3 minute read

posted: 28-Jan-2025 & updated: 31-Jan-2025

You can find the abstract and slides of the keynote presentation here.
Want to share this reflection? Use this link – https://k-privateai.github.io/seminar-reflections/03 – to share!

The third Silicon Valley Private AI Forum (K-PAI) seminar exceeded all expectations, marking its most successful gathering to date. The event, hosted at Stanford University’s Graduate Community Center, drew an impressive turnout of nearly 100 participants, showcasing the growing interest in AI’s multifaceted impact across industries and academia.

The evening kicked off with Sunghee’s comprehensive keynote on “The AI Knight Rises - From Deep Learning to Flourishing Societies,” which masterfully wove together diverse perspectives on AI’s evolution – from technical breakthroughs to market dynamics and societal implications, as it echoed themes recently explored at the Salzburg Global Seminar’s “Technology, Growth, and Inequality: The Case of AI” working group. The attendees particularly appreciated how the presentation brilliantly connected technical breakthroughs with broader societal implications, tracing AI’s journey through Big Data, Deep Learning, Transformers, LLMs, generative AI, and ultimately to AI Agents. A key insight that resonated strongly with the audience was the speaker’s observation that AI is rapidly becoming – or has already become – a commodity, suggesting that future technological breakthroughs and entrepreneurial success will hinge not on AI development itself, but on leveraging this powerful commodity through clear and differentiated domain expertise, with AI agents powered by advanced LLMs and multimodal generative AI playing pivotal roles in the years ahead.

After the main keynote, Professor Jung Hee Cheon, CEO of CryptoLab Inc. and distinguished professor at Seoul National University, delivered an illuminating presentation introducing SNU’s Homomorphic Cryptography Computer Research Center. His talk expertly bridged the technical and practical aspects of homomorphic computing, demonstrating how this breakthrough technology enables secure computation on encrypted data - a crucial foundation for private AI development. Using clear architectural comparisons and visual aids, Professor Cheon illustrated the evolution from traditional computing systems to homomorphic processing units (HPU), showing how this transformation enables AI applications to maintain data privacy without sacrificing functionality. The presentation sparked particular interest among attendees working at the intersection of AI and data security, generating thoughtful questions about the practical implications for future AI systems.

The highlight of the evening was the panel discussion “AI at the Inflection Point,” featuring an impressive lineup of experts including Jay Eum (CEO of GFT Ventures), Ernest Ryu (Professor @ UCLA), Youngsuk Park (Senior Applied Scientist & Manager @ AWS, Amazon), and Sunghee Yun (Co-Founder and CTO of Erudio Bio). Under the skilled moderation of Dayoung Jeong (Professor of Law @ Chungnam National University and Visiting Scholar @ UC Berkeley), the panel delivered on its promise to provide a truly holistic view of AI’s landscape including legal and ethical aspects of AI. The diverse backgrounds of the panelists - spanning venture capital, academia, big tech, and startups - contributed to a rich, multifaceted discussion that resonated strongly with the audience & societal and economic impacts.

The feedback from attendees was overwhelmingly positive, with participants especially valuing the rare convergence of investment, academic, industry, and entrepreneurial perspectives in a single forum. The networking session crackled with energy as conversations delved deeper into the insights shared during the formal program. In fact, the discussions proved so engaging that, just as in previous seminars, we had to reluctantly wind down the session as attendees remained thoroughly absorbed in their exchanges long after the scheduled end time after taking the final photos (below). ★^^★

The third K-PAI seminar achieved its ambitious goal of bridging the technical, business, and societal dimensions of AI, demonstrating how Silicon Valley’s technological innovation can be thoughtfully aligned with broader human values and social responsibility. This holistic approach, enriched by vibrant networking opportunities and elevated by an exceptional Korean fusion dining experience, has set a new standard for AI forums in the Silicon Valley. The overwhelming success of this gathering has generated unprecedented enthusiasm for the next K-PAI seminar scheduled for Feb 12, 2025, where this vital conversation will continue.

Updated: