device. When the electronic device is identified as a non-IoT type device, the method may include transmitting the request message through the non-IoT output channel and when the electronic device is identified as an IoT type device the method may include transmitting the request message through the IoT output channel. The method may further include using an IP address filter gateway for filtering incoming traffic to a web server, the filtering may include granting device access to the web server when the request message is received through the non-IoT output channel and denying access to the web server when the request message is received through the IoT output channel.

Systems and methods for preventing, through machine learning and access filtering, distributed denial of service ('DDoS') attacks originating from IoT devices

A media verification device receives baseline media, which includes videos confirmed to include a target subject. The device determines, based on the baseline media for the target subject, a set of baseline features associated with the target subject. A baseline profile is determined for the target subject based on the set of baseline features. When test media which includes a video purported to include the test subject is received, test features are determined for the test media. A test profile is determined for the test media based on the set of test features. The test profile is compared to the baseline profile for the test subject. Based on this comparison, a confidence score is determined. If the confidence score is not greater than a threshold value, the test media is determined to include a synthetic video of the target subject, and an alert is provided.

Growing community of inventors

New York, NY, United States of America

Elena Kvochko