Stories

Tag: research

  • A generative AI podcast dialogue exploring The Geneva Learning Foundation’s progress in 2024

    A generative AI podcast dialogue exploring The Geneva Learning Foundation’s progress in 2024

    This experimental podcast, created in collaboration with generative AI, demonstrates a novel approach to exploring complex learning concepts through a conversational framework that is intended to support dialogic learning. Based on TGLF’s 2024 end-of-year message and supplementary materials, the conversation examines their peer learning model through a combination of concrete examples and theoretical reflection. The dialogue format enables exploration of how knowledge emerges through structured interaction, even in AI-generated content.

    Experimental nature and limitations of generative AI for dialogic learning

    This content is being shared as an exploration of how generative AI might contribute to learning and knowledge construction. While based on TGLF’s actual 2024 message, the dialogue includes AI-generated elaborations that may contain inaccuracies. However, these limitations themselves provide interesting insights into how knowledge emerges through interaction, even in artificial contexts.

    You can read our actual 2024 Year in review message here.

    Pedagogical value and theoretical implications of a generative AI conversational framework

    Structured knowledge construction: The conversational framework illustrates how knowledge can emerge through structured dialogue, even when artificially generated. This mirrors TGLF’s own insights about how structure enables rather than constrains dialogic learning.

    Multi-level learning: The dialogue operates on multiple levels:

    • Direct information sharing about TGLF’s work
    • Modeling of reflective dialogue
    • Meta-level exploration of how knowledge emerges through interaction
    • Integration of concrete examples with theoretical reflection

    Network effects in learning: The conversation demonstrates how different types of knowledge (statistical, narrative, theoretical, practical) can be woven together through dialogue to create deeper understanding. This parallels TGLF’s observations about how learning emerges through structured networks of interaction.

      We invite listeners to consider:

      • How a conversational framework enables exploration of complex ideas
      • The role of structure in enabling knowledge emergence
      • The relationship between concrete examples and theoretical understanding
      • The potential and limitations of AI in supporting dialogic learning

      This experiment invites reflection not just on the content itself, but on how knowledge and understanding emerge through structured interaction – whether human or artificial.

      Your insights about how this generative AI format affects your understanding will help inform future explorations of AI’s role in learning.

      What aspects of the conversational framework enhanced or hindered your understanding?

      How did the interplay of concrete examples and reflective discussion affect your learning?

      What difference did it make that you knew before listening that the conversation was created using generative AI?

      We welcome your thoughts on these deeper questions about how learning happens through structured interaction.

    1. Thick knowledge

      Thick knowledge

      Toby Mundy on books as thick knowledge:

      “[…] Books have a unique place in our civilisation […] because they are the only medium for thick descriptions of the world that human beings possess. By ‘thick’ description, I mean an extended, detailed, evidence-based, written interpretation of a subject. If you want to write a feature or blog or wikipedia entry, be it about the origins of the first world war; the authoritarian turn in Russia; or the causes and effects of the 2008 financial crisis, in the end you will have to refer to a book. Or at least refer to other people who have referred to books. Even the best magazine pieces and TV documentaries — and the best of these are very good indeed — are only puddle-deep compared with the thick descriptions laid out in books. They are ‘thin’ descriptions and the creators and authors of them will have referred extensively to books to produce their work.”

      I’ve found myself going back to searching for well-written, comprehensive, in-depth books for sourcing both foundational and most-current knowledge. This notion of ‘thick knowledge’ makes a lot of sense.

      Photo: Rainbow (Katey/flickr).

    2. Know-where

      Know-where

      Six months after starting to develop LSi.io, I have 64 ongoing conversations with 150 interlocutors, connecting humanitarian and development learning leaders, Chief Learning Officers and academic researchers.

      Being independent has given me a unique vantage point from which to examine the humanitarian and development sector’s learning, education and training strategies. I believe that such perspective is indispensable if we are to give more than lip service to “cross-sector” approaches, in an extremely competitive industry faced with shrinking resources (think ECHO budget cuts) and growing needs (think climate change). And I’ve found learning leaders from our world to be a smart, thoughtful and active bunch, finely attuned to the sector’s changing landscape.

      I’ve also enjoyed profound and promising  discussions with CLOs from the corporate sector. One of the most humble I’ve met manages two large brick-and-mortar campuses, one in Asia and the other in Old Europe, running hundreds of courses and dozens of educational programs on twenty-first leadership, fueled by a vision of sustainability in a volatile world that goes beyond trite, wooden and hollow corporate social responsibility.