Boris Köpf

About

I am a scientist at Azure Research (formerly the Confidential Computing group at Microsoft Research), where I work on techniques for tracking information flow in microarchitecture and machine learning systems. Prior to joining Microsoft in November 2018, I was a tenured faculty at the IMDEA Software Institute, a postdoc at the Max Planck Institute for Software Systems, and a Ph.D. student at ETH Zurich.

Recent Work

Machine Learning Security and Privacy

Securing AI Agents with Information Flow Control (arXiv '25) [pdf]
Permissive Information-Flow Analysis for Large Language Models (arXiv '25) [pdf]
Closed-form Bounds for DP-SGD against Record-level Inference Attacks (USENIX Security '24) [pdf]
Bayesian Estimation of Differential Privacy (ICML '23) [pdf]
SoK: Let The Privacy Games Begin! A Unified Treatment of Data Inference Privacy in Machine Learning (S&P '23) [pdf]
(more)

Microarchitectural Security

Principled Microarchitectural Isolation on Cloud CPUs (CCS '24) [pdf]
Gaussian Elimination of Side-channels: Linear Algebra for Memory Coloring (CCS '24) [pdf]
Speculation at Fault: Modeling and Testing Microarchitectural Leakage of CPU Exceptions (USENIX Security '23) [pdf]
Hide and Seek with Spectres: Efficient discovery of speculative information leaks with random testing (S&P '23) [pdf]
(more)

Upcoming Events

uASC 2026, Microarchitecture Security Conference (PC member)
USS 2026, Usenix Security Symposium (PC member)
S&P 2026, IEEE Symposium on Security and Privacy (PC member)
(more)

Contact

email: boris.koepf@microsoft.com