Privacy-preserving text insight mining in a closed domain

This invention takes text and gives it to two different AI language models. It then gets a response from each model. The claims focus on simply obtaining these two responses, particularly when the models were trained on different datasets, potentially from different user groups.

Why it matters: The rapid advancement and accessibility of generative language models since 2022 make training and comparing specialized models on different datasets, as described, significantly more feasible. This also enhances the relevance of comparing model outputs for understanding biases or privacy implications.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedAI/ML expertise, heavy software system

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