※ URL(Korean): https://www.kisdi.re.kr/bbs/view.do?bbsSn=114704&key=m2101113055776&pageIndex=3&sc=&sw=

KISDI Publishes Report on “The Rise of AI Agents and Policy Challenges” 
– From “answering AI” to “acting AI” 
– Government’s proactive role needed in the coming “AI-as-an-industry” era 

In 2025, AI agents have emerged as one of the key technology trends to watch. The Korea Information Society Development Institute (KISDI, President Sangkyu Rhee) recently published KISDI Premium Report titled “The Rise of AI Agents and Policy Challenges,” which provides a comprehensive analysis of the emergence and outlook of AI agents, the opportunities and risks accompanying their proliferation, and the policy implications for effectively responding to the AI agent era.
 
AI agents are autonomous AI systems that can understand user intentions, plan and make decisions on their own, and perform tasks using external tools. They represent a major evolution from traditional “answering AIs,” which focus on generating information, to “acting AIs” capable of taking autonomous actions. While “answering AI” services—exemplified by ChatGPT—have demonstrated AI’s potential as a general-purpose technology through human-like communication and reasoning, “acting AI,” or AI agents, add problem-solving, execution capabilities, and autonomy, enabling them to help users achieve their goals more effectively across diverse environments.

The report projects that the spread of AI agents will create new high value-added market opportunities for AI suppliers, while offering AI user companies chances to boost productivity and lower barriers to AI adoption. In particular, KISDI expects that the proliferation of AI agents will usher in a new “AI-as-an-industry” era, in which AI becomes an integral part of industrial structures. As service models across industries are reorganized around AI agents, and as AI, platform, and hardware companies engage in deeper cross-sector collaboration and competition, new business opportunities are expected to emerge and expand throughout the digital ecosystem—including services, platforms, infrastructure, IoT, and hardware.

The report also notes that as demand grows for AI agent services in industries requiring real-time interaction and the handling of sensitive data, technologies such as edge computing and on-device AI—where computation occurs locally on the device—will advance even more rapidly.

Ultimately, these changes are expected to extend beyond industrial opportunities, contributing to greater efficiency, convenience, and innovation across society, thereby improving quality of life and enhancing overall national productivity.

However, these opportunities come with new risks. The report highlights potential threats such as deepening dependence on a small number of big tech platform companies, ambiguity surrounding the legal status and accountability of AI agents, the risk of human job displacement, and growing social dependence stemming from diminished human judgment and autonomy.

Accordingly, Research Fellow Kyungseon Lee proposed three main policy directions: (1) promoting a fair and vibrant domestic industrial ecosystem, (2) advancing related legal frameworks and international cooperation, and (3) establishing comprehensive social response systems.

KISDI will continue to closely monitor the evolution of AI technologies and expand its research to support the development of policies and institutional foundations needed for the era of AI agents.