And then here we are dealing with a classic case of epistemic injustice. Because if they do not say the things that have a way to make to the language models, like if this is already RLHF’d, already aligned and the alignment was opposing this kind of narratives, then these voices are kind of censored or muted. That’s just how alignment works, right? And so part of the community alignment or constitutional alignment is for the community to go back and see how, like I said these words, but I didn’t mean it this way. In your summary, I said these words, but not in that context. There was a forced context by the language model and so on.