Using Large Language Models (‘LLMs’) For Code Hardening In A Storage System

This invention describes a system that uses artificial intelligence to fix software bugs in a storage system. When an error is found, the system creates a temporary copy of the software for a different user. An AI model then suggests a code fix for the error, which is applied and tested on this temporary copy. Once the fix is confirmed to work safely on the copy, it is then applied to the main software running the storage system.

Why it matters: Filed when LLMs were still nascent in reliably generating production-grade code fixes for critical systems. By 2026, LLM capabilities for accurate, secure, and context-aware code generation and debugging have significantly matured, making automated deployment in sensitive environments more viable.

Status: Under examinationA pending application. It has not been granted, and it has not been abandoned.
How hard to build: SpecializedSoftware logic, LLM integration, storage system APIs

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10