Granted Patent B2

US 12,541,728 · App. 18/535,614 · Granted Feb 3, 2026

Inventors: Nathan Coles (San Francisco, CA); Darren M. Goetz (San Francisco, CA); Uma Meyyappan (San Francisco, CA); Dennis Montenegro (San Francisco, CA); Steve Perez (San Francisco, CA); Debarchana Roy (San Francisco, CA)

Assignee: WELLS FARGO BANK, N.A.

G06Q10/063114G06N5/04G06N20/00G06Q10/06316G06Q10/10G06Q30/016G06Q30/0185G06Q40/02

Abstract

Claims (85)

1 . A method comprising:

training, by a provider computing system associated with a provider, one or more artificial intelligence (AI) models to generate task prompts, wherein training the one or more AI models comprises:

training, by the provider computing system, the one or more AI models to generate the task prompts based on historical data comprising a plurality of example prompts created in various example scenarios;

receiving, by the provider computing system, feedback regarding the generated task prompts, the feedback comprising at least one of (i) at least one of usefulness, convenience, or appropriateness of the task prompts, or (ii) the task prompts being modified to a different format, include more detail, or include less detail; and

updating, by the provider computing system, the one or more AI models using the feedback to generate one or more updated AI models;

receiving, by the provider computing system, a requested action from a customer;

determining, by the provider computing system, contextual information about a plurality of customer devices that the customer has indicated for use with the provider computing system;

generating, by the provider computing system using the one or more updated AI models, a task prompt based on an identified customer pre-task for completion of the requested action, the task prompt requesting the customer to perform the customer pre-task based on the contextual information;

generating, by the provider computing system, an interaction board accessible by the customer via a customer device associated with the customer and accessible by an employee associated with the provider via an employee device associated with the employee, the interaction board being displayed via a graphical user interface and including the customer pre-task;

receiving, by the provider computing system, an indication that the customer pre-task has been completed prior to a customer interaction at a branch location, the customer interaction associated with the requested action, the branch location associated with the provider;

updating, by the provider computing system, the interaction board to indicate that the customer pre-task has been completed prior to the customer interaction;

receiving, by the provider computing system, information associated with the requested action from the customer device during the customer interaction; and

updating, by the provider computing system, the interaction board to include the information received from the customer device during the customer interaction.

2 . The method of claim 1 , further comprising:

identifying, by the provider computing system, the customer device from the plurality of customer devices based on the customer pre-task and the contextual information; and

transmitting, by the provider computing system, the task prompt to the customer device.

3 . The method of claim 1 , further comprising:

receiving, by the provider computing system, biometric data associated with the customer; and

authenticating, by the provider computing system, the customer via the biometric data.

4 . The method of claim 1 , further comprising:

receiving, by the provider computing system from the customer device via a near-field communication (NFC) device or a radio-frequency identification (RFID) device, a unique device identifier; and

authenticating, by the provider computing system, the customer via the unique device identifier.

5 . The method of claim 2 , wherein the task prompt is a request for the customer to complete and provide at least one of a document or a form associated with the requested action, and the information received from the customer during the customer interaction is at least one of another document or another form associated with the requested action.

6 . The method of claim 1 , further comprising:

uploading a document to the interaction board by at least one of the customer via the customer device or the employee via the employee device.

7 . The method of claim 6 , wherein, once uploaded to the interaction board, the document is viewable and/or editable by the customer device and the employee device.

8 . The method of claim 6 , further comprising:

receiving task details pertaining to the customer pre-task from at least one of the employee device or a document analysis circuit, the task details including an indication of pertinent portions of the document; and

updating, by the provider computing system, the interaction board to include the task details.

9 . The method of claim 8 , wherein receiving the task details comprises receiving the task details from the document analysis circuit, and the method further comprises:

identifying, by the document analysis circuit, the pertinent portions of the document using artificial intelligence models trained by studying at least one of user inputs over time relating to similar documents or historical data relating to similar documents.

10 . The method of claim 1 , further comprising:

determining, by the provider computing system, whether the customer is driving; and

providing, in response to determining that the customer is driving and by the provider computing system, the customer pre-task to the customer via voice communications.

11 . The method of claim 1 , further comprising:

identifying, by the provider computing system, a provider task for completion of the requested action to be performed by the employee;

updating, by the provider computing system, the interaction board to include the provider task;

generating, by the provider computing system, a provider task prompt requesting the employee to perform the provider task;

transmitting, by the provider computing system, the provider task prompt to the employee device;

receiving, by the provider computing system, a provider indication from the employee device that the provider task has been completed; and

updating, by the provider computing system, the interaction board to indicate that the provider task has been completed.

12 . A method comprising:

training, by a provider computing system associated with a provider, one or more artificial intelligence (AI) models to generate task prompts, wherein training the one or more AI models comprises:

training, by the provider computing system, the one or more AI models to generate the task prompts based on historical data comprising a plurality of example prompts created in various example scenarios;

receiving, by the provider computing system, feedback regarding the generated task prompts, the feedback comprising at least one of (i) at least one of usefulness, convenience, or appropriateness of the task prompts, or (ii) the task prompts being modified to a different format, include more detail, or include less detail; and

updating, by the provider computing system, the one or more AI models using the feedback to generate one or more updated AI models;

receiving, by the provider computing system, a requested action from a customer;

identifying, by the provider computing system, a customer pre-task for completion of the requested action to be performed by the customer;

determining, by the provider computing system, contextual information about a plurality of customer devices associated with the customer;

identifying, by the provider computing system, a customer device of the plurality of customer devices based on the customer pre-task and the contextual information;

generating, by the provider computing system using the one or more updated AI models, a task prompt requesting the customer to perform the customer pre-task based on the contextual information; and

transmitting, by the provider computing system, the task prompt to the customer device.

13 . The method of claim 12 , wherein the contextual information includes at least one of a device form, a set of device capabilities, a device status, or device location information of the customer device.

14 . The method of claim 12 , wherein the task prompt is a request for at least one of a document, a form, or information from the customer.

15 . The method of claim 12 , further comprising:

identifying, by the provider computing system, a provider task for completion of the requested action to be performed by an employee associated with the provider associated with the provider computing system;

generating, by the provider computing system, a provider task prompt requesting the employee to perform the provider task; and

transmitting, by the provider computing system, the provider task prompt to an employee device associated with the employee.

16 . A customer interaction system comprising:

a provider computing system associated with a provider and configured to communicate with a customer device associated with a customer and with an employee device associated with an employee of the provider via a network, the provider computing system comprising a processing circuit including a processor and a memory, the processing circuit structured to:

train one or more artificial intelligence (AI) models to generate task prompts, wherein training the one or more AI models comprises:

training the one or more AI models to generate the task prompts based on historical data comprising a plurality of example prompts created in various example scenarios;

receiving feedback regarding the generated task prompts, the feedback comprising at least one of (i) at least one of usefulness, convenience, or appropriateness of the task prompts, or (ii) the task prompts being modified to a different format, include more detail, or include less detail; and

updating the one or more AI models using the feedback to generate one or more updated AI models;

receive a requested action from the customer via the customer device;

identify a customer pre-task for completion of the requested action to be performed by the customer;

determine contextual information about a plurality of customer devices that the customer has indicated for use with the provider computing system;

generate, using the one or more updated AI models, a task prompt requesting the customer to perform the customer pre-task based on the contextual information;

generate an interaction board accessible by the customer via the customer device and the employee via the employee device, the interaction board being displayed via a graphical user interface and including the customer pre-task;

receive an indication that the customer pre-task has been completed prior to a customer interaction at a branch location, the customer interaction associated with the requested action, the branch location associated with the provider;

update the interaction board to indicate that the customer pre-task has been completed prior to the customer interaction;

receive information associated with the requested action from the customer device during the customer interaction; and

update the interaction board to include the information received from the customer device during the customer interaction.

17 . The customer interaction system of claim 16 , wherein the processing circuit is further structured to:

identify the customer device from the plurality of customer devices based on the customer pre-task and the contextual information; and

transmit the task prompt to the customer device.

18 . The customer interaction system of claim 17 , wherein the contextual information includes at least one of a device form, a set of device capabilities, a device status, or device location information of each of the plurality of customer devices.

19 . The customer interaction system of claim 16 , wherein the provider computing system is structured to pull the contextual information from the plurality of customer devices automatically.

20 . The customer interaction system of claim 16 , wherein the processing circuit is further structured to:

identify a provider task for completion of the requested action to be performed by the employee;

update the interaction board to include the provider task;

generate a provider task prompt requesting the employee to perform the provider task;

transmit the provider task prompt to the employee device;

receive a provider indication from the employee that the provider task has been completed; and

update the interaction board to indicate that the provider task has been completed.

Assignments (1)

Continuity (2)

References Cited (28)