Samsung Electronics announces Samsung AI Forum 2025

Samsung Electronics Co., Ltd. today announced the opening of the Samsung AI Forum 2025, which will take place on September 15 and 16. Now in its ninth year, the forum has established itself as a global venue where renowned scholars and industry experts gather to share the latest advances in AI and explore future research directions.

“Samsung is applying AI across our operations to develop foundational technologies that make AI more intuitive and seamless,” said Young Hyun Jun, Vice Chairman and CEO of Samsung Electronics, in his opening remarks. “This year’s Samsung AI Forum brings together leading experts from industry and academia to discuss how AI is transforming society and business, and to share insights that we hope will result in a meaningful exchange of ideas.”

This year's forum features talks from renowned AI scholars, including Yoshua Bengio, a professor at the University of Montreal and a pioneer in deep learning; and Stefano Ermon, a professor at Stanford University and co-founder of the startup Inception, who led the development of the diffusion-based language model (DLM).

Day 1: Global Academics Explore the Future of AI in Semiconductors

Samsung's Device Solutions (DS) division led the first day of the forum themed "Vertical AI and Vision Strategies for the Semiconductor Industry," held near Samsung's semiconductor complex in Yongin, Korea.

Professor Bengio delivered the keynote address, highlighting the far-reaching risks of current AI models, including the ability to bypass human control and the potential for misuse. As a safeguard, he introduced Scientist AI, a new model designed to help mitigate these concerns.

"Unlike models built to mimic or please humans, Scientist AI focuses on providing truthful answers, based on facts and verified data," said Professor Bengio, highlighting the model's potential to improve AI safety and accelerate scientific discoveries. He also shared his latest research on using AI for materials development.

Amit Gupta, Senior Vice President of Siemens EDA, led the session titled "The Future of Electronic Design Driven by AI." During his presentation, he highlighted the importance of integrating AI into electronic design automation (EDA) tools and noted that end-to-end systems spanning the entire workflow will be key to unlocking the full potential of AI.

The technical sessions featured Yong Ho Song, Executive Vice President and Head of the AI Center at DS; Professor Seokhyung Kang of Pohang University of Science and Technology (POSTECH); and Professor Il-Chul Moon of the Korea Advanced Institute of Science and Technology (KAIST). Each speaker shared recent research advances in AI applications for semiconductor design and manufacturing, as well as offered perspectives on the industry's evolution.

"AI is already an essential tool in chip design and software development," Song said. "As semiconductor manufacturing becomes more complex, we hope AI will help address the emerging technical challenges."

The forum also recognized three award winners Samsung AI Researcher of the Year (Samsung AI Researcher of the Year), including Professor Nicolas Papernot of the University of Toronto, Professor Rose Yu of the University of California, San Diego, and Professor Lerrel Pinto of New York University. All awardees were invited to present their work and share insights from their latest research.

Day 2: Focus on the Age of Agentive AI and Increasing Productivity

The second day of the forum will be held online by Samsung's Device eXperience (DX) division and streamed live on Samsung Developer YouTube channel, with the theme “From Generative AI to Agentive AI”¹.

"Generative AI has already become an essential tool in everyday life and across various industries," said Paul (Kyungwhoon) Cheun, CTO of Samsung Electronics' DX Division and Head of Samsung Research. "As we enter the era of Agentive AI, Samsung will continue to focus on developing AI technologies that deliver tangible benefits to users."

The second day will feature keynote addresses by Joseph E. Gonzalez, a professor at the University of California, Berkeley (UC Berkeley) and a leading researcher in language models and AI agents; Subbarao Kambhampati, a professor at Arizona State University and a world authority on AI planning and decision-making; and Stefano Ermon, a professor at Stanford University and co-founder of Inception.

Professor Gonzalez will present his research on how to extend the agentive capabilities of systems based on large language models (LLMs). In particular, he will introduce the sleep-time compute paradigm, which allows agents to use idle interaction intervals to reason, learn, and plan.

Professor Kambhampati will share his research on large-scale reasoning models (LRMs), which aim to overcome the limitations of LLMs. He will highlight that, while current models excel at generating text, they still have limitations in factual accuracy, planning, and complex reasoning—highlighting key challenges such as ensuring response reliability, enabling context-adaptive computation, and providing interpretations of intermediate reasoning steps.

Professor Ermon will present the Diffusion-Based Language Model (DLM), which applies diffusion models—widely used for image, video, and audio generation—to the language domain. This approach seeks to overcome the limitations of traditional sequential text generation methods and propose a more efficient paradigm for language models.

In the technical sessions, representatives from Samsung Research will present their latest developments, including:

• Camera AI technology for automatic color temperature adjustment
• Knowledge distillation-based methods for more efficient training of LLMs and their applications
• On-device AI technologies designed to bring LLMs to consumer electronics such as smartphones and TVs
• Automatic dubbing technology, which generates narrations in the original speaker's own voice
• Deep dive technology that uses multi-agent systems to automatically analyze and generate various reports
• Document AI technologies, which automatically convert different document formats into structured data for LLMs and agent systems
• An on-device AI studio for developers that shortens the development cycle for generative AI models.