Generative Ai is a still new and emergent technology capable of producing not only text that could be mistaken as human-generated, but also images, video, music, and “voice.” For all of the amazing opportunities opened up by generative AI, however, it does not come without its own risks. Secondary and post-secondary education, for example, was thrown into crisis in late 2022 when ChatGPT was released, and is still weathering that storm. Meanwhile, other AI models, known as “diffusion models” (which generate audio, images and video) have also been getting more sophisticated at a lightning pace. Yet, the average internet user has very little knowledge of how generative AI works, and far less the skills to distinguish its outputs from human-generated content.

Especially in an election year, should we worry about the circulation of products that generative AI models generate? What are the implications of the rapid and wide-spread proliferation of fake news and deepfakes? How do we guard against the “feedback loop” problem in generative AI learning models?

This week, we try to explain and de-mystify generative AI  in order to get to the root of what we should be concerned about and what we shouldn’t.

In this episode, we discuss the following thinkers/ideas/texts/etc:

• Florida House Bill 433
• The Radio Birdman
• Google AI search problems
• ChatGPT
• Speechify
• Large Language Models (explainer) – David J. Gunkel, AI for Communication (Routledge 2024)
• Diffusion Models (explainer) – David J. Gunkel, AI for Communication (Routledge 2024)
• “Feedback Loop” problem
• The many problems with PREDPOL (Predictive Policing)
• Facebook’s Artificial Intelligence Research Experiment (“Bob” and “Alice”)
• AI “Ghost Workers”
• Our season 10, episode 137 on “Originality”
• The “Lovelace Objection”
• Charles Babbage
• Emily M. Bender, Timnit Gebru, Angelina McMillan-Major, Shmargaret Shmitchell. “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?”
• Mark Coeckelbergh and David J. Gunkel. “ChatGPT: Deconstructing the Debate and Moving it Forward”
• G. W. F. Leibniz, Characteristica Universalis
• Ludwig Wittgenstein, Tractatus Logico-Philosophicus (1921)
• Ferdinand de Saussure , Course in General Linguistics
• “The Grand Academy of Legado” in Jonathan Swift’s Gulliver’s Travels (1726)
• Roland Barthes, “Death of the Author” (1967)
• Springer Nature – Editorial Policies
• APA reference for GenAI content
• APA’s guide to How to Cite ChatGPT
• Michel Foucault, “What is an Author?” (1969)
• Stephen Marche, “The College Essay is Dead” (The Atlantic, 2024)
• Technological Unemployment
• Paleography
• Cathy O’Neil, Weapons of Math Destruction
• Safiya Umoja Noble, Algorithms of Oppression
• Plato, Phaedrus