Granted Patent B2

US 12,267,431 · App. 17/662,941 · Granted Apr 1, 2025

Inventors: Peter Eric Sutherland (Matthews, NC); Laurence Henry Lipsmeyer (Chapel Hill, NC); Nelson Joseph Erb (Charlotte, NC)

Assignee: TRUIST BANK

H04L9/3228G10L15/22H04L63/105G10L2015/221G10L2015/223

Abstract

Claims (48)

1. A conversational interface system for step-up authentication using spoken passphrases, the system comprising:

at least one processor;

a communication interface communicatively coupled to the at least one processor; and

a memory device storing executable code that, when executed, causes the processor to:

receive a conversational request from a user for an action, and based thereon derive an intent by comparing, using a conversational interface, the intent to a collection of intents stored by an interaction model, the deriving determining the intent requires step-up authentication;

based on the determination, initiate a step-up authentication process that includes responding to the user using conversational artificial intelligence (AI) with a warning that an additional authentication step is required, the additional authentication step including an interaction with a virtual private cloud;

wherein a passphrase service of a datacenter transmits, over a communication channel separate from the conversational interface and to the user, a secret passphrase generated using a spoken passphrase dictionary;

receive a spoken submission from the user in response to the warning and based thereon determine a spoken intent of the spoken submission is to initiate an interaction with the virtual private cloud;

generate an authorization intent using an interaction model;

transmit the authorization intent and the spoken submission to the virtual private cloud (VPC), wherein the authorization intent and spoken submission is transmitted to the datacenter using one or more application programming interfaces (APIs) for validation of the spoken passphrase;

wherein the passphrase service of the datacenter confirms validity of the received spoken submission by comparing the spoken submission to the passphrase; and in response to confirmation of the spoken submission, authenticates the user, thereby authorizing the requested action; and

receive a confirmation of validation of the passphrase and that the action is being or will be performed.

2. The conversational interface of claim 1 , wherein the passphrase service of the datacenter accesses a spoken passphrase dictionary of words; and generates the secret passphrase by randomly selecting a plurality of words from the spoken passphrase dictionary.

3. The conversational interface of claim 1 , wherein transmission by the passphrase service of the datacenter of the secret passphrase over the communication channel separate from the conversational interface and receiving of the spoken submission occur asynchronously.

4. The conversational interface of claim 3 , wherein asynchronous occurrence refers to up to one hour of time passing between transmission and receiving.

5. The conversational interface of claim 1 , wherein the communication channel is configured to communicate using at least one selected from the group consisting of SMS, text, email, and app push notification.

6. The conversational interface of claim 1 , wherein the passphrase is a one-use passphrase that will be invalid if used more than once.

7. The conversational interface of claim 1 , wherein the passphrase expires a predetermined period of time after its transmission to the user.

8. The conversational interface of claim 1 , wherein the spoken passphrase dictionary words are subject to removal from the dictionary in the event they are identified as causing confusion with other words.

9. The conversational interface of claim 1 , wherein the passphrase comprises at least three words randomly selected from the spoken passphrase dictionary.

10. A method for step-up authentication for using spoken passphrases, the method performed by a conversational interface and comprising:

receiving a conversational request from a user for an action, and based therein deriving an intent by comparing, using a conversational interface, the intent to a collection of intents stored by an interaction model, the deriving determining the intent requires step-up authentication;

based on the determination, initiate a step-up authentication process that includes responding to the user using conversational artificial intelligence (AI) with a warning that an additional step is required, the additional authentication step including an interaction with a virtual private cloud;

wherein a passphrase service of a datacenter transmits, over a communication channel separate from the conversational interface and to the user, a secret passphrase generated using a spoken passphrase dictionary;

receiving a spoken submission from the user in response to the warning and based thereon determine a spoken intent of the spoken submission is to initiate an interaction with the virtual private cloud;

generating an authorization intent using an interaction model;

transmitting the authorization intent and the spoken submission to the virtual private cloud (VPC), wherein the authorization intent and spoken submission is transmitted to the datacenter using one or more datacenter application programming interfaces (APIs) for validation of the spoken passphrase;

wherein the passphrase service of the datacenter confirms validity of the received spoken submission by comparing the spoken submission to the passphrase; and in response to confirmation of the spoken submission, authenticates the user, thereby authorizing the requested action; and

receiving a confirmation of validation of the passphrase and that the action is being or will be performed.

11. The method of claim 10 , wherein the passphrase service of the datacenter accesses a spoken passphrase dictionary of words; and generates the secret passphrase by randomly selecting a plurality of words from the spoken passphrase dictionary.

12. The method of claim 10 , wherein transmission by the passphrase service of the datacenter of the secret passphrase over the communication channel separate from the conversational interface and receiving of the spoken submission occur asynchronously.

13. The method of claim 12 , wherein asynchronous occurrence refers to up to one hour of time passing between transmission and receiving.

14. The method of claim 10 , wherein the communication channel is configured to communicate using at least one selected from the group consisting of SMS, text, email, and app push notification.

15. The method of claim 10 , wherein the passphrase is a one-use passphrase that will be invalid if used more than once.

16. The method of claim 10 , wherein the passphrase expires a predetermined period of time after its transmission to the user.

17. A conversational interface system for step-up authentication using spoken passphrases, the system comprising:

at least one processor;

a communication interface communicatively coupled to the at least one processor; and

a memory device storing executable code that, when executed, causes the processor to:

receive a conversational request from a user for an action, and based thereon derive an intent by comparing, using a conversational interface, the intent to a collection of intents stored by an interaction model, the deriving determining the which requires step-up authentication of the user;

based on the determination, initiate a step-up authentication process that includes responding to a user using conversational artificial intelligence (AI) with a warning that an additional step is required, the additional authentication step including an interaction with a virtual private cloud;

wherein a passphrase service of a datacenter transmits, over a communication channel separate from the conversational interface and to the user, a secret passphrase generated using a spoken passphrase dictionary;

receive a spoken submission from the user in response to the warning and based thereon determine a spoken intent of the spoken submission is to initiate an interaction with the virtual private cloud;

wherein the user has been validated using basic authentication to the conversational interface, which stores user information in the memory device;

generate an authorization intent using an interaction model and at least a portion of the user information;

transmit the authorization intent and the spoken submission to the virtual private cloud (VPC), wherein the authorization intent and spoken submission is transmitted to the datacenter using one or more datacenter application programming interfaces (APIs) for validation of the spoken passphrase;

wherein the passphrase service of the datacenter confirms validity of the received spoken submission by comparing the spoken submission to the passphrase; and in response to confirmation of the spoken submission, authenticates the user, thereby authorizing requested action; and

receive a confirmation of validation of the passphrase and that the action is being or will be performed.

Assignments (1)

Continuity (1)

References Cited (16)

Cited By (1)

US 12,556,370