This was the post that got the concept of a “robust agent” to click into place in my mind. It also more concretely got me thinking about what it means to be honest, and how to go about that in a world that sometimes punished honesty.
Since then, I’ve thought a lot about meta-honesty as well as meta-trust (in contexts that are less about truth/falsehood). I have some half-finished posts on the topic I hope to share at some point.
This was the post that got the concept of a “robust agent” to click into place in my mind. It also more concretely got me thinking about what it means to be honest, and how to go about that in a world that sometimes punished honesty.
Since then, I’ve thought a lot about meta-honesty as well as meta-trust (in contexts that are less about truth/falsehood). I have some half-finished posts on the topic I hope to share at some point.
This also had some concrete impacts on how I think about the LessWrong team’s integrity, which made it’s way into several conversations that (I’d guess?) made their way into habryka’s post on Integrity, as well as my Robust Agency for People and Organizations.