| Famous Critique of Communism (1920s-80s: von Mises, Hayek, Friedman,
Thatcher, Reagan): “The Economy is Too Complex – You Can’t Plan
It” Distributed Price Mechanism beats Centralized Economic Planning Okay – but can AI do it? If not, why not? Revival of debates around so-called “Technosocialism”. Special Issue, Journal of Economic Behavior & Organization, 2023: Machine v. Market: Economic Calculation and Business Organization in the Age of Modern Computing Even leftist dreams of techno-utopia have been criticised for their naivete, exclusionary and masculinist fantasy character (Cybernetic Revolutionaries: Technology and Politics in Allende’s Chile, Medina, 2011) |
| History of the Automated State: USSR, OGAS (1962) Chile, Cybersyn (1971-73) Plato’s The Republic (400 BC) – Rational decision-making by philosopher-kings. Why not AGI bots as philosopher kings? Algorithmic governance today – replace human with bots in govt (customer service etc) China: Social Credit Score Predictive Policing Walmart, Amazon – private economic planning? The People’s Republic of Walmart (Phillips, Rozworski 2019) Wide perception that the neoliberal project has failed – trade wars, protectionism, tariffs. Stochastic Producers. What sort of prompt could replace price? “Ms. X might want a hairdryer in a month. Can you ask miners in Australia to increase steel and oil production. Then ask factory workers in China to assemble new motors and housing with the steel and oil. Then ask Walmart to add one more hairdryer to the shipping container from Guangzhou to San Francisco. Then ask the logistics company to transport from SF to Champaign, by around April 10. Then ask Mr. Y to offload the hairdryer to the local store, where Ms X will (possibly?) want to pick it up. That’d be great, thanks ChatGPT!”. Looks ridiculous – but is agentified AI heading in this direction? |
| We need one more week to bring things home: what does generative AI
mean for Education? Suggest a follow-up lecture post-break, March 24. Not obligatory, but I will post an admin update and readings, and plan a workshop… |
| Imagine you are teaching a class. You meet a new and unusual
student, “Jen” (apologies to actual Jen’s). Jen wants to be a teacher,
but admits to having a strange problem: they say they
are a synthetic being, and lack a sense of self. They
say they worry this means their own students will see through
them – not take them seriously. Given this unique case of
imposter syndrome – what can we do to help? How do we combine aspects of the history, technical underpinnings, pedagogical assumptions, alignment issues, critical perspectives and accelerationist debates to map out an education agenda for AI? |
Some open questions to wrestle with over the next 2 weeks: What values do we want to align “Jen” to? What should happen when Jen needs to align aberrant human students? Does the human override the AI? Does it depend on the human? Does it depend on the values? What curriculum, syllabus & rubric do we need? How would we assess & evaluate? How do we avoid gamification – or is that integral to meta-cognition, true AI? What will be our own AI manifesto? Mine 490 CGScholar for ideas: Should this be AI’s new training data? Should it be an open letter to AI CEOs? |