# US 2025/0168198 A1 · Methods and systems for ai-driven policy generation

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

- Number: US 2025/0168198 A1
- Document ID: US20250168198A1
- Status: Published application
- Kind: A1
- Application number: 18/748,461
- Filed: 2024-06-20
- Published: 2025-05-22
- Earliest claimed: 2023-06-30
- Inventors: STEPHEN TUCKER; RATHINASABAPATHY ARUMUGAM; SRIDHAR CHANDRASHEKAR
- Web page: https://patentgrep.com/patent/US20250168198A1

## CPC

- H04L 63/20: {for managing network security; network security policies in general (filtering policies H04L63/0227)}
- G06F 40/30: Semantic analysis
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

In one aspect, a method of an managing policies in a multi-cloud governance platform comprising: implementing AI-driven policy generation in the multi-cloud governance platform by: providing at least one large language model (LLM) with sufficient size to have near or better than human reasoning abilities as an emergent property of the LLM; providing a plurality of cloud-computing platform dynamically updated documentations; with the LLM, interpreting an existing policy of a cloud-computing platform as provided in the plurality of cloud-computing platform dynamically updated documentations; with the by the LLM, generating executable check, for a compliance with a policy of the cloud-computing platform; and with the LLM, creating and maintaining a plurality of resources or activities associated with the policy for at least one cloud instance of the cloud-computing platform.

## Claims

### Claim 1

A method of an managing policies in a multi-cloud governance platform comprising:
implementing AI-driven policy generation in the multi-cloud governance platform by:
providing at least one large language model (LLM) with sufficient size to have near or better than human reasoning abilities as an emergent property of the LLM;
providing a plurality of cloud-computing platform dynamically updated documentations;
with the LLM, interpreting an existing policy of a cloud-computing platform as provided in the plurality of cloud-computing platform dynamically updated documentations;
with the by the LLM, generating executable check, for a compliance with a policy of the cloud-computing platform; and
with the LLM, creating and maintaining a plurality of resources or activities associated with the policy for at least one cloud instance of the cloud-computing platform.

### Claim 2 (depends on 1)

The method of claim 1, wherein the LLM comprises a GPT model.

### Claim 3 (depends on 2)

The method of claim 2, wherein the GPT model comprises GPT-4 model.

### Claim 4 (depends on 2)

The method of claim 2, wherein the GPT model comprises a plurality of artificial neural networks that are based on a transformer architecture, pre-trained on a plurality of large data sets of unlabeled text.

### Claim 5 (depends on 4)

The method of claim 4 wherein the large data sets of unlabeled text comprises the plurality of dynamically-updated cloud computing platform documentations.

### Claim 6 (depends on 5)

The method of claim 5, wherein the GPT model is pre-trained on the plurality of dynamically-updated cloud computing platform documentations on a periodic basis.

### Claim 7 (depends on 6)

The method of claim 6, wherein the GPT model generates a novel human-like content summary of the plurality of cloud computing platform documentations based on a query from a user regarding at least one cloud computing platform documentation to a human-computer interface provided by the GPT model.

### Claim 8 (depends on 7)

The method of claim 7, wherein the GPT model automatically implements a Chain-of-thought (CoT) conduct with respect to the plurality of cloud computing platform documentations content based on the query from the human user to include a first judgment about the nature of the content of the cloud computing platform documentation of the plurality of cloud-computing platforms.

### Claim 9 (depends on 8)

The method of claim 8, wherein the GPT model automatically implements the CoT conduct with respect to the plurality of cloud computing platform documentations content based on the query from the human user to include a second judgment about the human-users intentions with respect to a user's intention for the query with respect to the plurality of cloud-computing documentations.

### Claim 10 (depends on 9)

The method of claim 9, wherein the GPT model automatically implements the CoT conduct with respect to the plurality of cloud computing platform documentations content based on the query from the human user to include a third judgment about a taxonomic structure of the plurality of cloud-computing documentations as the plurality of cloud-computing documentations are dynamically updated.

### Claim 11 (depends on 10)

The method of claim 10, wherein the taxonomic structure comprises a taxonomic substructure of a plurality of cloud instances each of the plurality of cloud-computing documentations.

### Claim 12 (depends on 11)

The method of claim 11, wherein a GPT response is subsequently used to dynamically manage the plurality of cloud instances.

### Claim 13 (depends on 12)

The method of claim 12, wherein the executable checks are generated for compliance with the policy of the cloud-computing platform using an SDKs for the target cloud-computing platform.

### Claim 14 (depends on 1)

The method of claim 1, further comprising:
with the LLM, validating a plurality of compliance functions by seeding a reference instances with a set of test configurations that are then checked via the SDK functions to ensure they match the configuration state.

### Claim 15 (depends on 15)

The method of claim 15, further comprising:
with the LLM, implementing an additional code to perform prompt engineering and a Retrieval Augmented Generation (RAG) operation to perform a semantic operation on a policy of a relevant cloud-computing platform.

### Claim 16 (depends on 16)

The method of claim 16, further comprising:
eliciting a correct SDK code for each rule required by the policy.
