What type of Master's is best for AI policy work?

By Milan_Griffes @ 2019-02-22T20:04 (+14)

80,000 Hours recommends a few different flavors of Master's as entry points into working on US-oriented AI policy: security studies, international relations, public policy, and machine learning.

Does anyone have opinions on which of these types of programs is the best to focus on?

(Clearly a large part of this revolves around personal fit, but perhaps some of these are much more relevant than others in a way that dominates personal fit considerations.)

Roxanne_Heston @ 2019-02-27T14:46 (+20)

Milan and I spoke about this, so I'm just commenting to let other readers know that I'm happy to be a resource on this, specifically if you're looking at US programs. For context, I'm a Master's student in Georgetown's Security Studies Program, in their Technology and Security concentration, but have considered and think well of other programs, too.

Khorton @ 2019-02-22T20:36 (+3)

I studied public policy at King's College London and wrote my dissertation on AI policy. I'm happy to answer questions about that particular programme.

Milan_Griffes @ 2019-02-22T22:47 (+2)

Awesome!

My first question: did you end up doing something after the program that you wouldn't have been able to do w/o having done the program first?

Khorton @ 2019-02-22T23:06 (+3)

No. I entered the UK Civil Service, which I could have done anyways. I'm probably one pay grade higher because of experience gained during my Master's / because I had a lot of time to apply for jobs rather than taking the first one I could get.

Milan_Griffes @ 2019-02-23T00:27 (+2)

Oh interesting. Reflecting back, do you feel like you should've just gone into the Civil Service directly, or are you happy with the route you chose?

Khorton @ 2019-02-23T09:58 (+10)

I'm happy with the route I chose. I love education, so the Master's was very pleasant and a real confidence boost. I also learned a lot about research methods, both quantitative and qualitative, which was one of the goals I had - I don't think I could have taught myself those skills. I also think it's useful to know the theoretical literature even if I don't use it every day because it's changed some of the assumptions I hold.

I also like having more options. It turns out that I really like the Civil Service and plan to stay in it for a few years, but it was good to test my fit in research and know that I could work in a think tank or do a PhD if I wanted to.

That said, I didn't previously know that the Civil Service will pay for employees to do a Master's if they have a reasonable business case. It probably would have been more rational for me to do this on the Civil Service's dime than my own. But overall, I think I made the best decision with the information I had available.

Milan_Griffes @ 2019-02-23T18:15 (+3)

Thanks, this is really helpful.