[Air-L] Invitation to Reflecting on Power and AI: The Case of GPT-3

[James Steinhoff]

SAVE THE DATE! You are invited to attend:


Workshop of the Society for Phenomenology and Media

Reflecting on Power and AI: The Case of GPT-3

Date: February 18th 2021, 09:00 – 18:45 (GMT-8), 18:00 – 19:45 (CET)

Location: Zoom, link will be provided later.



Speakers: Dr. James Steinhoff (eScience Institute at the University of Washington, USA), Dr. Olya Kudina (TU Delft, The Netherlands), Dr. Bas de Boer (University of Twente, The Netherlands)



Registration: No registration needed.



You are all cordially invited to the workshop Reflecting on Power and AI: The Case of GPT-3.



Description:

One of the latest hallmarks of current boom in Artificial Intelligence (AI) is  Generative Pre-Trained Transformer 3 (GPT-3), an unprecedentedly powerful autoregressive language model that uses deep learning to produce human-like texts ranging from historical fiction to functional code (Floridi & Chiriatti 2020). For the philosopher David Chalmers, GPT-3 is the most interesting and important AI system ever produced, and is suggested to be the first AI system displaying signs of general intelligence (Chalmers 2020). The possibilities presented by GPT-3 give rise to an array of ethical and political concerns that will be explored in this workshop: How to deal with potential biases GPT-3 inherited from training data? Can and should we use artificial language models to displace human workers? What role will language play in our lives when we know that it is often algorithmically produced? Can GPT-3 be used to eliminate existing inequalities, or is it likely to increase those? Last but not least, what can the companies behind GPT-3 do to ensure a responsible development and use of their product?

            The goal of this workshop is to explore the abovementioned issues from the perspective of two different philosophical approaches that each have a distinct outlook on the role of technologies in contemporary society but have rarely been put in conversation: Marxist media theory and the Technological Mediation Approach (TMA). In the first case, AI is approached as first and foremost an automation technology and is thus situated in the context of its commercial/industrial development and deployment. However, the Marxist approach also entails an ontological dimension insofar as it insists that capital (and its technologies) revolve around the central directive of the increase of the “real abstraction” of value, which is immaterial while configuring the concrete world. Two very different schools of Marxist theory will be drawn upon: labour process theory and value-form Marxism. In the second case,, AI can be approached as a mediating technology, through which the primary concern becomes how it shapes the way human beings understand and experience themselves and the world around them. In this workshop, we will use the framework of TMA to analyze how technologies generally shape our being-in-the-world, and point to how they do so very differently when integrated into different practices. After having outlined these two different perspectives, a plenary discussion will take place where the relevance of each of these perspectives for analysing the societal impact of GPT-3 will be explored.



Preparation:

An article will be distributed approximately one week before the date on which the workshop will take place.



Schedule:

09:00 – 09:15 (GMT-8) / 18:00 – 18:15 (CET): General introduction and an introduction into GPT-3

09:15 – 09:45 (GMT-8) / 18:15 – 18:45 (CET): Marxist media theory and GPT-3 (James Steinhoff) – Title: TBA

09:45 – 10:15 (GMT-8) / 18:45 – 19:15 (CET): The Technological Mediation Approach and GPT-3 (Olya Kudina & Bas de Boer) – Title: TBA

10:15 – 10:45 (GMT-8) / 19:15 – 19:45 (CET): Discussion with audience about the two approaches in relation to GPT-3



Bios:

James Steinhoff is a Postdoctoral Fellow at the eScience Institute at the University of Washington. His research interests include algorithmic media theory, political economy, philosophy of technology and Marxism. He is author of Automation and Autonomy: Labour, Capital and Machines in the Artificial Intelligence Industry (2021, Palgrave) and co-author of Inhuman Power: Artificial Intelligence and the Future of Capitalism (2019, Pluto).



Bas de Boer is a postdoctoral researcher at the University of Twente, The Netherlands. His research interests are in the philosophy of science technology, (post-)phenomenology, and the philosophy of medicine. Recently, his monograph How Scientific Instruments Speak: Postphenomenology and Technological Mediations in Neuroscientific Practice (2021, Lexington) was published.



Olya Kudina is an Assistant Professor in the Ethics of Technology at Delft University of Technology, the Netherlands. She explores the role of technology in the sense-making processes and the technologically-induced value change. Her research interests span across the intersection of (post)phenomenology, design, bioethics and AI.

References:

Chalmers, D. (2020). “GPT-3 and General Intelligence.” https://dailynous.com/2020/07/30/philosophers-gpt-3/#chalmers (Accessed, 18/12/2020).

Floridi, L., and M. Chiriatti. (2020). “GPT-3: Its Nature, Scope, Limits, and Consequences.” Minds and Machines 30, 681-694.



Abstracts:

TBA