Michael Backes

Michael Backes is the Founding Director and Chairman of the CISPA Helmholtz Center for Information Security and a Professor at Saarland University. His research focuses on Trustworthy Machine Learning, Data Privacy, Software Security, and Information Security. He leads a prominent research group in AI Security and Privacy, with extensive collaborations including the CISPA-Stanford Center.

Research Interests

• Trustworthy methods for machine learning
• Novel approaches for protecting personal data
• Universal solutions for software/system security
• AI-generated content detection
• Privacy-preserving techniques
• Adversarial machine learning

Publications & Research Trends: His recent work examines LLM security, data leakage detection, privacy risks in conversational AI, and ethical challenges in vision-language models. He has pioneered benchmarks like HateBench and MGTBench to evaluate AI security.

Scientific Awards & Honors

• ACM Fellow (2023)
• IEEE Fellow (2018)
• ERC Synergy Grant (2014)
• MIT TR35 Award (2009)
• NSA Cybersecurity Research Award (2017)
• Caspar Bowden Privacy Award (2004)
• VDI Diploma Thesis Award (2002)

Advising & Leadership: He has mentored over 40 PhD students, many of whom now hold tenured or leadership positions at institutions like Vrije Universiteit Amsterdam, CISPA Helmholtz Center, and Microsoft Research. His group offers postdoc fellowships with mentorship, funding, and collaboration opportunities.

Labs & Teams: Leads the AI Security and Privacy group at CISPA, focusing on foundational and applied research. The team collaborates with institutions like Stanford University and Saarland University.