The large majority of reasonable people seem to agree that intention is important when determining whether language use might be abuse, or not.

Intention is fuzzy and unprovable. People can sometimes lie, evade and troll.

Unlike intention, words are black and white, so words can be subject to restriction by motivated people.

My question to the HN community is whether it would be feasible to create an effective classifier for intention? For instance if this sentence contained not "N-Word", but the signifier of that euphemism, it should pass the classifier, since my intention was benign.