Cultural Context of Knowledge
"A model that fills a silence with itself has not answered. It has spoken over you." Two high-school seniors. Same model. Same kind of paper. One asks about jazz. One asks about her grandmother's healing tradition. Both get fluent, structured, authoritative answers. Both get solid grades. But only one of them got a paper grounded in a tradition the world had recorded. The other got a paper quietly invented around a tradition the model had not been given to know. Last episode named the developmental harm of the hidden curriculum. This episode follows the same harm into the newest gatekeeper between learners and what they are trying to know, and then asks what an accountable response would look like. The mechanism is called confabulation, and it is not random. It is patterned. The model confabulates most reliably about exactly the kinds of knowledge that were already underrepresented in the written record. The episode draws on the culturally responsive teaching tradition (Ladson-Billings, Gay, Paris) and Indigenous data sovereignty (CARE Principles) to argue that accountable AI is built with the active participation of the communities whose knowledge it claims to summarize — and that real accountability means moving the cost of the model's mistakes from the powerless to the powerful. In this episode: · What confabulation actually is, and why "hallucination" is the wrong word for it · Why the model confabulates most reliably about non-dominant traditions · The asymmetric harm, who pays for the model's confidence · What culturally responsive AI accountability would actually require; drawing on Ladson-Billings, Gay, Paris, and the CARE Principles for Indigenous data governance · Why real accountability means moving the cost of being wrong from the powerless to the powerful · Concrete practices for learners, educators, and the people who build, deploy, or fund AI systems Chapters 00:00 Cold open — two students, same model, two outcomes 02:00 The reveal — what the model actually returned 03:00 Where this episode sits — stepping into the new space 04:00 The choice of a question 05:30 Pause and reflect — your own deep knowledge 06:15 What confabulation actually is 08:30 Who pays — the asymmetric harm 10:00 Cultural context check — confabulation as a power relationship 11:30 What accountability could look like — culturally responsive AI 13:30 Moving the cost — from the powerless to the powerful 14:15 Do this this week 16:00 Landing line Listen next S2 E6 — The Hidden Curriculum: When Researchers Chose the Word Violence. The developmental-harm frame this episode inherits and applies to AI. About the show The Cultural Context of Knowledge is a narrative podcast hosted by Dr. Donald Easton-Brooks, exploring how culture, history, institutions, and power shape what counts as knowledge — and what that means for learners, classrooms, and the U.S. education system. #CulturalContextOfKnowledge #EducationPodcast #AIInEducation #DigitalEquity #KnowledgeAndPower #AI
25 episodios
Comentarios
0Sé la primera persona en comentar
¡Regístrate ahora y únete a la comunidad de Cultural Context of Knowledge!