Intel Agentic Computing for AI Agents

We’re showcasing how an AI Agent can autonomously implement the user defined task through a process of planning and execution, understanding what it needs to do to manipulate multiple applications, then taking actions using the mouse and keyboard to interact with those applications. This autonomous behavior may bring up concerns around vulnerabilities and security. This demo shows the research from Intel Labs around Agentic Computing and AI red-teaming that leads to improved AI Agents and shows our commitment to Responsible AI.