VERSEN event on Generative AI and Ethics

When: November 20th 2023
Where: University of Utrecht. Janskerkhof 3, 3512 BK Utrecht, The Netherlands

Generative Artificial Intelligence (AI) is an emerging technology that gained a lot of attention recently. Related bots such as ChatGPT represent great opportunities for businesses, developers and users interested in sophisticated and personalised conversational experiences, automated content creation, marketing and analytics, just to name a few.

This half-day workshop is intended for lecturers, researchers, students and IT professionals who are interested in generative AI and its potential impact on their respective fields. We hope this meeting will provide the perfect setting to understand more about generative AI, including current and future developments in the field, and to support lively and interactive discussions on the opportunities, ethical and legal challenges, and impact of Generative AI technologies.

The workshop will host two invited talks, and a panel discussion.

Speakers

Jakub Tomczak (Eindhoven University of Technology)
Generative AI: Why? How? What?

Large Language Models and image generators showed that Generative AI is the way to go in building a new generation of AI systems. But is it a temporary trend about generating synthetic data, or there is more to it and the future of AI is generative? In this talk, we will first look into why Generative AI is an important paradigm. Next, we will discuss how we can build (deep) generative models. Eventually, we will point out what we do with Generative AI. The talk will conclude by indicating the challenges Generative AI needs to face in the near and far future

Ibo van de Poel (Delft University of Technology)
What should we worry about in (generative) AI, and how might we (try to) do better

AI and its applications raise a host of ethical issues. Not just privacy concerns, but also issues related to fairness, explainability, potential harm, sustainability, and misuse. In my talk I will focus on some more specific ethical issues in relation to generative AI. For example, the results of generative AI may be (implicitly) biased due the underlying datasets. Applications like ChatGPT also raise questions about the reliability of results (and how to judge these). Sometimes, chatbots using generative AI may give wrong or harmful advice. AI applications may also have invisible costs in terms of sustainability and bad labor conditions in low-income countries. Finally, generative AI may challenge existing notions of creativity, authorship, and intellectual rights. I will suggest that coming to terms with these various ethical issues requires several things, including other design choices with respect to AI, attention for the political economy of AI, new user practices and skills, and broader societal discussion about issues like authorship, creativity, and plagiarism.

Registration

Participation is free of charge. Looking forward to receiving your registrations before November 15 at:

https://docs.google.com/forms/d/e/1FAIpQLScHw1gNG3j5723ClQbH9Rpg4KWfztvfw-C21OkQXKaigKjPtA/viewform

Program

12:00 - 13:05 Lunch & Welcome
13:15 - 14:00 Generative AI: Why? How? What? (Jakub Tomczak)
14:00 - 14:15 Short break
14:15 - 15:00 What should we worry about in (generative) AI, and how might we (try to) do better (Ibo van de Poel)
15:00 - 15:30 Coffee break
15:30 - 16:45 Panel (Emitzá Guzman Ortega, Hieke Keuning, Jakub Tomczak, and Ibo van de Poel)
16:45 - 18:00 Wrap-Up & Drinks

The VERSEN Community Events:
Önder Babur (WUR and TU/e)
Georgiana Caltais (U Twente)
Loek Cleophas (TU/e and research school IPA)
Samira Shirzadehhajimahmood (UU)
Mauricio Verano Merino (chair, VU and TU/e)\