In this lunation cycle, we were inspired by the saying: “You are what you eat” and wondered what it means for AI golems that have ingested terabytes of data from often questionable sources.
In this episode of the Pathfinders Podcast, we explore how AI models reflect our biases and why those biases surprise us. We seek the embodied aspect of wisdom, question existing mental models, examine similarities with rubber ducking, and wonder about how we both project and construct personalities of different AI models. We ponder how we might offer our lived experiences to the commons in a more fair value exchange that takes place when we provide data for training AI models.
We also aim to reclaim our collective agency as storytellers that can tip the AI bias in a different direction, and explore the need for greater diversity and woo in our data, and how we might make the AI data diets less WEIRD – Western, Educated, Industrialized, Rich and Democratic – and more weird, or even woo.
If you’d like to wonder and wander with us, join Tethix firekeepers Alja Isaković and Mathew Mytka in this meandering exploration inspired by the latest Pathfinders Newmoonsletter. In addition to the resources linked in the Newmoonsletter, we recommend exploring the following:
You can learn more about the Tethix pathfinding adventure at: https://tethix.co/pathfinders/
What data are we feeding to our Al models, and what are they turning into?