ENTITY MALICIOUSNESS ANALYSIS USING AUTONOMOUS ARTIFICIAL INTELLIGENCE AGENTS
개요
출원인
Microsoft Technology Licensing, LLC
발명자
Naveed Azeemi AHMAD; Lloyd Geoffrey GREENWALD; Muhammed Fatih BULUT; Yingqi LIU; Acar TAMERSOY
IPC 분류
CPC 분류
Techniques are described herein that are capable of performing entity maliciousness analysis using autonomous AI agents. A first autonomous AI agent selects relevant data from a corpus of data using a first selected AI tool in an AI model as a result of the relevant data being associated with an entity. A second autonomous AI agent generates a maliciousness determination, which indicates whether the entity exhibits malicious behavior, by analyzing the relevant data using a second selected AI tool in the AI model. A third autonomous AI agent generates a validity determination, which indicates whether the maliciousness determination is valid, by analyzing the maliciousness determination using a third selected AI tool in the AI model. As a result of an analysis that takes into consideration the validity determination, execution of an instruction that causes a security action to be performed with regard to the entity is triggered.
원문 (중국어)
Techniques are described herein that are capable of performing entity maliciousness analysis using autonomous AI agents. A first autonomous AI agent selects relevant data from a corpus of data using a first selected AI tool in an AI model as a result of the relevant data being associated with an entity. A second autonomous AI agent generates a maliciousness determination, which indicates whether the entity exhibits malicious behavior, by analyzing the relevant data using a second selected AI tool in the AI model. A third autonomous AI agent generates a validity determination, which indicates whether the maliciousness determination is valid, by analyzing the maliciousness determination using a third selected AI tool in the AI model. As a result of an analysis that takes into consideration the validity determination, execution of an instruction that causes a security action to be performed with regard to the entity is triggered.