I challenge you to define them, and will donate $10 to a charity of your choice if your definition gets a karma score of at least 3 points.

Ok, then. Here's my attempt.

Intrinsically interesting topics are topics which satisfy the following criteria:

1) The topic cannot be discussed by an adult human of average intelligence without putting in some cognitive effort and attention. (If you can be busy thinking about another topic while discussing it, then it probably isn't intrinsically interesting). If the topic cannot be discussed by a human of average intelligence then this condition is considered to be met.

2) The topic must have objective aspects which are a primary aspect of the topic.

3) The topic must have some overarching theories to connect the topic or have the possibility of overarching theories explain the topic. Thus for example, celebrity divorces would not fall into this category because they are separate unconnected data points. But differing divorces rates in different income brackets would be ok because one could potentially have interesting sociological explanations for the data.

4) The topic must have bridges to many other topics that aren't simply a variation of the topic itself. For example, AI bridges to programming, psychology, nature of human morality, evolution, neurobiology, and epistemology. In contrast, D&D rules don't connect to other topics in any strong way. There are some minimally interesting probability questions that you can ask if you are writing a quiz for an undergraduate probability course but that's about it. Most of the other topics that it is connected to are still variations of the same topic such as say what a society would look like in a universe that functioned under standard 3.5 D&D rules.