Project Overview

• - What is the Agentic AI Project?

  This GPAI project takes stock of the landscape of agentic AI systems. The OECD working paper, The agentic AI landscape and its conceptual foundations, examines how the terms “AI agents” and “agentic AI” are defined and used across the literature, with the aim of enhancing conceptual clarity, supporting coherent research, and informing evidence-based policymaking. Building on this work, the Tokyo Centre is involved in the analysis of use cases of agentic AI and their implications.

• - Background

  AI is increasingly evolving into more autonomous “agentic AI” systems that can pursue goals and take actions with higher level of autonomy. However, there is no unified definition of agentic AI. This project aims to take stock of existing definitions, and to identify illustrative examples and current applications of such systems and inform future research and policymaking.
  This project is conducted jointly by the OECD Secretariat, the Tokyo Centre and the OECD.AI Expert Group established for this project.

• - Implementation Method

  The project comprises two main strands of work:

  1. 1 Taking stock of the landscape of agentic AI systems, examining how the terms “AI agents” and “agentic AI” are defined and used across the literature, and providing a conceptual foundation for the field (released as an OECD working paper)
  2. 2 Building on this, analysing use cases and their implications (with the involvement of the Tokyo Centre; released as an OECD working paper)