We (AE Studio) are evaluating proposals. We are particularly excited to see ideas that have high potential upside for alignment and would likely be untested if they don’t run through this seedling (for all the reasons stated above)
I hesitate to say anything too concrete here. In line with our meta approach we want lots of different types of ideas without a limit on the approach or scope. We’re particularly interested in ideas that would otherwise go untested (they aren’t supported by your current employer, they aren’t withing the paradigm of funders in the space, etc).
Ideally you should have a vision in your mind of how your idea would impact the state of AI alignment if it worked out in full (even if its high risk and there is a good chance it won’t work out).
We (AE Studio) are evaluating proposals. We are particularly excited to see ideas that have high potential upside for alignment and would likely be untested if they don’t run through this seedling (for all the reasons stated above)
I like AE Studios!
Can you give an at-all-more-concrete operationalization of this?
I hesitate to say anything too concrete here. In line with our meta approach we want lots of different types of ideas without a limit on the approach or scope. We’re particularly interested in ideas that would otherwise go untested (they aren’t supported by your current employer, they aren’t withing the paradigm of funders in the space, etc).
Ideally you should have a vision in your mind of how your idea would impact the state of AI alignment if it worked out in full (even if its high risk and there is a good chance it won’t work out).