TLDR: South Korean tech giants, including Kakao, Naver, Samsung, and LG, are aggressively investing in agentic AI, a next-generation form of artificial intelligence capable of autonomously completing complex tasks. This strategic pivot aims to establish South Korea as a leader in the global AI revolution, with companies launching innovative services and platforms. The domestic enterprise agentic AI market is projected to surge from $61.4 million in 2024 to $875.6 million by 2030, reflecting a broader global trend towards AI-driven automation and intelligent collaboration.
South Korea is rapidly emerging as a formidable contender in the global artificial intelligence landscape, with its leading tech companies now locked in a fierce battle to dominate the burgeoning field of agentic AI. This advanced form of AI is characterized by its ability to act autonomously, reason independently, and make decisions to achieve specific goals, moving beyond simple command-response systems. The nation’s tech giants are not merely keeping pace with global trends but are actively striving to lead the next chapter of the AI revolution, aiming to carve out a distinct role in the global AI ecosystem.
Leading the charge is Kakao, the company behind South Korea’s ubiquitous messaging app, KakaoTalk. Kakao is taking a pioneering approach by integrating agentic AI directly into its core services. Their newly developed AI assistant, Kanana, is designed to manage a variety of tasks, such as planning trips, facilitating gift purchases, and scheduling appointments, all seamlessly within the KakaoTalk chat interface. This integration is bolstered by a partnership with OpenAI, bringing ChatGPT’s conversational intelligence to users. Kanana leverages ‘Tool Call’ capabilities to activate other Kakao services like KakaoMap, Melon, or Talk Calendar based on user requests, eliminating the need to switch between applications. Seung Kim, head of Kakao’s AI Synergy Task Force, emphasized that this allows users to access services without leaving the chat environment. Rollouts for these new features are anticipated to begin in October. To support this expansive ecosystem, Kakao has also introduced the Model Context Protocol (MCP), a standardized framework for AI agents to interact across different applications. Developers can contribute MCP-compatible tools to a marketplace called PlayTools, enabling users to customize their AI experience with modular add-ons.
Naver, South Korea’s dominant search engine, is pursuing a parallel strategy, embedding agentic AI deeply into its offerings. The company plans to transform its traditional keyword-based search portal into a conversational experience by introducing an ‘AI Tab’ by 2027. Kim Sang-bum, head of Naver AI Search, stated that Naver is developing both a ‘core agent’ to analyze user intent and ‘vertical agents’ to execute real-world actions. This initiative is powered by HyperCLOVA X, Naver’s proprietary large language model, which has been trained on extensive local data to ensure high linguistic and cultural precision, offering a distinct advantage over global rivals.
Beyond these consumer-facing applications, agentic AI is poised to revolutionize industrial and enterprise sectors. Samsung estimates that up to 70% of routine office work could eventually be automated by these intelligent systems. Meanwhile, LG is actively involved in smart factory innovation, with Woohyung Lim, Co-Director of LG AI Research, highlighting the shift towards collaborative agent ecosystems.
MakinaRocks, a prominent South Korean industrial AI company, recently hosted its flagship conference, ATTENTION 2025, in Seoul on September 4, 2025. The event, themed ‘AI that Thinks, Acts, and Transforms Industries,’ showcased the rapid rise of agentic AI, unveiled new features for its Runway platform, and outlined a roadmap for achieving hyperproductive industrial transformation. CEO Andre S. Yoon underscored the limitations of general-purpose AI for complex industrial tasks, advocating for domain-specific, vertical approaches. “Manufacturing, defense, and heavy industry hold the potential for a hundredfold productivity leap through agentic AI. Now is the time for organizations of all sizes to prepare to deploy thousands of AI agents,” Yoon stated. Panel discussions at the conference also focused on strategies for South Korea to secure AI sovereignty within its industrial sites and broader ecosystems. Real-world applications presented included AI-based water prediction (K-Water), engineering drawing analysis (Doosan Enerbility), AI agents on manufacturing floors (Hyundai AutoEver), and AI-based firepower planning (Korea Military Academy).
The economic impact of agentic AI is projected to be substantial. The enterprise agentic AI market in South Korea, valued at $61.4 million in 2024, is forecast to skyrocket to $875.6 million by 2030. Globally, Gartner predicts that by 2026, over 40% of enterprises will have integrated conversational AI agents, a significant increase from less than 5% in 2023. McKinsey further projects that generative and agentic AI could contribute an additional $2.6 to $4.4 trillion in annual value worldwide, with the largest gains expected from the automation of knowledge work.
Also Read:
- Agentic AI: Unpacking Autonomous Systems’ Thinking, Planning, and Action by 2026
- Knowledge Graphs Fueling the Ascent of Agentic AI in Enterprise
However, this rapid technological advancement is not without its challenges. Experts caution that the substantial energy demands of large AI models could place considerable strain on electricity grids and exacerbate carbon emissions, underscoring the critical need for renewable energy sources to power this evolving AI infrastructure. As agentic AI matures, it is expected to fundamentally redefine human-technology interaction, shifting towards intelligent collaboration where technology anticipates human intent rather than merely reacting to commands.
