In today’s conversation Ruben Puentedura acknowledges the limitations of “AI”, shares his research and experience with Retrieval Augmented Generation (RAG) technologies, and talks about the affordances of home grown LLMs (Large Language Models) combined with external resources. We are joined by Bryan Alexander and Tom Haymes.

Dr. Puentedura is the Founder and President of Hippasus, a consulting practice focusing on transformative applications of information technologies to education.

From Local to Global: A Graph RAG Approach to Query-Focused Summarization https://arxiv.org/pdf/2404.16130

Code a simple RAG from scratch https://huggingface.co/blog/ngxson/make-your-own-rag

More from Ruben Puentedura https://www.linkedin.com/in/rubenpuentedura/ Tom Haymes: https://www.linkedin.com/in/tomhaymes/ Bryan Alexander: https://bryanalexander.org/ Mark Corbett Wilson: https://www.linkedin.com/in/markcorbettwilson/

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media

In today’s conversation, I’m again joined by Bryan Alexander and Tom Haymes. We started our discussion around Tom’s recent post “The AI Revolution Will Be Small.” Tom wrote “Bigger is not better in computing technology. Small tech is where the actual power has been over the last 50 years. When we democratize computing power (or power in general), we create the seeds for actual change. In doing so, we have harnessed the power of millions of imaginations. That is where the actual power lies. Every technology goes through a phase where people can’t wrap their heads around what the new technology is for, how it works, or how it changes our social and economic paradigms. This is where we are now with Generative AI.” https://www.linkedin.com/pulse/ai-revolution-small-tom-haymes-paqac/

More from Tom Haymes: https://www.linkedin.com/in/tomhaymes/ Bryan Alexander: https://bryanalexander.org/ Mark Corbett Wilson: https://www.linkedin.com/in/markcorbettwilson/

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media

In this premier of the Talking with machines podcast, I’m joined by Bryan Alexander and Tom Haymes. Inspired by Bryan’s recent post “Some notes on how culture is responding to generative AI”, we discuss this horizon scan where Bryan asks, “How are we responding to AI through stories, art, religion, relationships, media, and symbols?” Link: https://aiandacademia.substack.com/p/some-notes-on-how-culture-is-responding More from Bryan: https://bryanalexander.org/ Tom Haymes: https://www.linkedin.com/in/tomhaymes/ Mark Corbett Wilson: https://www.linkedin.com/in/markcorbettwilson/

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media

The Western Institute for Social Research (WISR) is an accredited alternative graduate school in Berkeley, California. I created a RAG (Retrieval-Augmented Generation) using Google’s NotebookLM to collect dozens of documents and ‘chat’ about “The WISR Way” - WISR’s personalized learning method based on on the Dreyfus model of skill acquisition. I generated this podcast using NotebookLM’s podcast generator, edited all the places the ‘hosts’ mispronounced ‘WISR’ and added an introduction.

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media

Join Mark and Joe Lambert, co-founder and director of StoryCenter.org, innovators in digital storytelling (formerly the Center for Digital Storytelling). They have been leading digital storytelling workshops that employ storymapping (GIS) and “artificial intelligence” to explore possible futures for storytelling. And perhaps a surprise guest! We’ll be using digital technologies to tell stories about telling stories with digital technologies.

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media

Using Google’s NotebookLM, I added 50 eportfolio documents from my Drive. They were mostly journal articles. I included slides from a seminar I gave and a document I generated using Calude.ai. Then I used NotebookLM’s new “Audio Overview” feature to create this “podcast”. The LLM did a pretty good job of summarizing key points, but the references, especially to the slides, are made up: BS, or, bafflegab, if you prefer. Slide 37 was referenced several times. It was the end slide with the organization name and contact information. It will be a big improvement if Google adds the ability to edit the audio by editing a transcript.

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media

Earthquake!

Talking with machines by Mark Corbett Wilson @mcorbettwilson on social media