Google DeepMind and South Korea establish AI Campus in Seoul ten years after the AlphaGo match

Google DeepMind and the Ministry of Science and ICT of the Republic of Korea (MSIT) announced on 27 April 2026 a comprehensive partnership to accelerate scientific discovery and AI innovation. The timing of the announcement is not coincidental — it coincides with the tenth anniversary of the historic AlphaGo match in Seoul, which in 2016 marked one of the key turning points in the development of modern machine learning systems.

What was agreed?

The partnership between Google DeepMind and the Korean government encompasses several concrete initiatives. The central element is the establishment of an AI Campus in Seoul within Google’s offices, which will function as a research hub for Korean academic and research institutions. Alongside the infrastructure, a collaboration with the Korean AI Safety Institute on AI safety and best-practice standards has been agreed — a particularly important component in the context of global regulatory developments.

The partnership builds on Korea’s national program K-Moonshot Missions and aligns with the opening of the National AI for Science Center (NAIS) planned for May 2026.

AI Campus in Seoul: who are the first partners?

The AI Campus will be established within existing Google premises in Seoul. The initial academic partners are:

• Seoul National University (SNU) — South Korea’s leading university
• Korea Advanced Institute of Science and Technology (KAIST)
• Three AI Bio Innovation Hubs of the Korean ministry

The campus is conceived as a permanent point of collaboration between DeepMind and the Korean academic community, not a one-off project. Google DeepMind also announces that it will explore opportunities for internships for Korean students, pointing to a long-term strategy for developing local AI talent.

Frontier AI tools opened for research

One of the key elements of the partnership is access to advanced AI tools that DeepMind is making available to Korean researchers:

• AlphaEvolve — a Gemini-powered coding agent designed to accelerate algorithmic research
• AlphaGenome — a tool for genomic mutation analysis, relevant to biomedical research
• AlphaFold — a system for predicting 3D protein structures already used by more than 85,000 Korean researchers
• AI co-scientist — a multi-agent system designed as a research collaborator for hypothesis generation and experiment planning
• WeatherNext — a model for weather forecasting and energy consumption optimization

The figure of 85,000 Korean AlphaFold users signals that this is not a new collaboration — but with this agreement it gains an institutional framework and expands to tools that were not previously available to local academic institutions.

Collaboration with the Korean AI Safety Institute

Particularly relevant is the portion of the partnership concerning AI safety. Google DeepMind has announced collaboration with the Korean AI Safety Institute on research and development of best-practice standards. Korea established its own AI Safety Institute in 2024, joining a network of similar institutions that emerged in the US, UK, and EU following global negotiations on the safe development of AI systems.

The specific forms of this collaboration are not described in detail in the announcement, but the mere mention of collaboration with a national safety body in the context of a government-industry partnership suggests this is not a symbolic gesture — particularly at a moment when global regulatory dynamics are accelerating the establishment of mandatory AI system evaluation standards.

Context: ten years since the AlphaGo match

The announcement is framed around the tenth anniversary of AlphaGo’s victory over Korean grandmaster Lee Sedol in Seoul in 2016 — one of the most-watched events in the history of AI research. For DeepMind that match represents a symbolic point from which Korean public awareness of AI progress began; for Korea it is a national cultural marker that sparked strong investment in the domestic AI industry.

The partnership being built on that foundation — and now encompassing the government, universities, an AI safety body, and the opening of physical infrastructure — shows that both sides view the partnership as strategic rather than promotional.

What comes next?

The National AI for Science Center (NAIS) opens in May 2026, meaning the first concrete activity from this partnership will be visible within a week or two. The AI Campus still needs to be operationalized, and the modalities of tool access for universities have not yet been publicly specified in detail.

Korea thereby joins the circle of countries that have signed formal national AI agreements with DeepMind or Google — alongside the United Kingdom, Japan, and several EU countries — an increasingly clear pattern of market and regulatory positioning by major AI laboratories on a global scale.