Global artificial intelligence (AI) company Anthropic has officially opened its Seoul office, launching its push into the Korean market. With major domestic companies including Naver, Nexon, and Samsung Electronics already using Anthropic's AI model "Claude" in their operations, the company plans to expand its reach to domestic companies, startups, public institutions, and research bodies. Competition in the enterprise AI market with OpenAI, which entered Korea last year, is also expected to intensify further.
Announcing the official opening of the Seoul office Monday, Chris Ciauri, head of Anthropic International, said, "Korea is a country with a strong national-level AI vision and an AI regulatory framework, which aligns well with Anthropic's founding mission of providing humanity with a safe AI experience." Last month, Anthropic appointed Choi Ki-young as head of Anthropic Korea. Choi has handled enterprise sales at Google Cloud, Adobe, Microsoft, and Snowflake. "The starting point for the Seoul office is to stably support the AI projects that Korean companies are pursuing as a priority," Choi said. "We will collaborate to help companies make a safe AI transition, based on our sales team and partner ecosystem."
Anthropic Korea plans to broaden its influence among domestic companies, startups, and government and public institutions. In Korea, a significant number of companies had already been using Anthropic's AI model "Claude" before the Seoul office opened. A representative case is Naver. Naver recently rolled out the AI coding agent "Claude Code" across its entire engineering organization. This is considered one of the largest enterprise adoption cases in Asia. Nexon is also using Claude Code in its engineering organization to write, review, and deploy code for live-service games. LG CNS, Hanwha Solutions, and Samsung Electronics have also adopted Claude in their operations.
Anthropic Korea is also expanding its contact points with the startup and developer ecosystem. As soon as Tuesday, it will hold "Push to Prod Seoul," a hackathon for startups, together with Replit, an AI-based software development platform, and Korea Investment Partners. The hackathon, which Anthropic has operated in various parts of the world, is being held in Korea for the first time. At the event, domestic startups from various fields including AI, robotics, mobility, bio-healthcare, materials, parts and equipment, and defense will participate in teams and interact with developers from Anthropic and Anthropic Korea.
It is also strengthening cooperation with the domestic AI research ecosystem. Anthropic plans to cooperate with the National AI Research Lab (NAIRL) to support the use of Claude by its researchers. NAIRL is a research consortium that includes the Korea Advanced Institute of Science and Technology (KAIST), Korea University, Yonsei University, and the Pohang University of Science and Technology (POSTECH). Anthropic plans to provide free Claude accounts to up to 60 NAIRL researchers, supporting research on AI safety, model evaluation, and model robustness through this.
The industry views Anthropic's opening of the Seoul office as a catalyst for full-scale competition between OpenAI and Anthropic in the Korean market as well. OpenAI opened its Korea office in September last year and has strengthened sales targeting domestic companies. Recently, it has been expanding its stride into the AI transformation (AX) market in the public infrastructure sector by signing memorandums of understanding (MOU) with public institutions such as Korea Water Resources Corporation and Korea Technology Finance Corporation.
However, the fact that the U.S. government has restricted foreign access to Anthropic's latest AI models "Mithos 5" and "Fable 5" is cited as a burden. This is because the measure could also act as a constraint on domestic companies' use of U.S.-made AI models. On this, Ciauri was reticent, saying, "I believe this issue will be resolved soon."
On the other hand, following the access restrictions on Mithos 5 and others, "Project Canopy," a public-interest initiative aimed at strengthening the defenses of domestic public-interest infrastructure that is relatively lacking in security capabilities, has officially launched. Project participants have discovered hundreds of high-severity vulnerabilities, reported them to the relevant institutions and development entities, and supported the application of security patches. "AI has now become a national strategic asset beyond a simple technology," an industry official said. "This case has revealed that you can have negotiating power only when you use foreign technology while securing your own technology."
