Microsoft Technology Licensing, LLC

Tal Baumel

Hadas Bitran

Ran Efrati

Aaron Toby Bornstein

Einat Gruda Idelovitch

Tamar Sidi Maoz

Danny Karmon

Aviram Fireberger

Ksenya Kveler

Shahar Admati

Bert Jozef Prosper Hoorne

A method, computer program product, and computing system for processing a raw content portion to identify personal information, thus defining identified personal information; confirming the efficacy of the identified personal information; replacing the identified personal information within the raw content portion with one or more surrogates, thus defining a deidentified content portion; and performing a federated learning operation on the deidentified content portion to generate training data for a machine learning model.

System and method for federated learning

Context-aware content filtering adapted for a knowledge domain is provided. In certain examples, a classification confidence score by a classifier indicates a level of confidence that a prompt from a user is associated with the knowledge domain. The classification confidence score is compared with a threshold. When the score is below a threshold, a violation notice is provided to the user without submitting the prompt to a generative artificial intelligence (GAI) model. When the classification confidence score is above the threshold, the prompt is further processed to determine, according to rules, whether the prompt should be submitted to the GAI model. In various examples, the rules are applied to contextual information, safety score information, and intent information derived from the prompt.

Context-aware domain-specific content filtering

Growing community of inventors

Tel Aviv, Israel

Tal Baumel