MichaelPlant

• MichaelPlant 4 Jun 2019 21:44 UTC

  10 points

  in reply to: Milan_Griffes's comment on: There’s Lots More To Do

  I think Singer would ar­gue we should shame or lock up peo­ple if and only if that did the most good. It’s not at all clear, as a fact of the mat­ter, that would be the best option

• MichaelPlant 29 May 2019 9:49 UTC

  11 points

  in reply to: aarongertler's comment on: Crowd­fund­ing for Effec­tive Cli­mate Policy

  Just pick­ing up on the im­por­tance, ne­glect­ed­ness, tractabil­ity table, Hauke, can I ask you to ex­plain what you meant by those three terms (or, at least the last two) and how you see them as fit­ting to­gether to give you an es­ti­mate of cost-effec­tive­ness? I no­tice you did a fermi es­ti­mate too, so can you say what the re­la­tion­ship is be­tween I, N, T and the fermi es­ti­mate? This isn’t a crit­i­cal ques­tion—I’ve been think­ing about cause prio a lot and I’ve re­al­ised it’s not clear to me how peo­ple use these con­cepts in their de­ci­sion-mak­ing. Hence, if you could say a bit more that would helpful.

  As sug­ges­tions:

  is tractabil­ity: cost-effec­tive­ness, re­sources re­quires to solve the prob­lem, sub­jec­tively-per­ceived eas­i­ness, or some­thing else?

  Is ne­glect­ed­ness: re­sources go­ing to­wards the prob­lem? if so, how di­rectly tar­geted at the prob­lem did you have mind? Is it about coun­ter­fac­tual re­pla­ca­bil­ity? Some­thing else?

  Is the idea I, N, and T some­how give you an in­tu­itive cost-effec­tive­ness es­ti­mate and then you build the fermi es­ti­mate as an ex­plicit fol­low up?

  Sorry if this seems pedan­tic and I’m not en­gag­ing with the spirit of your post. The re­search looks very thor­ough and I’m glad you did it. As a non-ex­pert on the sub­ject mat­ter I prob­a­bly don’t have much of sub­stance of add to that.

• MichaelPlant 7 May 2019 8:19 UTC

  5 points

  on: EA Fo­rum Prize: Win­ners for March 2019

  Thanks for this write up and well done to the win­ners. Am I right in think­ing this is the last in­stance of the EA fo­rum prize, or will there be fur­ther months?

• MichaelPlant 3 May 2019 9:13 UTC

  12 points

  in reply to: aarongertler's comment on: I want an ethnog­ra­phy of EA

  Can you spell out why you’d like to see that? As read I your com­ment I im­me­di­ately thought ‘I would also like to see this’ and then re­al­ised I wasn’t sure why self-re­ports of rea­sons would be use­ful.

• MichaelPlant 1 May 2019 18:39 UTC

  5 points

  in reply to: Milan_Griffes's comment on: Why does EA use QALYs in­stead of ex­pe­rience sam­pling?

  I think that GWWC & GiveWell’s ear­lier use of QALYs cre­ated a lot of path de­pen­dence, such that cur­rent EA pri­ori­ti­za­tion re­mains in­fluenced by the QALY frame­work even though no or­ga­ni­za­tion ex­plic­itly uses it at pre­sent.

  I find this to be the most plau­si­ble ex­pla­na­tion of what has hap­pened. Your coun­ter­fac­tual story is rather helpful!

• MichaelPlant 29 Apr 2019 16:34 UTC

  15 points

  in reply to: Khorton's comment on: Psychedelics Normalization

  Peter Singer is not com­pet­i­tive with Usain Bolt when it comes to run­ning.

  He’s faster than he looks...

  But more se­ri­ously, now I un­der­stand your point, I think it’s plau­si­ble psychedelics could beat AMF (as­sum­ing we count the value of AMF the stan­dard way, look­ing just at the self-re­gard­ing effects of sav­ing lives, i.e. the value to the saved per­son) and more re­search would be use­ful to think through this. I had a go at com­par­ing AMF to drug policy re­form for psychedelics nearly 2 years ago. I think my model is not out of date but it’s at least in­dica­tive. The main prob­lem isn’t the po­ten­tial of psychedelics to be im­pact­ful, as that’s clear—the idea is psychedelics could be much bet­ter treat­ments for men­tal health, which is huge in scale, and chang­ing the law would im­prove treat­ments for huge num­bers of peo­ple—but about what the mostly coun­ter­fac­tual things (for EAs) to do are. It’s not ob­vi­ous what the best lev­er­age points for money/​time are and I haven’t been able to jus­tify the time to look (I’m try­ing to finish a PhD in philos­o­phy and this is not a philos­o­phy topic).

• MichaelPlant 29 Apr 2019 11:34 UTC

  4 points

  in reply to: Khorton's comment on: Psychedelics Normalization

  I’m un­sure what you mean by ‘not com­pet­i­tive with’. Aren’t all causes com­pet­i­tive with each other in the sense that one unit of re­sources (i.e. money or time) you spend on one isn’t one unit you can spend on an­other cause?

• MichaelPlant 29 Apr 2019 11:33 UTC

  29 points

  on: Psychedelics Normalization

  Hello and wel­come! If I can be for­given for toot­ing my own horn, I (with Lee Sharkey) wrote a de­tailed se­ries of fo­rum posts “High Time For Drug Policy Re­form” back in Au­gust 2017, which pri­mar­ily fo­cused on the po­ten­tial of psychedelics as a treat­ment for men­tal health. I also men­tioned it a promis­ing area in an EAGlobal talk in 2018.

  To ad­dress your point, I think the rea­son more EAs don’t pay at­ten­tion to psychedelics is a com­bi­na­tion of EAs not think­ing men­tal health is an im­por­tant prob­lem (some­thing I’ve also writ­ten about) and be­cause psychedelics are weird and un­fa­mil­iar. Re­gard­ing men­tal health’s im­por­tance, I think EAs are in­creas­ingly in­ter­ested in the longterm (this would also ex­plain a rel­a­tive lack of in­ter­est in poverty and an­i­mal welfare) or they are fo­cused on poverty but don’t be­lieve men­tal health treat­ments are com­pa­rably cost-effec­tive with anti-poverty ones. I think men­tal health treat­ments are com­pa­rably cost-effec­tive—at least in the same bal­l­park al­though it’s un­clear which is bet­ter on cur­rent ev­i­dence—when we use self-re­ported hap­piness scores to judge effec­tive­ness. You might then doubt we can sen­si­bly mea­sure hap­piness, which I ar­gue we can in this fo­rum post.

• MichaelPlant 27 Apr 2019 16:08 UTC

  5 points

  in reply to: Jonas Vollmer's comment on: Long-Term Fu­ture Fund: April 2019 grant decisions

  Can you say what these les­sons are? Would be good to have a write up of ad­vice and I would like to see an EA fo­rum post on this.

• MichaelPlant 25 Apr 2019 9:12 UTC

  5 points

  in reply to: Milan_Griffes's comment on: Le­gal psychedelic re­treats launch­ing in Jamaica

  I agree this is plau­si­ble, but I think you would ac­cept that this is con­jec­ture and still quite a long way from what we want, which I as­sume is some sort of quan­tified, ev­i­dence-based, com­par­a­tive anal­y­sis.

• MichaelPlant 25 Apr 2019 9:09 UTC

  4 points

  in reply to: Milan_Griffes's comment on: Why does EA use QALYs in­stead of ex­pe­rience sam­pling?

  I think your short ar­gu­ment misses the point. The ob­sta­cle isn’t the lack of such in­fras­truc­ture—I imag­ine aca­demics could use the ex­ist­ing tools if they asked po­litely or cre­ated their own—but the lack of de­mand for such in­fras­truc­ture.

• MichaelPlant 24 Apr 2019 11:19 UTC

  8 points

  on: Us­ing a mo­bile app to mea­sure well-be­ing for cause pri­ori­ti­za­tion research

  Thanks for writ­ing this up. I agree that ESM is the the­o­ret­i­cally ideal mea­sure of hap­piness. I made a few com­ments on a re­cent post about QALYs vs ESM I thought I should link too here.

  A cou­ple of other com­ments. First, I’d be happy to chat to you about this. Do get in con­tact.

  Se­cond, SWB mea­sures are in­creas­ingly be­ing taken se­ri­ously. See the global hap­piness policy re­port and the fact 170,000 ar­ti­cles and books have been pub­lished on SWB in the last 15 years and the graph be­low for a change over time. How­ever, the fo­cus on mainly on life satis­fac­tion, rather than on ESM mea­sures, and looks set to stay that way. The rea­son for this is a com­bi­na­tion of (a) some SWB re­searchers, e.g. Hel­liwell, think life satis­fac­tion, not hap­piness, is what mat­ters, (b) it’s eas­ier and cheaper to col­lect data on life satis­fac­tion, (c) as a re­sult of (b), there is much more work that has been done to es­tab­lished what will in­crease life satis­fac­tion, which is what is needed to guide policy and do cost-effec­tive­ness—see my hap­piness man­i­festo post and Ori­gins of Hap­piness for more, (d) as a re­sult of the fact more work has been done with life satis­fac­tion, there is now path de­pen­dence where it’s eas­ier to use life satis­fac­tion be­cause other re­searchers are/​have.

  

  Third, have you had an take up from re­searchers on this? If they’d said they aren’t in­ter­ested, did they give rea­sons?

• MichaelPlant 24 Apr 2019 11:06 UTC

  22 points

  on: Why does EA use QALYs in­stead of ex­pe­rience sam­pling?

  Three thoughts. First, it’s not re­ally the case that EAs use QALYs/​DALYs. GWWC and GiveWell used to use them , but GWWC no longer ex­ists as an in­de­pen­dent en­tity and GiveWell now use their own met­ric. 80k mostly fo­cus on the far fu­ture and so QALYs/​DALYs aren’t of pri­mary in­ter­est. Have I missed some­one? I think Founders Pledge do use them. Not sure what goes on ‘un­der the hood’ for The Life You Can Save’s recom­men­da­tions.

  Se­cond, even if you wanted to use the ex­pe­rience sam­pling method (ESM) as your mea­sure of wellbe­ing, you couldn’t be­cause there isn’t enough data on it. There are only two aca­demic pro­jects which have tried to col­lect data en masse—track­y­ourhap­piness and map­piness. The former is now de­funct (Killingsworth works for Microsoft now I be­lieve) and the lat­ter isn’t ac­tively be­ing used (I spoke to the cre­ator, Ge­orge MacKer­ron a cou­ple of months ago) I dis­cuss this in a pre­vi­ous fo­rum post. The best I think we can do, if we want to use sub­jec­tive wellbe­ing (SWB) mea­sure is life satis­fac­tion.

  Third, I think ESM is the the­o­ret­i­cally ideal mea­sure of hap­piness and thus EA—in­deed, ev­ery­one—should use it as the out­come mea­sure of im­pact (I as­sume wellbe­ing con­sists in hap­piness). What fol­lows is that ESM is su­pe­rior to all other mea­sures of wellbe­ing, in­clud­ing QALYs/​DALYs, wealth, etc. I’m hop­ing to do some re­search us­ing ESM at some point in the fu­ture if I can.

• MichaelPlant 20 Apr 2019 11:36 UTC

  5 points

  in reply to: Denise_Melchin's comment on: Thoughts on 80,000 Hours’ re­search that might help with job-search frustrations

  I think it would be mis­lead­ing if OP had said ‘sub­stan­tial pro­por­tion’. I read ‘sub­stan­tial num­ber’ as a com­ment on the ab­solute num­bers, which is vague (how many is ‘sub­stan­tial’) but not mis­lead­ing.

• MichaelPlant 18 Apr 2019 20:49 UTC

  2 points

  in reply to: anonymous_ea's comment on: Thoughts on 80,000 Hours’ re­search that might help with job-search frustrations

  Yup. In which case, it is a ‘big list’ for such folks.

• MichaelPlant 18 Apr 2019 8:29 UTC

  4 points

  in reply to: Ardenlk's comment on: Thoughts on 80,000 Hours’ re­search that might help with job-search frustrations

  Though I am say­ing that 80,000 Hours’ re­search can’t offer a sin­gle, definite rank­ing of what is best for ev­ery­one to do, that doesn’t mean that their re­search isn’t very use­ful for peo­ple figur­ing out what it is best for them to do

  Well, they do offer A list of the most ur­gent global prob­lems. I’ll grant this isn’t a list of what it is best for ev­ery­one to do, but it is (plau­si­bly, from their per­spec­tive) a list of what it is best for most peo­ple to do (or ‘most EAs’ or some nearby speci­fi­ca­tion). In­deed, given 80k has a con­cept of ‘per­sonal fit’, which is dis­tinct from their rat­ing of the prob­lems, the nat­u­ral read­ing of the list is that it pro­vides a gen­eral, im­per­sonal rank­ing of where (av­er­age?) in­di­vi­d­u­als can do the most good.

  I’m con­cerned you’re defend­ing a straw man - did any­one ever claim 80k’s list was true for ev­ery sin­gle pos­si­ble per­son? I don’t think so and such a claim would be im­plau­si­ble.

• MichaelPlant 18 Apr 2019 8:12 UTC

  22 points

  in reply to: Milan_Griffes's comment on: Le­gal psychedelic re­treats launch­ing in Jamaica

  A cou­ple of comments

  Al­most ev­ery­one in EA holds ei­ther a longter­mist view or a per­son-af­fect­ing view

  This puz­zled me slightly. One rea­son is that longter­mism and per­son-af­fect­ing views are differ­ent cat­e­gories; the former is a view about where, in prac­tice, value lies and the lat­ter is a view about where, in the­ory, value lies. You could be a to­tal­ist (all pos­si­ble peo­ple mat­ter), which is not a per­son-af­fect­ing view, but be a near-ter­mism. I think a bet­ter set up would have been: ‘psychedelics look good whether you just value the near-term or the long-term’. I sup­pose that leaves out the ‘medium-ter­mists’, but I don’t know how many peo­ple there are who hold this view, what­ever it is, in­side or out­side EA.

  Also ro­bust: in­ter­ven­tions that in­crease the set of well-in­ten­tioned + ca­pa­ble people

  CFAR & Paradigm Academy are aimed at this

  The psychedelic ex­pe­rience also seems like a plau­si­ble lever on in­creas­ing ca­pa­bil­ity (via re­duc­ing nega­tive self-talk & other men­tal blocks) and im­prov­ing in­ten­tions (via ego dis­solu­tion chang­ing one’s meta­phys­i­cal as­sump­tions)

  I would like you to say more about this. It seems plau­si­ble to me that train­ing ra­tio­nal­ity is or­ders of mag­ni­tude more im­pact­ful for the lon­grun, so this is an ob­jec­tion to counter.

  un­der a longter­mist view, psychedelic in­ter­ven­tions are plau­si­bly in the same bal­l­park of effec­tive­ness as x-risk in­ter­ven­tions

  I don’t think you’ve shown this. It’s more plau­si­ble to me that Xrisk is a top tier in­ter­ven­tion and ra­tio­nal­ity and the ‘mind­set-chang­ing­ness’ of psychedelics are in the lower tiers. This would still make them po­ten­tially very in­ter­est­ing from a long-ter­mist per­spec­tive—in the bucket of ‘things to do take se­ri­ously and pos­si­bly fund if X-risk has ab­sorbed as many re­sources as it can’.

  Just FYI, I wrote a mam­moth se­ries of ar­ti­cles on drug policy re­form 18 months or so ago where I ar­gued that psychedelics for men­tal health looks very promis­ing from the near term per­spec­tive. In other words, I ex­plic­itly claim what you’re claiming! I haven’t had a chance to do more work on it since and I add the usual caveats about not nec­es­sar­ily agree with ev­ery­thing past-Michael wrote.

  Also, just be­cause psychedelics are promis­ing as a cat­e­gory of in­ter­ven­tion, it doesn’t fol­low that set­ting up a re­treat of this kind is the best way to go within that (sub)cause area. You’d need to ar­gue for that too.

• MichaelPlant 18 Apr 2019 7:56 UTC

  13 points

  in reply to: Khorton's comment on: Le­gal psychedelic re­treats launch­ing in Jamaica

  This post did not con­vince me that the busi­ness was cre­ated ‘for EA rea­sons.’

  I think this is un­char­i­ta­ble and I gave small down­vote as a re­sult. Given those in­volved in this busi­ness are in­volved in the EA com­mu­nity and there is at least a plau­si­ble story to tell about why this is im­pact­ful, you’re es­sen­tially claiming ac­cus­ing the OP of act­ing in bad faith when there isn’t com­pel­ling rea­son to do so.

  And con­trary to Fo­rum stan­dards, it was writ­ten to per­suade, not to in­form

  I reread this and didn’t no­tice that it was writ­ten to per­suade vs in­form.

  oth­er­wise why would there be no stud­ies listed that found no effect or a nega­tive effect?

  I’ve been look­ing at the re­search of psychedelics for a while—see https://​​fo­rum.effec­tivealtru­ism.org/​​posts/​​wu9nEXWtvhEnYQTxG/​​high-time-for-drug-policy-re­form-part-1-4-in­tro­duc­tion-and and the other three posts in the se­ries. I can’t re­call a study claiming psychedelics have no or nega­tive effects. I agree that is po­ten­tially sus­pi­cious, but it’s in line with my view that they have pos­i­tive effects and there isn’t much re­search on this.

  But I don’t know any prac­tis­ing doc­tors in the EA com­mu­nity, so this is definitely the wrong place to ad­ver­tise

  Again, I think it’s bad faith to as­sume the pur­pose is sim­ply try­ing to make money from the par­ti­ci­pants of this fo­rum. I think it’s fine, good even, for peo­ple in the com­mu­nity to tell oth­ers what they are do­ing. Where else is one sup­posed to make these an­nounce­ments?

• MichaelPlant 17 Apr 2019 7:34 UTC

  11 points

  in reply to: cole_haus's comment on: Thoughts on 80,000 Hours’ re­search that might help with job-search frustrations

  So it doesn’t seem to be that there’s any in­sol­u­ble ten­sion be­tween tak­ing ac­count of in­di­vi­d­ual differ­ence and com­mu­ni­cat­ing the same mes­sage to a broad audience

  I don’t think the ten­sion is be­tween those things. The ten­sion is be­tween say­ing ‘our re­search is use­ful: it tells (group X) of peo­ple what it is best for them to do’ and ‘our re­search does not offer­ing a definite rank­ing of what it is be­fore for peo­ple to do (whether peo­ple in group X or oth­er­wise)’. I don’t think you can have this both ways.

  While this isn’t en­tirely per­son­al­ized (it’s based only on cer­tain at­tributes that 80,000 Hours high­lights), it’s also far from a sin­gle, defini­tive list

  Then it seems rea­son­able to in­ter­pret it as (an at­tempt at) a defini­tive list if you have those at­tributes.

  I un­der­stand why the au­thor is ar­gu­ing that 80k doesn’t offer a big list but I think that ar­gu­ment is un­der­mines the claim that 80k is use­ful (“Hey, we’re not tel­ling any­one what to do?” “Really? I thought that was the point”)

• MichaelPlant 16 Apr 2019 19:30 UTC

  34 points

  on: Thoughts on 80,000 Hours’ re­search that might help with job-search frustrations

  80,000 Hours’ re­search does not and can­not yield a “big list” of the best ca­reer paths, be­cause no such thing ex­ists. In­stead, we should use 80,000 Hours con­tent to map out our own per­sonal lists and figure out how to do the top things on them.

  Th­ese two sen­tences seem to be in a lot of ten­sion. If giv­ing ad­vice about which ca­reers did the most good were en­tirely per­sonal, then it nec­es­sar­ily fol­lows that you could make no gen­eral recom­men­da­tions at all about which ca­reers are bet­ter in terms of im­pact and there­fore 80k should stop what they are do­ing. How­ever, if you can make gen­eral recom­men­da­tions and thus say which ca­reers have more im­pact that oth­ers, then there is a ‘big list’ af­ter all.

  We might dis­agree about who this is a ‘big list’ for—the av­er­age per­son, an omni-skil­led grad­u­ate of a top uni­ver­sity, the av­er­age reader of 80k’s con­tent—but how­ever we fill that out, it’s still pos­si­ble to see it as a ‘big list’.

  I’m en­tirely with you that it doesn’t make sense to feel bad if some­one else can do more good than you. The aim is to do the most good you can do, not the most good some­one else who isn’t you can do. De­spite recog­nis­ing this on a con­cep­tual level, I still find it hard to be­lieve and of­ten feel guilty (or shame or sad­ness) when I think of peo­ple whose ‘al­tru­is­tic suc­cess­ful­ness’ sur­passes mine.