We should distinguish risk aversity, transparency, and bureaucracy. They’re obviously related but different concepts. I would argue that transparency is far more important than risk aversity, the more so the less risk averse you are—and unfortunately nontransparency often seems to be correlated with risk-taking. This is sometimes justified on infohazard logic (cf MIRI in general) or some harder-to-pin-down lack of urgency to communicate controversial decisions (cf Wytham Abbey). Increasing transparency necessarily increases bureaucracy, but there are many other ways bureaucracy can increase, so we shouldn’t expect it to balloon uncontrollably just because of one upward pressure.
I feel like most core EA organisations would come nowhere near meeting the transparency requirements Givewell place on charities they recommend (though Givewell themselves do impressively well on this score, so it’s clearly not impossible for metacharities).
We should distinguish risk aversity, transparency, and bureaucracy. They’re obviously related but different concepts. I would argue that transparency is far more important than risk aversity, the more so the less risk averse you are—and unfortunately nontransparency often seems to be correlated with risk-taking. This is sometimes justified on infohazard logic (cf MIRI in general) or some harder-to-pin-down lack of urgency to communicate controversial decisions (cf Wytham Abbey). Increasing transparency necessarily increases bureaucracy, but there are many other ways bureaucracy can increase, so we shouldn’t expect it to balloon uncontrollably just because of one upward pressure.
I feel like most core EA organisations would come nowhere near meeting the transparency requirements Givewell place on charities they recommend (though Givewell themselves do impressively well on this score, so it’s clearly not impossible for metacharities).