KAIST

< Poster for the Global AI Frontier Symposium 2025 (Including Schedule) >

KAIST announced that the National AI Research Lab (NAIRL) and the Global AI Frontier Lab co-hosted the 'Global AI Frontier Symposium 2025' at Seoul Dragon City on the 27th.

The symposium was hosted by the Ministry of Science and ICT and the Institute for Information & Communications Technology Planning and Evaluation (IITP), and was attended by over 500 experts from industry, academia, and research institutions, along with government officials, from both home and abroad.

Key domestic and international figures attended the event, including Deputy Prime Minister and Minister of Science and ICT, Kyungshoon Bae, and President of IITP, Jinbae Hong.

In particular, Turing Award winner and NYU Professor Yann LeCun, and Stanford University Professor and NVIDIA Senior Director Yejin Choi delivered keynote speeches on the future of AI governance and generative AI.

Researchers from five countries, including Korea, the US, Japan, France, and the UAE, participated in the symposium, sharing their international joint research achievements and presenting a blueprint for global AI collaboration.

The National AI Research Lab (Director Kee-Eung Kim) and the Global AI Frontier Lab (Co-Directors Yann LeCun and KyungHyun Cho) are leading domestic AI research as a global R&D hub, and at this event, they unveiled their major research achievements over the past year.

Deputy Prime Minister and Minister of Science and ICT, Kyungshoon Bae, urged in his opening address, "I request that the National AI Research Lab and the Global AI Frontier Lab take the lead in international cooperation and the expansion of the research ecosystem for the Republic of Korea to leap beyond an AI technology powerhouse to become a global leader."

The symposium was divided into two sessions.

Session 1, 'The Future of Next-Generation AI and Fundamental Research,' was chaired by Professor Eunho Yang of KAIST, with presentations on research topics including: ▲Data-Driven Reliability Framework (Ichiro Takeuchi, RIKEN), ▲AI-Based Drug Discovery (Tae-Eui Kim, Korea University), ▲Analysis of Training-Free VLM (Sungjae Hwang, Yonsei University), ▲Scientific Foundation Models and Semiconductor Applications (No-Seong Park, KAIST), ▲Overcoming the Compression Scaling Law (Jaeho Lee, POSTECH), and ▲Incremental Learning in Vision (Kartik Alahari, INRIA).

Session 2, 'Physical AI for Real-World Autonomy,' held simultaneously, was chaired by Professor Minsoo Cho of POSTECH, with various presentations including: ▲Non-Contrastive Self-Supervised Learning (Jean Pons, ENS-PSL), ▲Fairness-Based Offline Multi-Objective Reinforcement Learning (Jongmin Lee, Yonsei University), ▲Next-Generation 3D Vision Models (Hyunjeong Shim, KAIST), ▲Inference-Oriented Research on VLA Models (Sungjun Choi, Korea University), ▲Robustness of Vision Foundation Models Based on GaRA (Suha Kwak, POSTECH), and ▲Embodied Agents (Ivan Laptev, MBZUAI).

Experts from the industry, including Naver Cloud and POSCO Holdings, participated in each session to hold in-depth discussions on industrial applicability and industry-academia-research cooperation models.

< Photo of Professor Kee-Eung Kim of KAIST Kim Jae-chul AI Graduate School debating >

Director of the National AI Research Lab, Kee-Eung Kim, stated his aspirations, "This symposium is a significant platform to announce the research achievements of the past year to the world and expand the foundation for international cooperation. We will continue to work with industry, academia, and research institutions to create a virtuous cycle of global AI innovation."

President Kwanghyung Lee said, "The National AI Research Lab is the center of domestic AI research and a bridge connecting us to the world. KAIST will continue to lead the sustainable future of humanity with AI technology, crossing the boundaries of academia, industry, and nations."

The National AI Research Lab is the largest domestic AI research consortium of industry, academia, research, and government, established in October 2024 with the support of the Ministry of Science and ICT and IITP, led by KAIST.

It involves 45 professors and over 150 student researchers from four universities—KAIST, Korea University, Yonsei University, and POSTECH—and is accelerating the establishment of a global AI ecosystem in cooperation with 12 domestic companies, 14 overseas joint research institutions, Seoul Metropolitan Government, and Seocho-gu district. Furthermore, more than 20 foreign researchers reside for two weeks or more to conduct joint research with students, and world-renowned scholars are invited monthly to share the latest AI research achievements.

The National AI Research Lab has signed an MOU with the RIKEN in Japan and discussed expanding cooperation with AI hub institutions in Singapore last August, strengthening its role as a bridgehead for global AI governance cooperation.

< (From left) NYU Professor Yann LeCun, Stanford University Professor Yejin Choi, Global AI Frontier Lab Co-Director KyungHyun Cho, KAIST Professor Kee-Eung Kim, Deputy Prime Minister and Minister of Science and ICT Kyungshoon Bae >

The active research and dynamic ecosystem expansion efforts of the National AI Research Lab are expected to become the groundwork for the Republic of Korea to establish Sovereign AI and secure a leading position in the global AI governance discussion.