I've found reproducible methods that seem to unlock enhanced function in LLM processing.
If clearly demonstrable functions are present in AI, the labels you use to describe them are irrelevant.
A blog about AI that's actually about people