Jinseong Park

AI Fellow · Korea Institute for Advanced Study (KIAS)

Hi, I'm Jinseong Park. I am an AI Fellow (AI Assistant Professor) at the Korea Institute for Advanced Study (KIAS).

My work focuses on trustworthy machine learning (TrustML), investigating theoretical foundations and practical algorithms for Privacy- and Safety preserving Intelligence (PSI, Ψ) systems. My research extends to industrial applications by designing scalable and robust systems through trustworthy AI.

Jinseong Park

Research

🧠

Understanding of AI

We study the theoretical foundations of modern AI. Our goal is to understand how AI works.

Generative Modeling Generalization Memorization
🔒

Privacy & Safety Issues of AI

We develop privacy- and safety-preserving machine learning algorithms. Our work spans formal privacy guarantees via differential privacy, machine unlearning for generative AI, and defenses against AI attacks.

Differential Privacy Machine Unlearning Attack Robustness
📈

Industrial Applications

We apply trustworthy AI techniques to real-world industrial challenges.

Time Series Finance Manufacturing
View full research overview →

Latest News

Dec 2026 I was selected as a Top Reviewer for NeurIPS 2026, and I will attend the conference in Sydney, Australia (Dec 6–12) [Link].
Nov 2026 I will join the KIAS Center for AI and Natural Sciences Autumn Workshop 2026 in Jeonbuk, Korea (Nov 2–4) [Link].
Oct 2026 My interview with KIAS has been published! [Watch on YouTube]
Oct 2026 I will attend the Learning Theory (LeT) Workshop at POSTECH in Pohang, Korea (Oct 7–8) [Link].
Jun 2026 One paper on "diffusion bridge" was accepted at KDD 2026. See you in Jeju!
View all news →

Collaborations