# US 2025/0094325 A1 · Synthetic test data generation using generative artificial intelligence

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0094325 A1
- Document ID: US20250094325A1
- Status: Published application
- Kind: A1
- Application number: 18/470,234
- Filed: 2023-09-19
- Published: 2025-03-20
- Earliest claimed: 2023-09-19
- Assignees: Acronis International GmbH
- Applicants: Acronis International GmbH
- Inventors: Candid Wuest; Serg Bell; Stanislav Protasov
- Web page: https://patentgrep.com/patent/US20250094325A1

## CPC

- G06F 11/3684: {for test design, e.g. generating new test cases}
- G06F 11/3688: {for test execution, e.g. scheduling of test suites}
- G06F 11/3692: {for test results analysis}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems and methods for generating synthetic test data for testing a software solution. Systems and methods include receiving a testing task from a user, identifying test properties of the testing task, gathering initial information based on the test properties of the testing task, forming a training dataset based on the initial information and the test properties, pretraining a generative AI model based on a large language model (LLM) using the training dataset, configuring synthetic test data based on the test properties, and generating synthetic test data according to the testing task using the generative AI model.

## Claims

### Claim 1

A method of generating synthetic test data for testing a software solution on a computing system, the method comprising:
receiving a testing task from a user, wherein the testing task is indicative of a software solution to be tested and a type of testing to be performed;
identifying test properties of the testing task, including at least one required attribute for synthetic test data;
gathering initial information based on the test properties of the testing task;
forming a training dataset based on the initial information and the test properties;
pretraining a generative AI model based on a large language model (LLM) using the training dataset;
configuring synthetic test data based on the test properties; and
generating synthetic test data according to the testing task using the generative AI model.

### Claim 2 (depends on 1)

The method of claim 1, further comprising feeding the synthetic test data into the computing system.

### Claim 3 (depends on 1)

The method of claim 1, further comprising generating a feedback loop about the synthetic test data for retraining of the generative AI model.

### Claim 4 (depends on 3)

The method of claim 3, further comprising retraining the generative AI model based on the test properties, the initial information, and the feedback loop.

### Claim 5 (depends on 1)

The method of claim 1, further comprising validating the generated synthetic test data against predefined criteria.

### Claim 6 (depends on 1)

The method of claim 1, further comprising implementing data anonymization processes during synthetic data generation.

### Claim 7 (depends on 1)

The method of claim 1, wherein the initial information comprises at least one of:
existing data of the computing system, existing threat data,
relevant threat data associated with the software solution,
system configuration data including hardware specifications, operating system settings, software configurations,
user behavior data including browsing patterns, application usage, and response to various stimuli.

### Claim 8 (depends on 1)

The method of claim 1, wherein the testing task properties include at least one objective of the software solution testing and the targeted environment for testing.

### Claim 9 (depends on 1)

The method of claim 1, wherein the configuring is implemented through a configuration interface wherein the user specifies at least one of a synthetic data property, synthetic data variation level, or protocol adherence through the configuration interface.

### Claim 10

A system for generating synthetic test data for testing a software solution, comprising:
a hardware processor implemented on a computing device;
instructions that, when executed on the hardware processor, cause the computing device to implement:
an input module configured to receive a testing task from a user by using an input module, wherein the testing task is indicative of a software solution to be tested and a type of testing to be performed,
a data processing module configured to:
identify properties of the testing task, by using a data processing module, including at least one required attribute for synthetic test data,
gather initial information, by using the data processing module, based on the properties of the testing task,
form a training dataset, by using the data processing module, based on the initial information and the test properties, and
configure synthetic test data based on a plurality of testing task properties,
a generative AI model pretrained based on a large language model (LLM) using the training dataset and configured to generate synthetic test data according to the testing task.

### Claim 11 (depends on 10)

The system of claim 10, wherein the instructions that, when executed on the hardware processor, cause the computing device to further implement a data feeder module configured to supply the synthetic test data to the computing system.

### Claim 12 (depends on 10)

The system of claim 10, wherein the AI model is further configured to receive feedback and retrain based on the feedback.

### Claim 13 (depends on 10)

The system of claim 10, wherein the data processing module is further configured to validate the synthetic test data against predefined criteria.

### Claim 14 (depends on 10)

The system of claim 10, wherein the data processing module is further configured to collect at least one of:
existing data of the computing system, existing threat data, relevant threat data associated with the software solution;
system configuration data including hardware specifications, operating system settings, software configurations;
user behavior data including browsing patterns, application usage, and response to various stimuli.

### Claim 15 (depends on 10)

The system of claim 10, wherein the input module is further configured to receive user-defined properties including amount of synthetic data, variation levels, and protocol adherence.

### Claim 16 (depends on 10)

The system of claim 10, wherein the AI model is further configured to perform data anonymization process during synthetic data generation.

### Claim 17 (depends on 10)

The system of claim 10, wherein the input module is further configured to identify testing task properties including at least one objective of the software solution testing and the targeted environment.

### Claim 18 (depends on 10)

The system of claim 10, wherein the data processing module collaborates with the AI model to maintain properties of synthetic test data.

### Claim 19

A method for generating synthetic test data for testing a software solution, the method comprising:
receiving input data from a user, the input data defining at least one property of the synthetic test data;
configuring the synthetic test data properties based on the input data;
training an AI model with use-case specific data; and
generating synthetic test data using the AI model.

### Claim 20 (depends on 19)

The method of claim 19, further comprising applying the synthetic test data to the software solution.
