US Patent Issued to Vijay Madisetti on Oct. 28 for "Method and system for protecting and removing private information used in large language models" (American, Indian Inventors)

ALEXANDRIA, Va., Oct. 28 -- United States Patent no. 12,455,909, issued on Oct. 28, was assigned to Vijay Madisetti (Alpharetta, Ga.).

"Method and system for protecting and removing private information used in large language models" was invented by Vijay Madisetti (Alpharetta, Ga.) and Arshdeep Bahga (Chandigarh, India).

According to the abstract* released by the U.S. Patent & Trademark Office: "A method for generating adversarial data for use in an LLM including receiving an input condition including authentic data records, extracting by a PII extraction module an authentic data record from the input condition, assigning a utility score to each data field of the plurality of data fields of the authentic data record by a generator neural ...