Granted Patent B2

US 12,562,163 · App. 18/214,082 · Granted Feb 24, 2026

Inventors: Shivani (Hyderabad, IN); Prasanthi PVL (Hyderabad, IN); Suneela Pothula (Hyderabad, IN); Aditya Ramamurthy (Hyderabad, IN); Swaroop Thummalapalli (Hyderabad, IN)

Assignee: ServiceNow, Inc.

G10L15/22G06F8/33

Abstract

Claims (53)

1 . A method, comprising:

receiving conversation data that indicates a user requested function, wherein the user requested function is associated with a workflow;

determining, via a machine learning model, computer instructions corresponding to the user requested function;

determining information associated with the computer instructions not provided by the user requested function;

performing at least one of: determining a placeholder for the information or outputting a request for the information;

updating the workflow to include the computer instructions corresponding to the user requested function; and

updating the computer instructions corresponding to an edit made directly to the workflow.

2 . The method of claim 1 , wherein the workflow includes a digitally managed workflow.

3 . The method of claim 1 , wherein:

the conversation data includes a natural language input; and

the user requested function includes a desired result associated with the workflow.

4 . The method of claim 1 , wherein the conversation data includes text input.

5 . The method of claim 1 , wherein the conversation data includes voice input.

6 . The method of claim 1 , wherein the conversation data is characterized by a conversation between a user and a computer system.

7 . The method of claim 1 , further comprising using a natural language understanding model to determine a response to the conversation data.

8 . The method of claim 7 , wherein the natural language understanding model includes a language model.

9 . The method of claim 1 , wherein determining the computer instructions corresponding to the user requested function includes:

obtaining statements from the machine learning model;

determining at least one parameter specific to the workflow;

determining an environment in which the statements are to be used; and

generating the computer instructions by updating the statements to include the at least one parameter and formatting the statements to be intelligible for the environment.

10 . The method of claim 1 , wherein updating the workflow includes updating parameters in an action of the workflow based at least on the conversation data.

11 . The method of claim 1 , further comprising executing the computer instructions to perform the user requested function.

12 . The method of claim 1 , further comprising:

receiving user input with respect to an element of the workflow;

determining conversation data corresponding to the received user input; and

outputting the conversation data corresponding to the received user input.

13 . The method of claim 1 , further comprising:

using the workflow to receive user input including metadata;

determining a representation of the user input; and

outputting the representation of the user input, wherein the representation of the user input is in natural language.

14 . The method of claim 1 , wherein the conversation data is associated with a conversation with a graphical user interface.

15 . The method of claim 1 , wherein the conversation data is included in a group of one or more tiles.

16 . A system, comprising:

processing circuitry; and

memory accessible by the processing circuitry, the memory storing instructions that, when executed by the processing circuitry, cause the processing circuitry to:

receive conversation data that indicates a user requested function, wherein the user requested function is associated with a workflow;

determine, via a machine learning model, computer instructions corresponding to the user requested function;

determine information associated with the computer instructions not provided by the user requested function;

perform at least one of: determining a placeholder for the information or outputting a request for the information;

update the workflow to include the computer instructions corresponding to the user requested function; and

update the computer instructions corresponding to an edit made directly to the workflow.

17 . The system of claim 16 , wherein the workflow includes a digitally managed workflow.

18 . The system of claim 16 , wherein:

the conversation data includes a natural language input; and

the user requested function includes a desired result associated with the workflow.

19 . A computer program product embodied in a non-transitory computer readable medium and comprising computer instructions for:

receiving conversation data that indicates a user requested function, wherein the user requested function is associated with a workflow;

determining, via a machine learning model, computer instructions corresponding to the user requested function;

determining information associated with the computer instructions not provided by the user requested function;

performing at least one of: determining a placeholder for the information or outputting a request for the information;

updating the workflow to include the computer instructions corresponding to the user requested function; and

updating the computer instructions corresponding to an edit made directly to the workflow.

Assignments (1)

Continuity (1)

References Cited (19)