ScaDS.AI Dresden/Leipzig announces and welcomes you to join its public colloquium session on Thursday, April 03, 2025 at 11 am CEST. The colloquium takes place at seminar room “Collmberg” at ScaDS.AI Dresden/Leipzig (details below) and parallel online (link to Zoom session).

Dr. Patrick Ebel (ScaDS.AI Leipzig, CIAO junior research group leader) invites

The Human-AI Integration Challenge: Designing a Future We Can Trust

How will spring 2030 feel AI-wise? How will we look back at 2025? And how do we now reflect on 2020 – when few predicted the mass adoption of ChatGPT et al.? Humans struggle to predict future states of human-technology relationships. Could this become a real problem within the coming years? This talk takes a 70-year retrospective on one of the biggest predictive failures in engineering psychology – the Fitts List, a foundational framework for human-machine function allocation. It shaped how we design automation, but it also represents a monumental ‘miscalculation’. To understand why, I will try to deliberately recreate this mistake in my talk – examining why psychology, despite its predictive ambitions, struggles with nonlinear technological disruption. We often say past behavior is the best predictor of future behavior – but what happens when AI reshapes the very nature of human-technology interaction? 

Through insights from research of my team and me, I will explore:

• Key paradigms of Human-AI coupling (Cooperation | Teaming | Partnership)
• Two case studies of AI-driven ‘empowerment’:
  • Hybrid closed-loop AI in diabetes management – real-time AI-assisted decision-making.
  • Assistive AI in energy-efficient driving – helping humans navigate complex multi-criteria action regulation.

For both contexts, I briefly (i) present the simulation environments we have developed at the University of Lübeck and (ii) our theoretical modeling approaches that may help us to formulate a very first idea on what an empowering AI could look like.

This talk wants to inspire discussion:

• How do we design AI that doesn’t just function but aligns with human needs & values?
• What will it take to make AI not just intelligent but truly trustworthy & relatable?
• And when AI reaches a pivotal threshold – how do we ensure that humans remain an inseparable part of the system?

About Prof. Dr. Thomas Franke

Thomas Franke is a full professor of Engineering Psychology and Cognitive Ergonomics at the University of Lübeck. His research focuses on understanding and supporting resource regulation in human-technology environments. Herein he focuses on theory and method development and topics related to user diversity and effects of interface design on user experience and behavior.

Location

ScaDS.AI Dresden/Leipzig
Löhrs Carré, Humboldtstrasse 25, 04105 Leipzig
5th floor, large seminar room (B 05.23 “Collmberg”)