Milan_Griffes

• Milan_Griffes 30 Oct 2020 21:30 UTC

  4 points

  on: Des­per­a­tion Ham­ster Wheels

  Thank you for writ­ing this fan­tas­tic post!

  I think un­der­stand­ing & grow­ing more skil­lful in this ter­ri­tory is ex­tremely im­por­tant (from an im­pact-max­i­miz­ing per­spec­tive and from other per­spec­tives too).

  I’m re­minded of some of the prac­tices I listed here, and of Jes­sica’s post On hid­ing the source of knowl­edge as well.

  How do you think the pro­fes­sional and so­cial spaces you were par­ti­ci­pat­ing in dur­ing this time in­ter­acted with the ham­ster-wheel feel­ing?

• Milan_Griffes 30 Oct 2020 19:57 UTC

  2 points

  in reply to: AidanGoth’s comment on: Founders Pledge Re­port: Psychedelic-As­sisted Men­tal Health Treatments

  Fur­ther elab­o­ra­tion of the rescal­ing hy­poth­e­sis and Griffiths et al. 2006 here: https://​​enthea.net/​​founders-pledge-re­port-psychedelics-and-sub­jec­tive-wellbe­ing.html

• Milan_Griffes 30 Oct 2020 19:27 UTC

  10 points

  on: Differ­ences in the In­ten­sity of Valenced Ex­pe­rience across Species

  Thanks for this!

  I’m cu­ri­ous about how these frame­works in­ter­act with Log­a­r­ith­mic Scales of Plea­sure and Pain

• Milan_Griffes 19 Oct 2020 19:57 UTC

  4 points

  in reply to: MichaelDickens’s comment on: The Risk of Con­cen­trat­ing Wealth in a Sin­gle Asset

  Yes, I think the crux here is how good you think your judg­ment/​taste is rel­a­tive to that of all the oth­ers who are try­ing.

• Milan_Griffes 18 Oct 2020 22:39 UTC

  2 points

  in reply to: Matt_Lerner’s comment on: The Risk of Con­cen­trat­ing Wealth in a Sin­gle Asset

  Right, so if you had good judg­ment and a rea­son­able net­work, I think you could find a lot of the promis­ing op­por­tu­ni­ties.

  Some­one would be like “I have a de­vice that pro­duces bright light at a low cost, and I need fund­ing to man­u­fac­ture it on a mas­sive scale.”

  So you’d be like “Really? Let me see this de­vice.”

  You’d in­spect the de­vice, see that it does in fact do the thing the in­ven­tor says it does, and then af­ter fur­ther dili­gence on the busi­ness plan & team you’d in­vest.

  I think this is roughly how to go about sep­a­rat­ing the real op­por­tu­ni­ties from the snake oil.

• Milan_Griffes 18 Oct 2020 18:05 UTC

  2 points

  on: The Risk of Con­cen­trat­ing Wealth in a Sin­gle Asset

  Thanks for this! Su­per in­ter­est­ing.

  A ques­tion I of­ten have about the in­dex-fund ap­proach: if we were in a past his­tor­i­cal mo­ment, wouldn’t VC-style in­vest­ment on moon­shots perform bet­ter than in­dex in­vest­ing?

  e.g. if we were in 1910, wouldn’t it be bet­ter to in­vest in a port­fo­lio of moon­shots that in­cluded Ford and Edi­son, rather than in a broad mar­ket in­dex? (Most of the moon­shots would fail but Ford and/​or Edi­son would bring in mas­sively out­sized re­turns)

• Milan_Griffes 18 Oct 2020 17:54 UTC

  4 points

  on: [Link] “Where are all the suc­cess­ful ra­tio­nal­ists?”

  I’m re­minded of Romeo’s com­ment about ra­tio­nal­ity at­tract­ing “the walk­ing wounded” on a similar post from a cou­ple years back.

  I ac­tu­ally think ra­tio­nal­ity is do­ing pretty good all things con­sid­ered, though I definitely res­onate with Ap­plied Div­inity Stud­ies’ view­point. Tsuyoku Nar­i­tai!

[Link] “Where are all the suc­cess­ful ra­tio­nal­ists?”

• Milan_Griffes 17 Oct 2020 19:36 UTC

  2 points

  on: Five New EA Char­i­ties with High Po­ten­tial for Impact

  This is awe­some!

  Has Char­ity En­trepreneur­ship looked into Wren? https://​​pro­jectwren.com

  (Monthly sub­scrip­tion ser­vice for coun­ter­fac­tu­ally-ad­justed car­bon offsets via re­foresta­tion etc.)

• Milan_Griffes 17 Oct 2020 19:32 UTC

  4 points

  in reply to: kokotajlod’s comment on: Avoid­ing Mu­nich’s Mis­takes: Ad­vice for CEA and Lo­cal Groups

  Agree that the Blues can’t can­cel Trump. Note that be­ing af­fili­ated with Red Tribe isn’t suffi­cient to avoid can­cel­la­tion (though it prob­a­bly helps) – see Pe­traeus, see the Repub­li­cans on these lists: 1, 2

  Jor­dan Peter­son seems ba­si­cally im­pos­si­ble to can­cel due to a com­bi­na­tion of his shame­less­ness & his virtue (he isn’t re­ally Blue Tribe though). Same for Joe Ro­gan and Tyler Cowen.

• Milan_Griffes 16 Oct 2020 15:35 UTC

  2 points

  in reply to: Wei_Dai’s comment on: Avoid­ing Mu­nich’s Mis­takes: Ad­vice for CEA and Lo­cal Groups

  Trump demon­strates that thor­ough­go­ing shame­less­ness effec­tively wards off can­cel­la­tion, at least in the short run.

• Milan_Griffes 14 Oct 2020 20:30 UTC

  2 points

  in reply to: Milan_Griffes’s comment on: Avoid­ing Mu­nich’s Mis­takes: Ad­vice for CEA and Lo­cal Groups

  Also of The Apol­ogy, though that’s ob­vi­ously an ex­treme case.

• Milan_Griffes 14 Oct 2020 20:27 UTC

  12 points

  in reply to: rohinmshah’s comment on: Avoid­ing Mu­nich’s Mis­takes: Ad­vice for CEA and Lo­cal Groups

  I’m re­minded of this.

• Milan_Griffes 14 Oct 2020 20:11 UTC

  20 points

  in reply to: Khorton’s comment on: Avoid­ing Mu­nich’s Mis­takes: Ad­vice for CEA and Lo­cal Groups

  I read this piece as propos­ing a stance to­wards a so­cial dy­namic (“how EA should ori­ent to can­cel cul­ture”), rather than con­tin­u­ing liti­ga­tion of any­one’s char­ac­ter.

• Milan_Griffes 9 Oct 2020 22:47 UTC

  3 points

  in reply to: Thomas Kwa’s comment on: [Linkpost] Some Thoughts on Effec­tive Altruism

  I think some­thing like 3a is right, es­pe­cially given our clue­less­ness.

• Milan_Griffes 9 Oct 2020 17:24 UTC

  4 points

  in reply to: MichaelPlant’s comment on: [Link] How un­der­stand­ing valence could help make fu­ture AIs safer

  … if we could ex­plain to an AGI what hap­piness is, then we could get it to cre­ate more hap­piness (or, at least, not cre­ate more un­hap­piness)?

  I think this cap­tures #1, #2, #4, #6, #8.

  But not #3 and #5, and not re­ally #7, not re­ally #9, not re­ally #10.

• Milan_Griffes 8 Oct 2020 18:55 UTC

  2 points

  in reply to: MichaelA’s comment on: Cru­cial ques­tions for longtermists

  Some re­lated ma­te­rial in this blog post: How un­der­stand­ing valence could help make fu­ture AIs safer

[Link] How un­der­stand­ing valence could help make fu­ture AIs safer

• Milan_Griffes 8 Oct 2020 18:22 UTC

  2 points

  in reply to: AidanGoth’s comment on: Founders Pledge Re­port: Psychedelic-As­sisted Men­tal Health Treatments

  I don’t have a strong view on whether hav­ing chil­dren in­creases or de­creases he­do­nis­tic well-be­ing (though it seems likely to in­crease well-be­ing in de­sire/​prefer­ence terms).

  Yes, I haven’t looked closely but it seems like a com­pli­cated topic.

  Pol­l­mann-Schult 2018 thinks that the hav­ing kids<>life satis­fac­tion re­la­tion­ship de­pends a lot on the con­text:

  There are, how­ever, con­sid­er­able cross-coun­try vari­a­tions in life satis­fac­tion be­tween par­ents and non-par­ents. Within Europe, par­ent­hood is more pos­i­tively as­so­ci­ated with hap­piness in so­cial demo­cratic coun­tries than in con­ser­va­tive, liberal, and Eastern Euro­pean coun­tries (Aassve et al. 2012).

  Notably Swedish, Dan­ish, and Nor­we­gian par­ents are more satis­fied with their lives than their child­less coun­ter­parts (Kohler et al. 2005; Hansen et al. 2009; Daukan­taite and Zukauskiene 2006). A pos­i­tive as­so­ci­a­tion be­tween par­ent­hood and life satis­fac­tion has also been found in Rus­sia (Mikucka 2016).

  Par­ents in the USA and con­ti­nen­tal Euro­pean coun­tries, in con­trast, ex­pe­rience equal or lower lev­els of life satis­fac­tion than child­less in­di­vi­d­u­als (Alesina et al. 2004; Um­ber­son and Gove 1989; Keizer et al. 2010; Myrskylä and Mar­go­lis 2014; Pol­l­mann-Schult 2014; Rizzi and Mikucka 2015).

• Milan_Griffes 8 Oct 2020 18:03 UTC

  2 points

  in reply to: AidanGoth’s comment on: Founders Pledge Re­port: Psychedelic-As­sisted Men­tal Health Treatments

  I won­der which of he­do­nis­tic and prefer­ence util­i­tar­i­anism you’re more sym­pa­thetic to, or which of he­do­nism and prefer­ence/​de­sire the­o­ries of well-be­ing you’re more sym­pa­thetic to. The former tend to go with ex­pe­ri­en­tial SWB and the lat­ter with eval­u­a­tive or eu­daimonic SWB (see Michael Plant’s re­cent pa­per).

  
  

  As far as I can tell, ex­pe­ri­en­tial and eu­daimonic well-be­ing con­verge in the limit, but it’s im­por­tant to pri­ori­tize eu­daimonic well-be­ing along the way to avoid pre­ma­ture op­ti­miza­tion.

  e.g. Jhanic states are more he­do­nic than co­caine or Twit­ter, but also more difficult to ac­cess.