Khorton

• Khorton 16 Feb 2019 8:54 UTC

  8 points

  in reply to: John_Maxwell_IV's comment on: Three Bi­ases That Made Me Believe in AI Risk

  Not nec­es­sar­ily, be­cause peo­ple can be­lieve that mul­ti­ple kinds of work are sig­nifi­cant. I will never be in the mil­i­tary, but I be­lieve there are Gen­er­als out there whose de­ci­sions are life-and-death for a lot of peo­ple. I could pre­sum­ably be­lieve the same about AI safety.

• Khorton 15 Feb 2019 20:58 UTC

  2 points

  in reply to: Khorton's comment on: The Need for and Vi­a­bil­ity of an Effec­tive Altru­ism Academy

  Or, frankly, that it would make EA seem even more like a cult than it already does.

• Khorton 15 Feb 2019 20:55 UTC

  18 points

  in reply to: aarongertler's comment on: The Need for and Vi­a­bil­ity of an Effec­tive Altru­ism Academy

  I was go­ing to raise some similar ob­jec­tions, but you did it bet­ter. :) The only ex­tra point I would raise is that my worst fear with pro­ject isn’t it be­ing a waste of re­sources—my worst fear would be that it leads to sig­nifi­cantly worse group-think as the “best and bright­est” ab­sorb the “right an­swers” from one or two teach­ers.

• Khorton 14 Feb 2019 22:05 UTC

  3 points

  in reply to: David_Moss's comment on: Does giv­ing to char­ity make it more likely you‘ll be al­tru­is­tic in fu­ture?

  Is this the differ­ence be­tween “I’m a per­son who eats healthy” vs “I did such a good job avoid­ing cake at the staff party, so I de­serve a cou­ple cook­ies”?

• Khorton 12 Feb 2019 8:32 UTC

  8 points

  in reply to: aarongertler's comment on: The Nar­row­ing Cir­cle (Gw­ern)

  Elderly peo­ple in the West are treated with less re­gard than they might been a cen­tury ago, I think (al­though I’m not cer­tain—maybe el­derly peo­ple in the West have always been treated poorly?)

• Khorton 12 Feb 2019 8:25 UTC

  1 point

  in reply to: LucyW's comment on: Are men more likely to at­tend EA Lon­don events? At­ten­dance data, 2016-2018.

  I have done some in­ter­views, ac­tu­ally—the Fo­rum post has just got­ten held up by my per­sonal life.

• Khorton 9 Feb 2019 16:58 UTC

  2 points

  on: Reflec­tions on do­ing good with lump sums—the re­tired per­son’s dilemma

  “Given that there is no use­ful anal­y­sis which could be mean­ingful, I need to make an ar­bi­trary, in­stinc­tive de­ci­sion.”

  I agree that analysing which char­i­ties best match your goals is difficult, but I’m sur­prised that you don’t think there’s any use­ful anal­y­sis out there—es­pe­cially since you’re ob­vi­ously in­volved with EA! Can you ex­plain that to me?

• Khorton 8 Feb 2019 14:09 UTC

  3 points

  in reply to: Ramiro's comment on: Open Thread #43

  Risks of an Effec­tive Th­e­sis Prize

  Poor qual­ity:

  -The en­tries (and win­ning dis­ser­ta­tion) are low qual­ity and make EA look bad.

  -The judge(s) don’t have suffi­cient ex­per­tise to eval­u­ate the sub­mis­sions prop­erly.

  -The per­son who or­ga­nizes the com­pe­ti­tion doesn’t have the skills to run it prop­erly (eg the prize is awarded very late or not at all).

  Waste of re­sources com­pared to coun­ter­fac­tual:

  -Thou­sands of dis­ser­ta­tions could be en­tered. Even if you spent 5 min­utes eval­u­at­ing each, that’s 100+ hours of your time. Depend­ing on what your time is worth, if you spent the time work­ing in­stead of eval­u­at­ing dis­ser­ta­tions, you could po­ten­tially save a tod­dler’s life from malaria.

  -It’s un­likely that an­nounc­ing a prize now will im­pact cur­rent PhD stu­dents, who prob­a­bly chose their topic years ago (al­though it could po­ten­tially in­fluence new stu­dents).

  -Are we sure that prizes in­fluence dis­ser­ta­tions? Is there any ev­i­dence base for this?

• Khorton 3 Feb 2019 12:49 UTC

  3 points

  on: What Courses Might Be Most Use­ful for EAs?

  Tak­ing one pro­gram­ming class to test your ap­ti­tude (or spend­ing the equiv­a­lent time learn­ing on­line) might also be worth­while. If your one pro­gram­ming class goes well, you might de­cide that pro­gram­ming would be a good back-up op­tion in case you don’t end up work­ing at MIRI/​GPI/​CEA/​wher­ever else you re­ally want to work.

  I think peo­ple don’t stress enough how im­por­tant it is to be able to pay your stu­dent loan and keep a roof over your head af­ter your grad­u­a­tion. Learn­ing a pro­gram­ming lan­guage and, if that goes well, in­tern­ing or work­ing part time or tak­ing a sum­mer job work­ing in that lan­guage, is an ex­cel­lent way to keep your­self out of poverty.

• Khorton 2 Feb 2019 17:38 UTC

  2 points

  on: What are the eas­iest highly pos­i­tive and effec­tive things that peo­ple can do?

  Easy ways to do more good:

  1. Set up a di­rect debit to one of GiveWell.com’s top char­i­ties. You can always start with $20 a pay­check and change it later; it’s just use­ful to get started and know you’re helping peo­ple.

  2. Work through the 80,000 Hours ca­reer guide. Use their ca­reer plan­ning tool to think about how you could help more peo­ple with your ca­reer.

• Khorton 2 Feb 2019 17:28 UTC

  4 points

  on: What Courses Might Be Most Use­ful for EAs?

  I’m not sure I can an­swer from an EA per­spec­tive, but I’ll try to per­son­ally re­flect on the most use­ful uni­ver­sity courses I’ve taken:

  -I took two post­grad­u­ate re­search meth­ods courses (Quan­ti­ta­tive Re­search Meth­ods and Eval­u­at­ing Public Policy). I am now much bet­ter at eval­u­at­ing the claims made in so­cial sci­ences pa­pers, which I’m re­ally happy about.

  -My Chris­tian the­ol­ogy courses helped de­velop my think­ing about my faith, which I’ve found per­son­ally valuable.

  -So­ciol­ogy /​ Psy­chol­ogy courses have in­fluenced some of my core be­liefs (na­ture vs nur­ture; re­wards vs pun­ish­ment; etc), so in a sense I use this knowl­edge on a daily ba­sis.

  -Drama classes can be great for de­vel­op­ing your com­mu­ni­ca­tion skills, as well as your abil­ity to put your­self in some­one else’s shoes.

  -I trained as a teacher, and my class­room man­age­ment class was sur­pris­ingly trans­fer­able to an office set­ting. It re­ally helped me think about how I can in­fluence other peo­ple’s be­havi­our by choos­ing a meet­ing’s en­vi­ron­ment and agenda and through my own ver­bal and non­ver­bal com­mu­ni­ca­tion.

• Khorton 1 Feb 2019 22:29 UTC

  3 points

  on: Re­search on Effec­tive Strate­gies for Equity and In­clu­sion in Move­ment-Building

  I’m glad you de­cided to cross-post this to the Fo­rum af­ter all. :)

• Khorton 30 Jan 2019 11:12 UTC

  5 points

  on: What Are Effec­tive Alter­na­tives to Party Poli­tics for Effec­tive Public Policy Ad­vo­cacy?

  It’s worth not­ing that state- or province-level pow­ers are usu­ally not “less im­por­tant,” they’re just based on par­tic­u­lar cat­e­gories, so they’re worth pay­ing at­ten­tion to. For ex­am­ple, Canada stopped ex­port­ing as­bestos to de­vel­op­ing coun­tries in 2012 due to a provin­cial de­ci­sion. http://​​www.pre­vent­cancer­now.ca/​​canada-has-stopped-min­ing-and-ex­port­ing-as­bestos-but-the-bat­tle-to-pro­tect-peo­ple-from-as­bestos-harm-con­tinues/​​

• Khorton 30 Jan 2019 8:53 UTC

  8 points

  on: What Are Effec­tive Alter­na­tives to Party Poli­tics for Effec­tive Public Policy Ad­vo­cacy?

  There’s a lot to be said for pub­lic opinion or “poli­ti­cal mood.” I think the le­gal­iza­tion of gay mar­riage in Canada would be a good case study. Fac­tors that led to le­gal­iza­tion in­cluded:

  -lobbying

  -court decisions

  -Pri­vate Mem­ber’s bills

  -a change in pub­lic opinion

  -poli­ti­cal circumstances

  For a large-scale is­sue, work­ing on any of the above would prob­a­bly be worth your time.

• Khorton 27 Jan 2019 18:12 UTC

  5 points

  on: We should make aca­demic knowl­edge easier

  Slight down­vote be­cause I don’t think this is what EAs should fo­cus on. The claim that we could make aca­demic knowl­edge 50% sim­pler isn’t well-ev­i­denced in the ar­ti­cle, and the area seems quite in­tractable for rea­sons ex­plained in the ar­ti­cle.

• Khorton 26 Jan 2019 12:01 UTC

  1 point

  on: Vo­ca­tional Ca­reer Guide for Effec­tive Altruists

  The UK Civil Ser­vice has sev­eral posts which don’t re­quire a de­gree, in­clud­ing their ap­pren­tice­ship scheme. https://​​www.gov.uk/​​gov­ern­ment/​​or­gani­sa­tions/​​civil-ser­vice-fast-track-apprenticeship

• Khorton 23 Jan 2019 8:38 UTC

  2 points

  in reply to: rhys_lindmark's comment on: How can I in­ter­nal­ize my most im­pact­ful nega­tive ex­ter­nal­ities?

  I don’t think it makes sense to donate to an “effec­tive char­ity” in the typ­i­cal sense when it comes to obe­sity be­cause your ex­ter­nal­ity is only af­fect­ing tax­pay­ers in your own health­care sys­tem.

• Khorton 23 Jan 2019 8:37 UTC

  1 point

  in reply to: rhys_lindmark's comment on: How can I in­ter­nal­ize my most im­pact­ful nega­tive ex­ter­nal­ities?

  Given that the table listed the cost of obe­sity as af­fect­ing so­cial­ized medicine, you would pre­sum­ably donate to some­thing lo­cal and med­i­cal, like a hos­pi­tal. You could also po­ten­tially donate to (eg di­a­betes) re­search or a char­ity that makes healthy eat­ing eas­ier in your area.

• Khorton 21 Jan 2019 23:21 UTC

  1 point

  on: [Offer, Paid] Help me es­ti­mate the so­cial im­pact of the startup I work for.

  Please tell me you found some­one—the small num­ber of com­ments makes me sad!

• Khorton 21 Jan 2019 23:19 UTC

  1 point

  on: [Offer, Paid] Help me es­ti­mate the so­cial im­pact of the startup I work for.

  Are you look­ing for peo­ple to eval­u­ate your im­pact based on the data you have, or is your com­pany gen­uinely will­ing to re­search its coun­ter­fac­tual im­pact if you push for it?