Christian Nickel

Hey there, my name is Christian Nickel. I am a full-time PhD researcher at the Research Center Trustworthy Data Science and Security (RC-Trust) and the University of Duisburg-Essen, working in Nils Köbis' group Human Understanding of Algorithms and Machines (HUAM).

Grounded in Computer Science and fascinated by Artificial Intelligence, I explore the interplay between human and AI agents.

This interdisciplinary research is in increasing demand, as it is crucial to AI safety and effectiveness. What if an agent misunderstands its users' goals, becomes misaligned, behaves sycophantically, or even deceives? What if results from an in-silico experiment mislead decision-makers due to poor simulation fidelity? Could hybrid systems of natural and artificial agents - such as financial markets - become unstable? How do these factors influence human behavior? In highly consequential domains like embodied AI, these failure modes can have dire consequences.

Thus, on the one hand,

• I investigate how AI agents model the goals, knowledge, strategies, and emotions of others (Theory of Mind, ToM),
• how they can be steered according to distinct personality traits,
• how they stay aligned in scenarios with complex incentives, and
• how model reasoning affects the above.

On the other hand, a key focus is rooted in human behavior,

• understanding how interaction dynamics and strategies of both human and artificial agents shift depending on the personality and nature of the counterpart - whether biological or artificial and
• ultimately how outcomes change in hybrid agent scenarios.

To that end, I draw from an interdisciplinary toolbox spanning Computer Science, behavioral science, and economics.

My goal is to understand how mixed-agent scenarios play out while also accounting for human behavior, how to build steerable, safe, and trustworthy AI agents, and in which contexts to deploy them. I further aim to understand how they could be leveraged for simulation and strategizing to inform executive and political decision-making - ultimately laying the groundwork for understanding the organizational and societal implications of the agent economy.

Before my time as a PhD researcher, I obtained my Master's in Computer Science from the University of Bonn. During my studies, I worked at the Fraunhofer Institute for Applied Information Technology (Fraunhofer FIT) on Distributed Ledger Technology (DLT/Blockchain).

Laura Schrewe and I recently received the Best Poster Award at EMNLP's WiNLP workshop for our work on Artificial Theory of Mind.

I am passionate about interdisciplinary approaches that push the boundaries of AI and its applications, and I look forward to collaborating with researchers and practitioners across all relevant fields.