본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

목차보기

Title page 1

Contents 6

Abstract 4

Acknowledgements 5

Executive summary 7

1. Background 9

2. What is an AI system? 10

2.1. OECD definition of an AI system 10

2.2. Main elements of the OECD definition of an AI system 10

3. What are AI agents and agentic AI? 13

3.1. AI Agents 15

3.1.1. Conceptual foundations of an AI agent 15

3.2. Agentic AI 19

3.2.1. Key features of agentic AI 19

3.3. Key elements and distinct characteristics 24

3.4. Agentic AI as a socio-technical paradigm 24

4. Recent trends in the uptake of AI agents and agentic AI 26

5. Discussion 29

References 30

Tables 6

Table 3.1. Illustrative definitions of AI agents 17

Table 3.2. Most definitions characterise AI agents as systems that produce outputs to achieve objectives with a certain degree of autonomy 19

Table 3.3. Illustrative definitions of agentic AI 22

Table 3.4. Key differences between AI agents and agentic AI systems 24

Table 4.1. Most common tools related to AI agents by use case 28

Figures 6

Figure 3.1. Global searches for "agentic AI" on Google surged in 2025 13

Figure 4.1. Half of developers on Stack Overflow plan to use AI agents; 38% remain unswayed 26

Figure 4.2. Vast majority of developers using AI agents concerned over privacy, security and accuracy 27

Figure 4.3. Software engineering most common use for AI agents among developers 28

Boxes 14

Box 3.1. Evolution of agentic research 14