Send email Copy Email Address

Email

Address

Stuhlsatzenhaus 5
66123 Saarbrücken (Germany)

Awards (selection)

2023: ACM Fellow

2019: Karl-Heinz Beckurts Prize

2018: IEEE Fellow

2018: Honorary Doctorate

2017: NSA Cybersecurity Research Award

2017: IEEE Golden Core Award

2017: CNIL-INRIA Privacy Award

2014: ERC Synergy Grant

2014: IEEE Outstanding Community Service Award

2011: IEEE Outstanding Community Service Award

2009: MIT TR35

2009: ERC Starting Grant

2008: IBM Faculty Award

2007: Fellow of the German Max Planck Society

2005: IBM Outstanding Achievement Award for Linking Formal Verification and Cryptography

2005: IBM Research Division Award for contributions to Web Services Security

2004: Microsoft Award for outstanding research in privacy enhancing technologies

2004: IBM Outstanding Achievement Award for contributions to Enterprise Privacy Architectures

2002: VDI Diploma Thesis Award

2001: Scholarship of the DFG graduate studies program ”Quality Guarantees for Computer Systems”

Short Bio

Michael Backes is the founding director and CEO of the CISPA Helmholtz Center for Information Security.


In his research, Prof. Backes focuses primarily on trustworthy methods for machine learning, on novel approaches for protecting personal data, and on universal solutions for software and system security. His research results have shaped these scientific fields internationally, as is evidenced by over 300 peer-reviewed and highly cited publications in renowned international journals and conference proceedings.

His research has earned him internationally renowned scientific awards and honors, in particular the ERC Synergy Grant, which is Europe's most highly endowed EU group research award, the Karl Heinz Beckurts Prize, an ERC Starting Grant, the Caspar Bowden Privacy Award, the IEEE and ACM Fellowship, as well as various Career Awards and Best Paper Awards.

His research results have regularly been translated into highly innovative applications that have brought him widespread attention beyond the scientific community. In particular, Prof. Backes was the first German to receive the MIT TR35 Award, which recognized him as one of the top 35 researchers and innovators in the world under the age of 35 (across disciplines) whose “accomplishments are poised to have a dramatic impact on the world as we know it”.

Michael Backes is regularly listed in rankings as one of Germany's most important IT personalities. He holds an honorary doctorate from the Université de Lorraine and is both honorary citizen and future ambassador of the city of St. Ingbert.

 

 

CV: Last stations

Since 2017
CISPA Helmholtz Center for Information Security - Founding Director and Chairman (CEO)
Since 2005
Saarland University - Professor
2011-2018
CISPA, Saarland University - Scientific Director
2007 - 2017
Max Planck Institute for Software Systems - Max Planck Fellow

Publications by Michael Backes

Year 2026

Conference / Medium

Annual Meeting of the Association for Computational Linguistics (ACL)

Conference / Medium

Annual Meeting of the Association for Computational Linguistics (ACL)
The Art of (Mis)alignment: How Fine-Tuning Methods Effectively Misalign and Realign LLMs in Post-Training

Conference / Medium

International Conference on Machine Learning (ICML)
Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs

Conference / Medium

AAAI 2026 Workshop on AI Governance
Frequency-Domain Model Fingerprinting for Image Autoregressive Models

Conference / Medium

ICLR 2026 Workshop: Principled Design for Trustworthy AI

Conference / Medium

Annual Meeting of the Association for Computational Linguistics (ACL)
InferPilot: Autonomous Inference Attacks Against ML Services With LLM-Based Agents

Conference / Medium

Annual Meeting of the Association for Computational Linguistics (ACL)
Rethinking Assessments of Prompt Injection Attacks

Year 2025

Conference / Medium

Conference on Neural Information Processing Systems (NeurIPS)
Adjacent Words, Divergent Intents: Jailbreaking Large Language Models via Task Concurrency

Conference / Medium

Conference on Neural Information Processing Systems (NeurIPS)
Finding and Reactivating Post-Trained LLMs’ Hidden Safety Mechanisms

Conference / Medium

Conference on Empirical Methods in Natural Language Processing (EMNLP)
Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification