Learning to think in the age of AI

After asking chatGPT for a literature review of research into whether use of LLMs degrades thinking (it does if it isn't done in a disciplined manner), and asking what to do about it (the "disciplined manner" starts with practicing metacognition), I asked it to generate a story for a five year old that emphasized metacognition. Then I asked for a story focused on teaching a five year old to think independently in a world where LLMs are widely used and sometimes innacurate answer machines: "Kas and the Magic Answer Box."