What gets written about here
Articles cover the questions that come up repeatedly among teachers, students and self-directed learners using AI tools and do not have a tidy answer — where the interesting material usually is.
Recurring subjects include prompt design, question scaffolding, and the persistent problem of telling understanding apart from fluency.
The editorial line
Nothing here promises an outcome, ranks a shortcut, or claims a method works for everyone. Where evidence is thin, the article says so.
Where we have changed our mind, the article says that too, because that a fluent AI answer is a correct one is a mistake worth admitting to having made.
How to read an article here
Each piece is written to be argued with. If you find yourself disagreeing halfway through, that is the intended use — the question How would you check whether this AI response is actually correct? applies to what you read as much as to what you study.
The articles that are most worth your time are the ones that leave you with a better question than you arrived with.
Where the archive leads
Articles are companions to the guides rather than replacements. When a piece raises something you want to work through properly, start with the prompt patterns in resources, then the advanced guide on scaffolded questioning.