Show Notes
Josh and Ray talk about the real risks of AI with Anna Colom, a social science researcher working at the intersection of democratic processes and digital technologies. We are going through unprecedented times in the pace and scale of technological development driven by data and AI systems. Yet, these developments and their applications are being led and decided by a few in a regulatory vacuum, supported by narratives driven by hype and power asymmetries. There is however a large and more diverse community of technology developers across the globe who can claim back the authority if it refuses to be blinded by hype; if it questions decisions made by a few; if it favours evidence and ethics over magic; and if it works collaborative with the rest of society to answer some of the most pressing questions of out times on how data and AI can be responsibly used to help people make this world a better place. Mentioned in this episode:From hype to responsibility: what works & matters for whom in data and AI? - Anna Colom's keynote at Heart of Clojure 2024 [talk description] Isabelle Ferreras - Professor UC at Louvain University who has done work on the democratisation of companies and organisations Organising Tech in Sweden: The Union Perspective - Per Holfve, an advisor with the Swedish Association of Professional Scientists, talks about how scientists fit into labour organising in tech [PERMALINK] Dr. Joy Buolamwini - founder of the Algorithmic Justice League, an award-winning researcher, and poet of code AI, Ain't I A Woman? - a spoken word piece that highlights the ways in which artificial intelligence can misinterpret the images of iconic black women, by Dr. Joy Buolamwini
We'd love to hear from you if you disagree with something we've said. Please read our rules of engagement for good faith discussions, and then send an email to [email protected] or hit us up on socials. To view full show notes, including transcripts, please visit the episode page. Be sure to follow us on BlueSky and Mastodon! Cover art by Anya K. Jordan @anyakjordan.bsky.social Theme music by Xylia Politechs is provided under the terms of the Attribution-NonCommercial-NoAI licence
More from Politechs
Paris Marx, host of the Tech Won't Save Us podcast, lays out a vision for an alternative model of technology that serves the people, rather than the shareholders of a handful of US tech companies. We talk about how the Trump regime's overreach (unpredictable foreign trade policies, kidnapping the leader of a sovereign state, escalating support for the Zionist entity's genocide in Gaza, an unprovoked and illegal attack on Iran, including the assassination of Iran's supreme leader, his family, and other high-ranking officials of the Iranian government) has presented us with a unique opportunity for an internationalist response that rejects US tech and builds a new technical stack in service of the public good. Mentioned in this episode:Road to Nowhere - book by Paris Marx [permalink] Hyperscale - book by Paris Marx [permalink] Tech Won't Save Us - podcast by Paris Marx Disconnect - newsletter by Paris Marx Ludicrous: The Unvarnished Story of Tesla Motors - book by Ed Niedermeyer [permalink] See the transcript footnotes for more! We'd love to hear from you if you disagree with something we've said. Please read our rules of engagement for good faith discussions, and then send an email to [email protected] or hit us up on socials. To view full show notes, including transcripts, please visit the episode page. Be sure to follow us on BlueSky and Mastodon! Cover art by Anya K. Jordan @anyakjordan.bsky.social Theme music by Xylia Politechs is provided under the terms of the Attribution-NonCommercial-NoAI licence
At the beginning of the season, we felt we had a pretty good handle on HELLISH AI: a Hyper-scaled Expansive Large Language model which is Imperial, Sucks up all the electricity and water, and causes Human suffering. We wondered whether a different way was possible, a HEAVENLY AI that is High quality, with Ethical curation of data, Accesible to individuals and smaller organisations, Viable, Environmentally sustainable, with Narrowly focused applications, using a Language model, that fit with the true spirit of YOLO: since you only live once, you might as well live well. In this conversation, we apply what we've learned from our guests this season to try and answer that question. Mentioned in this episode:I Am An AI Hater - essay by Anthony Moser [permalink] A Butlerian Hauntology - paper by Dr. Syed Mustafa Ali Empire of AI - book by Karen Hao British Post Office scandal - article about wrongful prosecution of British Post Office workers Library Freedom Project - a network of values-driven librarians working together to build information democracy [permalink] See the transcript footnotes for more! We'd love to hear from you if you disagree with something we've said. Please read our rules of engagement for good faith discussions, and then send an email to [email protected] or hit us up on socials. To view full show notes, including transcripts, please visit the episode page. Be sure to follow us on BlueSky and Mastodon! Cover art by Anya K. Jordan @anyakjordan.bsky.social Theme music by Xylia Politechs is provided under the terms of the Attribution-NonCommercial-NoAI licence
Friend of the show Dr. Chris Gilliard returns to catch us up on what's been happening in the world of surveillance since last time we talked. From Ring-powered "pet finder" systems to Meta/Rayban-wearing Glassholes to Anthropic's performative handwringing over their Department of War contract, it's not very encouraging out there. You might be able to guess what Chris thinks about the question of whether any of this technology can be redeemed. Mentioned in this episode:On the Impossibility of Ethical Surveillance - essay by Torin Monahan Facial Recognition is the Plutonium of AI - paper by Luke Stark [PDF] Cute videos, but little evidence: Police say Amazon Ring isn't much of a crime fighter - article by Cyrus Farivar [permalink] Careless People - book by Sarah Wynne-Williams The TESCREAL bundle: Eugenics and the promise of utopia through artificial general intelligence - paper by Timnit Gebru and Emile P. Torres [permalink] See the transcript footnotes for more! We'd love to hear from you if you disagree with something we've said. Please read our rules of engagement for good faith discussions, and then send an email to [email protected] or hit us up on socials. To view full show notes, including transcripts, please visit the episode page. Be sure to follow us on BlueSky and Mastodon! Cover art by Anya K. Jordan @anyakjordan.bsky.social Theme music by Xylia Politechs is provided under the terms of the Attribution-NonCommercial-NoAI licence
Dr. Dan McQuillan, Senior Lecturer in Critical AI at Goldsmiths, University of London, defines Decomputing as "a 50:50 hybrid of decolonial and degrowth approaches." He joins us to unpack that further and pick up on some of the themes from his prescient book from June of 2022, "Resisting AI: An Anti-fascist Approach to Artificial Intelligence", which came out a mere four and a half months before OpenAI unleashed ChatGPT on the world. Mentioned in this episode:Resisting AI: An Anti-Fascist Approach to Artificial Intelligence - book by Dan McQuillan [permalink] Tools for Conviviality - book by Ivan Illich danmcquillan.org - Dan McQuillan's blog [permalink]
Josh and Ray are joined by computational linguist Dr. Emily M. Bender to discuss what "AI" really is, and why the slippery nature of the term is very much intentional. We look at the eugenicist history of the quest to define and measure human intelligence and try to understand what the project of "artificial intelligence" is, then explore LLMs as synthetic text extruding machines and automation technology, as opposed to a machine god on the cusp of being born. Mentioned in this episode:Mystery AI Hype Theater 3000 - podcast by Drs. Emily M. Bender and Alex Hanna [permalink] THE AI CON - How to Fight Big Tech’s Hype and Create the Future We Want - book by Drs. Emily M. Bender and Alex Hanna [permalink] On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜 - paper by Emily M. Bender, Timnit Gebru, et al. [permalink] Ghost in the Machine - film by Valerie Veatch [permalink] The TESCREAL Bundle - paper by Timnit Gebru and Émile Torres [permalink] See the transcript footnotes for more! We'd love to hear from you if you disagree with something we've said. Please read our rules of engagement for good faith discussions, and then send an email to [email protected] or hit us up on socials. To view full show notes, including transcripts, please visit the episode page. Be sure to follow us on BlueSky and Mastodon! Cover art by Anya K. Jordan @anyakjordan.bsky.social Theme music by Xylia Politechs is provided under the terms of the Attribution-NonCommercial-NoAI licence