US Patent Issued to Palo Alto Networks on Sept. 15 for "Detection of indirect prompt injection attacks with malicious instructions detection models" (California Inventors)

ALEXANDRIA, Va., Sept. 15 -- United States Patent no. 12,737,462, issued on Sept. 15, was assigned to Palo Alto Networks Inc. (Santa Clara, Calif.).

"Detection of indirect prompt injection attacks with malicious instructions detection models" was invented by Chien-Hua Lu (San Jose, Calif.), Bo Qu (Saratoga, Calif.), Xu Zou (Saratoga, Calif.) and Sergey Sviridov (Santa Clara, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A malicious instructions detection model ("detector") intercepts augmented prompts destined for a large language model ("LLM"). Each augmented prompt was augmented with data from potentially compromised data sources susceptible to indirect prompt injection attacks. The detector tokeni...