and partly in order to publicify an excuse for why I’ve ceased (and aim to cease more) writing/communicating
I see. I’ve found our communication valuable, and it also makes me a little sad because I only have a few people to infrequently communicate with about alignment. But that would be a selfish (or ‘your-inside-view discrediting’) reason to advise against it.
I do endorse an underlying meta-strategy: I think it’s valuable for some of us—those of us who are so naturally inclined—to try some odd research/thinking-optimizing-strategy that, if it works, could be enough of a benefit to push at least that one researcher above the bar of ‘capable of making serious progress on the core problems’.
One motivating intuition: if an artificial neural network were consistently not solving some specific problem, we’d probably try to improve or change that ANN somehow or otherwise solve it with a ‘different’ one. Humans, by default, have a large measure of similarity to each other. Throwing more intelligent humans at the alignment problem may not work, if one believes it hasn’t worked so far.
In such a situation, we’d instead want to try to ‘diverge’ something like our ‘creative/generative algorithm’, in hopes that at least one (and hopefully more) of us will become something capable of making serious progress.
Social caveat: To me this is logically orthogonal, but I imagine it might be complicated for others to figure out when to be concerned for someone, and when to write it off as them doing this.
(My intuition): More ‘contextualness’ could help, e.g trying to ask some questions to assess someone’s state.
I don’t usually think about ‘what community norms should be’
I see. I’ve found our communication valuable, and it also makes me a little sad because I only have a few people to infrequently communicate with about alignment. But that would be a selfish (or ‘your-inside-view discrediting’) reason to advise against it.
I do endorse an underlying meta-strategy: I think it’s valuable for some of us—those of us who are so naturally inclined—to try some odd research/thinking-optimizing-strategy that, if it works, could be enough of a benefit to push at least that one researcher above the bar of ‘capable of making serious progress on the core problems’.
One motivating intuition: if an artificial neural network were consistently not solving some specific problem, we’d probably try to improve or change that ANN somehow or otherwise solve it with a ‘different’ one. Humans, by default, have a large measure of similarity to each other. Throwing more intelligent humans at the alignment problem may not work, if one believes it hasn’t worked so far.
In such a situation, we’d instead want to try to ‘diverge’ something like our ‘creative/generative algorithm’, in hopes that at least one (and hopefully more) of us will become something capable of making serious progress.
Social caveat: To me this is logically orthogonal, but I imagine it might be complicated for others to figure out when to be concerned for someone, and when to write it off as them doing this.
(My intuition): More ‘contextualness’ could help, e.g trying to ask some questions to assess someone’s state.
I don’t usually think about ‘what community norms should be’