# LLMs Cannot Structure Synthesized Explanations See my work on [*Draft 1 - Moravec's Mistake*](*Draft%201%20-%20Moravec's%20Mistake*.md). There will end up being a very clear through line to this argument. I will be able to parse away a ton of unnecessary detail. At no point has chatgpt or notebooklm helped with that. It just helps provide information to me quickly and I—using creativity and an understanding of my problem—criticize and conjecture and move forward with the bits I find useful. I keep asking if they solve my problem. I keep continuing my arguments. --- Date: 20250424 Links to: Tags: References: * []()