Semester Kickoff
Can we actually make AI safe?
Join AI Safety Saarland for our Semester Kickoff — an evening of talks, panel discussion, and open conversation with three researchers working at the cutting edge of AI safety, security, and ethics.
🎙 Speakers
• Dr. Christoph Endres (sequire technology) Co-discoverer of Indirect Prompt Injection — one of the most critical vulnerabilities in modern LLMs — and founder of sequire technology, building defenses for AI systems under the EU AI Act.
• Felix Jahn (DFKI / RAIME) PhD researcher at Kevin Baum's Responsible AI and Machine Ethics group at DFKI, and first author of the IASEAI 2026 paper on reason-based neuro-symbolic architectures for safe and ethical AI alignment.
• Adam Dziedzic (CISPA / SprintML) Tenure-track faculty at CISPA, co-leading the SprintML lab. His research focuses on trustworthy machine learning — privacy, robustness, and data provenance for generative AI models.
🗣 Panel Discussion After the talks, all three speakers will join a moderated panel tackling questions like: — Is a human in the loop enough, or is that just safety theater? — LLMs are already deployed everywhere. Are we too late to make them safe? — Is the EU AI Act helping or hurting real AI safety work? — What's the difference between AI safety, AI security, and AI ethics? — If you could fix one thing about how AI is being built today, what would it be?
📍 E2.5 Hörsaal 1, Saarland University, Campus Nord 🗓 Thursday, April 16th ⏰ 16:15 – 20:00
Free entry. Open to everyone — students, researchers, and anyone curious about where AI is heading and whether we're building it safely.
Networking with drinks after the panel. See you there!