[lamarr-nlp] Guest Talk by Thilo Hagendorff from University of Stuttgart | From Machine Psychology to AI Safety

Wednesday 12 Aug 2026, 11:00 → 12:00 Europe/Berlin

As part of the Lamarr NLP Colloquium, we have the pleasure to host Thilo Hagendorff from University of Stuttgart. Thilo will give a talk on how he moved from machine psychology to AI safety.

Title: From Machine Psychology to AI Safety

Abstract:

LLMs and LLM-based agents are currently at the forefront of integrating AI systems with human communication and everyday life. For this reason, it is of great importance to thoroughly assess and scrutinize their capabilities. This talk will provide an overview of how I moved from "machine psychology" - where LLMs are treated as participants in psychology experiments - to AI safety. I will focus on research projects covering phenomena that are core problems of AI safety, including deception, emergent misalignment, behavioral self-awareness, evaluation awareness, jailbreaking, and more.

 

Bio:
Thilo Hagendorff is a Research Group Leader at the University of Stuttgart. Previously, he worked at the Cluster of Excellence "Machine Learning" at the University of Tuebingen. He was a visiting scholar at Stanford University, UC San Diego, and the European Laboratory for Learning and Intelligent Systems (ELLIS) in Alicante.

 

Looking forward to your participation.

Date: Wednesday, August 12, 2026
Time: 11:00 - 12:00 pm (CET).
Location:  Friedrich-Hirzebruch-Allee 6, 53115 Bonn, Germany
Room: 2.122 + Zoom
Zoom:  https://uni-bonn.zoom-x.de/j/63819604806?pwd=64PSGa9HyTym9j1bjy6jhcJF3eHebi.1