> There isn’t a fundamental limitation into building something like a truth classifier into a generative model so that it optimized for outputting “true” statem

> There isn’t a fundamental limitation into building something like a truth classifier into a generative model so that it optimized for outputting “true” statements. Problem is, they didn't do that