Systems and methods for a generative artificial intelligence model with confidence indication

1 . A provider computing system comprising:

at least one processing circuit having at least one processor coupled to at least one memory device, the at least one memory device storing instructions thereon that, when executed by the at least one processor, cause the at least one processing circuit to:

receive, from a user device, a query corresponding to a first topic;

determine, responsive to receipt of the query, one or more data sources associated with the first topic based on data included in the query;

receive, from the user device, a first set of information that identifies a data source of the one or more data sources having data associated with the query;

interface with the identified data source to retrieve the data associated with the query;

generate, using a machine learning model, a first response to the query based on information received from the identified data source including the data associated with the query;

determine, responsive to evaluation of the first response, a correlation between the first response generated using the machine learning model and the information received from the identified data source used to generate the first response;

generate a first score for the first response based on the correlation, the first score indicating a confidence of the first response generated using the machine learning model; and

transmit one or more signals to cause the user device to display a user interface including the first response and the first score.

2 . The provider computing system of claim 1 , wherein the instructions further cause the at least one processing circuit to:

receive, from the user device, an indication to generate a second response to a second query without a second score that indicates a confidence of the second response; and

prevent, responsive to receipt of the second query, generation of the second score by providing data included in the second query to the machine learning model and displaying the second response responsive to generation of the second response.

3 . The provider computing system of claim 1 , wherein the instructions further cause the at least one processing circuit to:

receive, responsive to displaying the user interface, an indication to generate a second response to the query, the indication including a first set of information to identify a data source having a second set of information to generate the second response;

interface, responsive to identification of the data source, with the data source to retrieve the second set of information; and

generate, using the machine learning model, the second response based on the second set of information.

4 . The provider computing system of claim 3 , wherein the instructions further cause the at least one processing circuit to:

transmit one or more second signals to cause the user device to update the user interface to include the second response and an element to indicate that the second response was generated with the second set of information.

5 . The provider computing system of claim 1 , wherein the instructions further cause the at least one processing circuit to:

determine, responsive to receipt of the query, a credential associated with the user device;

identify, responsive to determination of the credential, one or more sets of information that are accessible based on the credential; and

generate, using the machine learning model responsive to retrieval of the one or more sets of information, the first response based on the one or more sets of information.

6 . The provider computing system of claim 1 , wherein the instructions further cause the at least one processing circuit to:

generate, using the machine learning model based on the information received from the identified data source, a plurality of responses including the first response;

determine, responsive to generation of the plurality of responses, correlations between respective responses of the plurality of responses and the information received from the identified data source;

generate, based on the correlations, a plurality of scores for the plurality of responses, each of the plurality of scores associated with a respective response of the plurality of responses, wherein the plurality of scores include the first score, and each of the plurality of scores indicate a confidence of each respective response of the plurality of responses; and

output, responsive to a determination that the first score exceeds a predetermined threshold, the first response.

7 . The provider computing system of claim 1 , wherein the user interface includes a graphical representation to indicate the first score, wherein the graphical representation includes at least one of an icon, a button, or an element, and wherein the graphical representation further includes an indication of data used to generate the first response.

8 . The provider computing system of claim 1 , wherein the machine learning model includes at least one of:

a Large Language Model;

a generative pre-trained transformer; or

a generative artificial intelligence model.

9 . A method, comprising:

receiving, by a provider computing system from a user device, a query corresponding to a first topic;

determining, by the provider computing system responsive to receipt of the query, one or more data sources associated with the first topic based on data included in the query;

receiving, by the provider computing system from the user device, a first set of information that identifies a data source of the one or more data sources having data associated with the query;

interfacing, by the provider computing system, with the identified data source to retrieve the data associated with the query;

generating, by the provider computing system using a machine learning model, a first response to the query based on information received from the identified data source including the data associated with the query;

determining, by the provider computing system responsive to evaluation of the first response, a correlation between the first response generated using the machine learning model and the information received from the identified data source used to generate the first response;

generating, by the provider computing system, a first score for the first response based on the correlation, the first score indicating a confidence of the first response generated using the machine learning model; and

transmitting, by the provider computing system, one or more signals to cause the user device to display a user interface including the first response and the first score.

10 . The method of claim 9 , further comprising:

receiving, by the provider computing system from the user device, an indication to generate a second response to a second query without a second score that indicates a confidence of the second response; and

preventing, by the provider computing system responsive to receipt of the second query, generation of the second score by providing data included in the second query to the machine learning model and displaying the second response responsive to generation of the second response.

11 . The method of claim 9 , further comprising:

receiving, by the provider computing system responsive to displaying the user interface, an indication to generate a second response to the query, the indication including a first set of information to identify a data source having a second set of information to generate the second response;

interfacing, by the provider computing system responsive to identification of the data source, with the data source to retrieve the second set of information; and

generating, by the provider computing system using the machine learning model, the second response based on the second set of information.

12 . The method of claim 11 , further comprising:

transmitting, by the provider computing system, one or more second signals to cause the user device to update the user interface to include the second response and an element to indicate that the second response was generated with the second set of information.

13 . The method of claim 9 , further comprising:

determining, by the provider computing system responsive to receipt of the query, a credential associated with the user device;

identifying, by the provider computing system responsive to determination of the credential, one or more sets of information that are accessible based on the credential; and

generating, by the provider computing system using the machine learning model responsive to retrieval of the one or more sets of information, the first response based on the one or more sets of information.

14 . The method of claim 9 , further comprising:

generating, by the provider computing system using the machine learning model based on the information received from the identified data source, a plurality of responses including the first response;

determining, by the provider computing system responsive to generation of the plurality of responses, correlations between respective responses of the plurality of responses and the information received from the identified data source;

generating, by the provider computing system based on the correlations, a plurality of scores for the plurality of responses, each of the plurality of scores associated with a respective response of the plurality of responses, wherein the plurality of scores include the first score, and each of the plurality of scores indicate a confidence of each respective response of the plurality of responses; and

outputting, by the provider computing system responsive to a determination that the first score exceeds a predetermined threshold, the first response.

15 . The method of claim 9 , wherein the user interface includes a graphical representation to indicate the first score, wherein the graphical representation includes at least one of an icon, a button, or an element, and wherein the graphical representation further includes an indication of data used to generate the first response.

16 . A non-transitory computer-readable storage media having instructions stored thereon that, when executed by at least one processor of a provider computing system, cause the provider computing system to perform operations comprising:

receiving, from a user device, a query corresponding to a first topic;

determining, responsive to receipt of the query, one or more data sources associated with the first topic based on data included in the query;

receiving, from the user device, a first set of information that identifies a data source of the one or more data sources having data associated with the query;

interfacing with the identified data source to retrieve the data associated with the query;

generating, using a machine learning model, a first response to the query based on information received from the identified data source including the data associated with the query;

determining, responsive to evaluation of the first response, a correlation between the first response generated using the machine learning model and the information received from the identified data source used to generate the first response;

generating a first score for the first response based on the correlation, the first score indicating a confidence of the first response; and

transmitting one or more signals to cause the user device to display a user interface including the first response and the first score.

17 . The non-transitory computer-readable storage media of claim 16 , wherein the instructions, when executed by the at least one processor of the provider computing system, further cause the provider computing system to perform operations comprising:

receiving, from the user device, an indication to generate a second response to a second query without a second score that indicates a confidence of the second response; and

preventing, responsive to receipt of the second query, generation of the second score by providing data included in the second query to the machine learning model and displaying the second response responsive to generation of the second response.