Kaj_Sotala comments on Earning money with/for work in AI safety - Less Wrong

7 Post author: rmoehn 18 July 2016 05:37AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (31)

You are viewing a single comment's thread. Show more comments above.

Comment author: Kaj_Sotala 19 July 2016 12:19:39PM *  5 points [-]

I'm currently working on an AI strategy project for the Foundational Research Institute; they are hiring and do not require plenty of experience:

Requirements

  • Language requirement is research proficiency in English.
  • We anticipate that an applicant is dedicated to alleviating and preventing suffering, and considers it the top global priority.
  • A successful applicant will probably have a background in quantitative topics such as game theory, decision theory, computer science, physics, or math. But we welcome applicants regardless of background.
  • Peer-reviewed publications or a track record of completed comparable research output is not required, but a plus.
  • There is no degree requirement, although a PhD is an advantage, all else equal.

Their open research questions include a number of AI-related ones, and I expect many of them to still have plenty of low-hanging fruit. I'm working on getting a better handle on hard takeoff scnearios in general; most of the my results so far can be found on my website under the "fri-funded" tag. (Haven't posted anything new in a while, because I'm working on a larger article that's been taking some time.)

Comment author: qmotus 21 July 2016 05:14:13PM 0 points [-]

Will your results ultimately take the form of blog posts such as those, or peer-reviewed publications, or something else?

I think FRI's research agenda is interesting and that they may very well work on important questions that hardly anyone else does, but I haven't yet supported them as I'm not certain about their ability to deliver actual results or the impact of their research, and find it a tad bit odd that it's supported by effective altruism organizations, since I don't see any demonstration of effectiveness so far. (No offence though, it looks promising.)

Comment author: Kaj_Sotala 22 July 2016 07:24:13AM 0 points [-]

The final output of this project will be a long article, either on FRI's website or a peer-reviewed publication or both; we haven't decided on that yet.