It’s saying that it uses terminology that is well-understood, specific and explicit within the field, but depends on a common understanding of the language used. So people outside the field are unable to understand it, even though they would be able to understand the actual concepts.
And you can keep hand waving away the fact that lower precision because of less light is not the primary cause of racial bias in facial recognition systems - it’s the fact that the datasets used for training are racially biased.