I think this is a really nice write-up! As someone relatively new to the idea of AI Safety, having a summary of all the approaches people are working on is really helpful as it would have taken me weeks to put this together on my own.
Thanks!
Obviously this would be a lot of work, but I think it would be really great to post this as a living document on GitHub where you can update and (potentially) expand it over time, perhaps by curating contributions from folks.
I probably won’t do this, but I agree it would be good.
In particular it would be interesting to see three arguments for each approach: a “best argument for”, “best argument against” and “what I think is the most realistic outcome”, along with uncertainties for each.
I agree that this would be good, but especially hard to do in a manner endorsed by all parties. I might try to write a second version of this post that tries to write this out, specifically, trying to clarify the assumptions on what the world has to look like for this research to be useful.
Thanks!
I probably won’t do this, but I agree it would be good.
I agree that this would be good, but especially hard to do in a manner endorsed by all parties. I might try to write a second version of this post that tries to write this out, specifically, trying to clarify the assumptions on what the world has to look like for this research to be useful.
Maybe the “AI Watch” page could incorporate ideas from this post and serve as an equivalent to “a living document on GitHub.”