Considerations on Cryonics

Is cry­on­ics worth it, and if yes, should one cry­ocras­ti­nate (i.e. post­pone sign­ing up for cry­on­ics to a later date)? Bet­teridge’s law of head­lines only ap­plies par­tially here: Yes, it is prob­a­bly worth it (un­der plau­si­ble as­sump­tions $2.5m for a 20 year old, and more for older peo­ple), and no, cry­ocras­ti­na­tion is usu­ally ir­ra­tional. A cost-benefit anal­y­sis writ­ten in Lua. I also perform a Monte-Carlo simu­la­tion us­ing Guessti­mate, and find that sign­ing up for cry­on­ics at age 20 is worth in the mean $35m , me­dian -$100k (90% con­fi­dence in­ter­val: -$1.59m, $63.2m). It there­fore seems recom­mend­able to sign up for cry­on­ics im­me­di­ately.

Con­sid­er­a­tions on Cryonics

If I died, would I be rid of my senses?
Or will it re­tain, trapped within my corpse, in sta­sis?
If I died, would I be a woman in heaven?
Or would I fall asleep, not know­ing what it’s like to feel al­ive?

If I died, would I be­gin with a new life?
Or would I be gone as quickly as the breath I give last?
If I died, would I be a woman in heaven?
Or would I fall asleep, not know­ing what it’s like to feel al­ive?

Pa­tri­cia Taxxon, “De­con­struct” from “Foley Artist”, 2019

One day they woke me up
So I could live for­ever
It’s such a shame the same
Will never hap­pen to you

Jonathan Coul­ton, “Want You Gone” from “Por­tal 2: Songs to Test By (Vol­ume 3)”, 2011

Many would-be cry­on­i­cists cry­ocras­ti­nate, i.e they put off sign­ing up for cry­on­ics un­til a later point in their life. This has of­ten been ex­plained by the fact that sign­ing up for cry­on­ics seems to re­quire high con­scien­tious­ness and can be eas­ily be de­layed un­til an­other point in life: “I’ll get around to do­ing it even­tu­ally” – per­son who was cre­mated. How­ever, it hasn’t yet been ex­plored whether this pro­cras­ti­na­tion might be ra­tio­nal: Many cry­on­ics or­gani­sa­tions have high mem­ber­ship fees, which might be avoided by wait­ing.

To find this out, I pre­sent a point-es­ti­mate model of whether (and if yes, when) to sign up for cry­on­ics. The model is writ­ten in Lua.

Note

This write-up is not in­tended as an in­tro­duc­tion to the con­cept of cry­on­ics. For a pop­u­lar in­tro­duc­tion to the topic that clar­ifies many com­mon mis­con­cep­tions about the prac­tice, see “Why Cry­on­ics Makes Sense” by Tim Ur­ban.

For more ba­sic in­for­ma­tion about the topic, the Cry­on­ics FAQ by Ben Best, a former di­rec­tor of the Cry­on­ics In­sti­tute, an­swers many ques­tions, as well as Al­cor’s Cry­on­ics FAQ.

Th­ese texts should an­swer most ques­tions peo­ple usu­ally have about cry­on­ics.

Cost-Benefit Calcu­la­tion for Cryonics

If you make 50k$/​yr now, and value life-years at twice your in­come, and dis­count fu­ture years at 2% from the mo­ment you are re­vived for a long life, but only dis­count that fu­ture life based on the chance it will hap­pen, times a fac­tor of 12 be­cause you only half iden­tify with this fu­ture crea­ture, then the pre­sent value of a 5% chance of re­vival is $125,000, which is about the most ex­pen­sive cry­on­ics price now.

Robin Han­son, “Break Cry­on­ics Down”, 2009

To find out whether to sign up for cry­on­ics at all, one needs to make a cost-benefit calcu­la­tion. This has been at­tempted be­fore, but that anal­y­sis has been rather short (dis­re­gard­ing sev­eral im­por­tant fac­tors) and it might be pro­duc­tive to ap­proach the topic in­de­pen­dently.

The costs of cry­on­ics are com­par­a­tively easy to calcu­late and con­tain lit­tle un­cer­tainty: The price of cry­op­reser­va­tion and life-in­surance are widely known, and can be eas­ily added to­gether. The benefits of cry­op­reser­va­tion, how­ever, con­tain a lot more un­cer­tainty: It is not at all clear that the tech­nol­ogy for reusci­ta­tion will be de­vel­oped, cry­on­ics or­ga­ni­za­tions (or hu­man­ity) sur­vive to de­velop such tech­nol­ogy, or that the fu­ture will be in­ter­ested in reusci­tat­ing peo­ple from cry­op­reser­va­tion.

The model pre­sented makes the as­sump­tion that a per­son has a given age and has the op­tion of wait­ing for sign­ing up for cry­on­ics ev­ery year up to their ex­pected year of death. So, for ex­am­ple, a per­son that is 20 years old now is able to plan sign­ing up when they are 20 years old, 21 years, 22 years and so on up to 78 years. The value of cry­on­ics is calcu­lated, and the value of a reg­u­lar death is tac­itly as­sumed to be $0.

curage=20
act­val={78.36, 78.64, 78.66, 78.67, 78.68, 78.69, 78.69, 78.70, 78.71, 78.71, 78.72, 78.72, 78.73, 78.73, 78.74, 78.75, 78.75, 78.77, 78.79, 78.81, 78.83, 78.86, 78.88, 78.91, 78.93, 78.96, 78.98, 79.01, 79.03, 79.06, 79.09, 79.12, 79.15, 79.18, 79.21, 79.25, 79.29, 79.32, 79.37, 79.41, 79.45, 79.50, 79.55, 79.61, 79.66, 79.73, 79.80, 79.87, 79.95, 80.03, 80.13, 80.23, 80.34, 80.46, 80.59, 80.73, 80.88, 81.05, 81.22, 81.42, 81.62, 81.83, 82.05, 82.29, 82.54, 82.80, 83.07, 83.35, 83.64, 83.94, 84.25, 84.57, 84.89, 85.23, 85.58, 85.93, 86.30, 86.68, 87.08, 87.49, 87.92, 88.38, 88.86, 89.38, 89.91, 90.47, 91.07, 91.69, 92.34, 93.01, 93.70, 94.42, 95.16, 95.94, 96.72, 97.55, 98.40, 99.27, 100.14, 101.02, 101.91}

for age=curage,math.floor(act­val[curage]) do
	print(value(age) .. “: ” .. age)
end

curage con­tains the cur­rent age of the user of the pro­gram. act­val is an ac­tu­ar­ial table that con­tains at the nth po­si­tion the me­dian life ex­pec­tancy of a per­son that is n years old at the mo­ment for a west­ern na­tion (in this case Ger­many).

This model usu­ally tries to err on the side of con­ser­va­tive es­ti­mates, think of the lower range of a 50% con­fi­dence in­ter­val.

The Dis­value of Waiting

Two im­por­tant fac­tors play into the value (or dis­value) of wait­ing to sign up for cry­on­ics: Mo­ti­va­tion drift and the pos­si­bil­ity of dy­ing be­fore sign­ing up.

func­tion value(age)
	re­turn prob_signup(age)*prob_liveto(age)*(benefit(age)-cost(age))
end

Mo­ti­va­tion Drift

prob_signup is a func­tion that calcu­lates the prob­a­bil­ity of sign­ing up for cry­on­ics af­ter hav­ing waited up to hav­ing a cer­tain age. It seems clear that peo­ple loose mo­ti­va­tion to finish plans over time, es­pe­cially if they are un­pleas­ant or com­plex. A good ex­am­ple for this is peo­ple be­ing mo­ti­vated at the start of the year to do reg­u­lar ex­er­cise: How many of those ac­tu­ally keep their promises to them­selves? They might start off ex­er­cis­ing, but af­ter the first few weeks the first peo­ple drop out, and and a cou­ple of months there is nearly no­body left still go­ing to the gym ex­cept the ones who already did it be­fore. It seems like there is a strong re­gres­sion to the mean in re­gards to ac­tion: Most reg­u­lar ac­tions are re­placed by in­ac­tion, most strong val­ues are re­placed by ap­a­thy over time. A similar phe­nomenon seems likely for sign­ing up for cry­on­ics: At first, peo­ple are very en­thu­si­as­tic about sign­ing up, but then loose in­ter­est as time pro­gresses.

It doesn’t seem ob­vi­ous how strong mo­ti­va­tion drift is and how it de­vel­ops over time (some peo­ple might re­gain mo­ti­va­tion af­ter some time), but in­tu­itively it seems like a ge­o­met­ric dis­tri­bu­tion. The rea­son­ing is as fol­lows: Imag­ine that a thou­sand peo­ple have the mo­ti­va­tion to perform a given ac­tion n years into the fu­ture. Every year, a cer­tain per­centage p of the peo­ple still mo­ti­vated loses in­ter­est in perform­ing that ac­tion and drop out. After n years, the num­ber of peo­ple who perform the ac­tion is (the per­centage of peo­ple still mo­ti­vated is ).

When try­ing to find out what the value of p is for one­self, one can imag­ine a thou­sand in­de­pen­dent iden­ti­cal copies of one­self plan­ning a com­plex plan one year ahead. How many of those would ac­tu­ally fol­low through on that plan? In­tu­itively, I’d say that it can’t be much higher than 95%, pos­si­bly much lower, es­pe­cially for some­thing as com­plex and time-con­sum­ing as sign­ing up for cry­on­ics.

de­cay=0.95
func­tion prob_signup(age)
	re­turn de­cay^(age-curage)
end

In­ter­est­ingly, this does not mean that the de­ci­sion of whether to be cry­on­i­cally pre­served or not is then set in stone as soon as pos­si­ble: Cry­on­ics mem­ber­ships are very easy to can­cel, in nearly all cases a sim­ple email and a ces­sa­tion of pay­ing mem­ber­ship fees suffices. Sign­ing up for cry­on­ics ear­lier pro­tects against re­gres­sion to the mean, which means ap­a­thy or lack of mo­ti­va­tion to­wards cry­on­ics, but does not pro­tect against chang­ing ones mind about cry­on­ics: If one be­comes con­vinced it’s bul­lshit later, one can eas­ily get out (much more eas­ily than get­ting in). On the other hand, there might be a feel­ing of con­sid­er­able sunk cost due to already paid mem­ber­ship fees and the ac­quired life in­surance.

It will be as­sumed that once one is signed up for cry­on­ics, one stays signed up for it.

Dy­ing Be­fore Sign­ing Up

If you die be­fore sign­ing up, all pos­si­ble value (or dis­value) of cry­on­ics gets lost. So we want to calcu­late the prob­a­bil­ity of dy­ing be­fore hav­ing a cer­tain age given be­ing cur­rently curage years old.

Mor­tal­ity rates are of­ten calcu­lated us­ing a so-called Gom­pertz dis­tri­bu­tion. I de­ter­mined the b and eta val­ues by eye­bal­ling Wolfram Alpha and us­ing a calcu­la­tor in To­masik 2016

b=0.108
eta=0.0001

func­tion gom­pertz(age)
	re­turn math.exp(-eta*(math.exp(b*age)-1))
end

gom­pertz re­turns the prob­a­bil­ity of reach­ing age start­ing from birth, but I need the prob­a­bil­ity of reach­ing age given one is already curage years old. With Bayes the­o­rem one can calcu­late that

is equal to be­cause be­ing older than age is (in this calcu­la­tion) a sub­set of be­ing older curage, and . Some pre­cau­tions have to ap­ply in the case that the prob­a­bil­ities of reach­ing age is not in­de­pen­dent of the prob­a­bil­ity of reach­ing curage, but those are difficult to es­ti­mate and will not be im­ple­mented here.

This way, one can im­ple­ment the prob­a­bil­ity of liv­ing un­til age given curage the fol­low­ing way:

func­tion prob_liveto(age)
	re­turn gom­pertz(age)/​gom­pertz(curage)
end

Longevity Es­cape Velocity

Longevity Es­cape Ve­loc­ity (short LEV) is the name for the pos­si­ble year when anti-ag­ing tech­nol­ogy be­comes so good that peo­ple can be re­ju­ve­nated faster than they age. Although the con­cept is con­sid­ered idle spec­u­la­tion in many cir­cles, many fu­tur­ists jus­tify not sign­ing up for cry­on­ics be­cause they ex­pect that LEV will ar­rive dur­ing their life­time, and see no rea­son to sign up for a cry­on­ics mem­ber­ship they are prob­a­bly not go­ing to need any­way. In this text, I will con­sider LEV by as­sum­ing there will be a cer­tain year af­ter which the prob­a­bil­ity of death is prac­ti­cally zero.

I some­what ar­bi­trar­ily set this year to 2080, though many fu­tur­ists seem more op­ti­mistic:

levyear=2080

Calcu­lat­ing the Cost

Calcu­lat­ing the cost is com­par­a­tively straight­for­ward, but there are some hid­den vari­ables (like op­por­tu­nity costs and so­cial costs) that have to be con­sid­ered (not all of these are con­sid­ered in this text).

The raw cost for cry­on­ics de­pends heav­ily on the or­gani­sa­tion choosen for preser­va­tion, the ba­sic price range is from ~$20000 to ~$250000. In this case, I chose the costs for neu­ro­cry­op­reser­va­tion at Al­cor, though this anal­y­sis should be ex­tended to other or­gani­sa­tions.

Raw cry­on­ics cost can be split into three differ­ent parts: mem­ber­ship fees, com­pre­hen­sive mem­ber standby costs and the cost for cry­op­reser­va­tion.

func­tion cost(age)
	re­turn mem­ber­ship_fees(age)+pres_cost(age)+cms_fees(age)
end

Mem­ber­ship Fees

Mem­ber­ship fees for Al­cor are calcu­lated us­ing the age of the mem­ber and the length of their mem­ber­ship.

Direct Fees

Cur­rent Mem­ber­ship Dues, net of ap­pli­ca­ble dis­counts, are:

  1. First fam­ily mem­ber: $525 an­nu­ally or $267 semi-an­nu­ally or $134 quar­terly.

  2. Each ad­di­tional fam­ily mem­ber aged 18 and over, and full-time stu­dents aged 25 and un­der: $310 an­nu­ally or $156 semi-an­nu­ally or $78 quar­terly.

  3. Each minor fam­ily mem­ber un­der age 18 for the first two chil­dren (no mem­ber­ship dues are re­quired for any ad­di­tional minor chil­dren): $80 an­nu­ally or $40 semi-an­nu­ally or $20 quarterly

  4. Full-time stu­dent aged 26 to 30: $460 an­nu­ally or $230 semi-an­nu­ally or $115 quar­terly.

  5. Long-term mem­ber (to­tal mem­ber­ship of 20 − 24 years): $430 an­nu­ally or $216 semi-an­nu­ally or $108 quar­terly.

  6. Long-term mem­ber (to­tal mem­ber­ship of 25 − 29 years): $368 an­nu­ally or $186 semi-an­nu­ally or $93 quar­terly.

  7. Long-term mem­ber (to­tal mem­ber­ship of 30 years or longer): $305 an­nu­ally or $154 semi-an­nu­ally or $77 quar­terly.

  8. Long-term mem­ber (to­tal mem­ber­ship of 40 years or longer): $60.00 an­nu­ally or $30.00 semi-an­nu­ally or $15.00 quarterly

Al­cor Life Ex­ten­sion Foun­da­tion, “Al­cor Cry­op­reser­va­tion Agree­ment—Sched­ule A”, 2016

The fol­low­ing as­sump­tions will be made in the im­ple­men­ta­tion:

  1. The per­son con­sid­er­ing sign­ing up for cry­on­ics is over 18 years old.

  2. If the per­son is un­der 25 years old, they are a stu­dent. Con­sid­er­ing the fact that cry­on­ics mem­bers seem to be more likely to be rich and ed­u­cated, this seems likely, though maybe a bit clas­sist. The code can be changed if per­sonal need arises.

  3. If the per­son is over 25 years old, they are not a stu­dent.

  4. The per­son stays a mem­ber un­til their death (oth­er­wise the cry­on­ics ar­range­ment doesn’t work).

  5. The mem­ber­ship fees will not be changed dras­ti­cally over time. In fact, in­fla­tion ad­justed prices for cry­on­ics have mostly stayed con­stant, so this is a rea­son­able as­sump­tion.

  6. The cry­on­i­cist will know when LEV has oc­curred, and will can­cel their mem­ber­ship start­ing from that year.

The im­ple­men­ta­tion is quite straight­for­ward:

func­tion al­cor_fees(age)
	lo­cal left=math.min(math.floor(act­val[age])-age, levyear-curyear)
	lo­cal cost=0

	if age<25 then
		newage=25
		cost=(newage-age)*310
	end
	if left>=30 then
		cost=cost+(left-30)*305
		left=30
	end
	if left>=25 then
		cost=cost+(left-25)*368
		left=24
	end
	if left>=20 then
		cost=cost+(left-20)*430
		left=20
	end
	if age<=25 then
		cost=cost+(left-(25-age))*525
	else
		cost=cost+left*525
	end

	re­turn 300+cost
end

Com­pre­hen­sive Mem­ber Standby

For Mem­bers re­sid­ing in the con­ti­nen­tal U.S. and Canada: Al­cor will provide Com­pre­hen­sive Mem­ber Standby (CMS) to all Mem­bers (standby in Canada may be sub­ject to de­lays due to cus­toms and im­mi­gra­tion re­quire­ments), which in­cludes all res­cue ac­tivi­ties up through the time the legally pro­nounced Mem­ber is de­liv­ered to the Al­cor op­er­at­ing room for cry­opro­tec­tion. This charge is waived for full-time stu­dents un­der age 25 and minors (un­der age 18).

Al­cor Life Ex­ten­sion Foun­da­tion, “Al­cor Cry­op­reser­va­tion Agree­ment—Sched­ule A”, 2016

Em­pha­sis mine.

Cur­rent CMS charges are:
$180 an­nu­ally, $90 semi-an­nu­ally, or $45 quarterly

Al­cor Life Ex­ten­sion Foun­da­tion, “Al­cor Cry­op­reser­va­tion Agree­ment—Sched­ule A”, 2016

I will as­sume that the cry­on­ics mem­ber starts pay­ing a CMS fee start­ing 10 years be­fore their ac­tu­ar­ial age of death.

cms=180

func­tion cms_age(age)
	re­turn act­val[age]-10
end

func­tion cms_fees(age)
	re­turn cms*(act­val[age]-cms_age(age))
end

Preser­va­tion Cost

There are sev­eral differ­ent meth­ods of fund­ing cry­on­ics, the most pop­u­lar of which seems to be life in­surance. I haven’t spent much time in­ves­ti­gat­ing the ex­act in­ner work­ings of life in­surances, so I will make the as­sump­tion that the in­surance com­pa­nies price their prod­ucts ad­e­quately, so one doesn’t have much of a fi­nan­cial ad­van­tage by choos­ing life in­surance as op­posed to sim­ply sav­ing money & pay­ing the cry­on­ics mem­ber­ship in cash. I also as­sume that life in­surance com­pa­nies can ac­cu­rately price in the ar­rival date of LEV.

Min­i­mum Cry­op­reser­va­tion Fund­ing:
• $200,000.00 Whole Body Cry­op­reser­va­tion […].
• $80,000.00 Neu­ro­cry­op­reser­va­tion […].
[…]
Sur­charges:
• $10,000 Sur­charge for cases out­side the U.S. and Canada other than China.
• $50,000 Sur­charge for cases in China.
[…]

Al­cor Life Ex­ten­sion Foun­da­tion, “Al­cor Cry­op­reser­va­tion Agree­ment—Sched­ule A”, 2016

I as­sume that the per­son con­sid­er­ing sign­ing up lives out­side of the U.S (but not in China), since a lot more peo­ple live out­side the U.S than in­side of it. I also as­sume that the per­son wants to sign up for neu­ro­cry­op­reser­va­tion. With these as­sump­tions, the func­tion that re­turns preser­va­tion costs be­comes quite sim­ple:

func­tion pres_cost(age)
	re­turn 90000
end

Other Pos­si­ble Costs

There is a num­ber of differ­ent ad­di­tional costs that have not been con­sid­ered here be­cause of their (per­ceived) small scale or difficult tractabil­ity.

Th­ese in­clude op­por­tu­nity costs for the time spent in­form­ing one­self about cry­on­ics (tens of hours spent), op­por­tu­nity costs for the time spent sign­ing up (tens of hours spent), so­cial costs by seem­ing weird (though cry­on­ics is easy to hide, and most cry­on­i­cists seem to be rather vo­cal about it any­ways), and alienat­ing fam­ily mem­bers who nec­es­sar­ily come into con­tact with cry­on­ics (con­sid­er­ing the “Hos­tile Wife Phenomenon”).

Calcu­lat­ing the Benefit

Calcu­lat­ing the benefit of cry­on­ics car­ries a great un­cer­tainty, but ba­si­cally it can be di­vided into six dis­tinct com­po­nents: The prob­a­bil­ity of be­ing pre­served, the prob­a­bil­ity of re­vival, the amount of years gained by cry­on­ics, the value of one lifeyear, the prob­a­bil­ity of liv­ing to the year when one will sign up, and the prob­a­bil­ity of then dy­ing be­fore LEV.

func­tion benefit(age)
	re­turn prob_pres*prob_succ*years_gain*val_year*prob_liveto(age)*prob_diebe­forelev(age)
end

Here, I will only take point es­ti­mates of these val­ues.

Value of a Lifeyear in the Future

Much ink and pix­els have been spilled on the ques­tion of the qual­ity of the fu­ture, very lit­tle of it try­ing to make ac­cu­rate or even re­solv­able pre­dic­tions. One way to look at the ques­tion could be to cre­ate clear crite­ria that en­cap­su­late the most im­por­tant hu­man val­ues and ask a pre­dic­tion mar­ket to start bet­ting. This could in­clude the power of hu­man­ity to make most im­por­tant de­ci­sions re­gard­ing its de­vel­op­ment and re­source man­age­ment, di­ver­sity among hu­man be­ings, av­er­age hap­piness and lifes­pans and other vari­ables such as in­equal­ity re­gard­ing re­sources.

But a much sim­pler way of ap­proach­ing the topic could be the fol­low­ing: One takes ar­gu­ments from both sides (pro­claiming pos­i­tive fu­tures and nega­tive fu­tures) and pre­ma­turely con­cludes that the fu­ture is on av­er­age go­ing to be neu­tral, with a high var­i­ance in the re­sult. But some prob­lems pre­sent them­selves: In differ­ent value sys­tems, “neu­tral” means very differ­ent things. Strictly speak­ing, a util­i­tar­ian would see hu­man ex­tinc­tion as neu­tral, but not net neu­tral (the util­ity of a world with­out any sen­tient be­ings is ex­actly 0, which is pre­sum­ably lower than the cur­rent value of the world), anti-na­tal­ists con­sider an empty world to be a pos­i­tive thing, and most peo­ple work­ing on pre­vent­ing hu­man ex­tinc­tion would con­sider such a world to be a gi­gan­tic loss of op­por­tu­nity, and there­fore net nega­tive.

There seems to be no sim­ple way to re­solve these con­flicts, oth­er­wise it would have been writ­ten down up to now. But it seems like most peo­ple would take the cur­rent state of af­fairs as neu­tral, with im­prove­ments in hap­piness, mean­ing and wealth to be pos­i­tive, and de­creases in those to be nega­tive. Also, they don’t see dy­ing to­mor­row as a neu­tral event.

Caveats

Here I will as­sume that

  • Fu­ture life years can be av­er­aged in their qual­ity, and that av­er­age has mon­e­tary value

  • Fu­ture lifeyears are not tem­po­rally dis­counted, i.e a lifeyear 1000 years in the fu­ture is as valuable as the next lifeyear

  • There are no diminish­ing marginal re­turns to lifeyears, i.e 1000 life years are 1000 times as valuable as one lifeyear

Th­ese are pre­sum­ably con­tro­ver­sial as­sump­tions, but they sim­plify the anal­y­sis. I will con­tinue to read about philoso­phers’ and economists’ analy­ses of the re­la­tion be­tween ad­di­tional lifeyears and util­ity, and up­date this sec­tion.

QALYs and VSL

There are two differ­ent meth­ods of putting a value on hu­man life: the VSL and the QALY. The Wikipe­dia page on VSL lists $182000 for the value of a year of life in Aus­tralia, and $50000 as the “de facto in­ter­na­tional stan­dard most pri­vate and gov­ern­ment-run health in­surance plans wor­ld­wide use to de­ter­mine whether to cover a new med­i­cal pro­ce­dure”. This num­ber seems like a good con­ser­va­tive es­ti­mate.

In­ter­est­ingly, this ap­prox­i­mately equals a year of wak­ing hours worth the min­i­mum wage ($$10167*52=$58240$).

In­tu­itively, the prob­a­bil­ity dis­tri­bu­tion over the value of a year of life in the fu­ture should then look like this:

.l(“nplot”)
.l(“nstat”)

grid([-20000 120000 20000];[0 0.00004 0.000004])
xti­tle(“Dol­lar value of a fu­ture life year”)
yti­tle(“Prob­a­bil­ity”)
plot({n.pdf(x;50000;500000000)})
draw()

Probability distribution over the value of a lifeyear in the future

Note that this graph is not based on real data and only for illus­tra­tive pur­poses.

But one can take an­other fac­tor into ac­count: Most nega­tive fu­ture sce­nar­ios don’t lead to reusci­ta­tion (civil­i­sa­tional col­lapse, sta­ble to­tal­i­tar­i­anism, ex­is­ten­tial catas­tro­phes like AI failure, nu­clear war, biotech­nolog­i­cal dis­aster and nat­u­ral catas­tro­phe all re­duce hu­man ca­pa­bil­ities or keep them con­stant, pre­vent­ing the de­vel­op­ment of reusci­ta­tion tech­nol­ogy). In most of the nega­tive fu­tures, there are ei­ther no more hu­mans around or peo­ple don’t have time, en­ergy or re­sources to bring peo­ple back from cry­onic preser­va­tion (if in­deed they still are in preser­va­tion by that point), and for mal­i­cious ac­tors, in most sce­nar­ios it is eas­ier to cre­ate new peo­ple than to bring pre­served peo­ple back.

This effect might be damp­ened by the con­sid­er­a­tion that most pos­si­ble fu­tures have net-nega­tive value, but on the other hand, nearly all of those fu­tures don’t lead to reusci­ta­tion.

This would mean that the prob­a­bil­ity dis­tri­bu­tion over the value of a lifeyear in the fu­ture con­di­tional on be­ing reusci­tated could look like this:

.l(“nplot”)
.l(“nstat”)

grid([-20000 120000 20000];[0 0.00004 0.000004])

xti­tle(“Dol­lar value of a fu­ture life year”)
yti­tle(“Prob­a­bil­ity”)
plot({:[x>50000;n.pdf(x;50000;500000000);0.4472*n.pdf(x;50000;100000000)]})
draw()

Probability distribution over the value of a lifeyear in the future conditional on being reuscitated

Note that this graph is not based on real data and only for illus­tra­tive pur­poses.

Nega­tive Scenarios

How­ever, I can think of 3 very spe­cific (and thereby highly un­likely) sce­nar­ios where peo­ple could be reusci­tated into a (for them) net-nega­tive world.

As­cended Economy

The as­cended econ­omy is a sce­nario where the de­vel­op­ment of cap­i­tal­ism di­verges sig­nifi­cantly from the de­sires of hu­mans, lead­ing to most (if not all) of hu­man­ity be­com­ing ex­tinct. It seems highly un­likely, but pos­si­ble that cry­op­re­served hu­mans are placed into the hands of an al­gorithm that in­vests the money in the rele­vant funds to reusci­tate the cry­op­re­served hu­mans at a cer­tain point. This al­gorithm could re­ceive lit­tle (or no) in­for­ma­tion on what to do with the reusci­tated hu­mans af­ter­wards, lead­ing ei­ther to these hu­mans quickly dy­ing again be­cause of an econ­omy where they are worth­less, or be­ing kept al­ive solely for fulfilling the con­tract that is em­bed­ded in the al­gorithm. This might lead to in­san­ity-in­duc­ing bore­dom as the hu­mans are kept al­ive as long as al­gorithm man­ages to, pos­si­bly hun­dreds or thou­sands of years. This would have net-nega­tive value for the peo­ple reusci­tated.

Malev­olent Fu­ture Actors

A su­per­in­tel­li­gence be­comes a sin­gle­ton and starts be­hav­ing malev­olently be­cause of a near miss in its im­ple­men­ta­tion or or be­cause it has been set up by a malev­olent hu­man. This would lead to cry­op­re­served peo­ple be­ing reusci­tated, hav­ing their brains scanned and ex­e­cuted as a brain em­u­la­tion, copied and put into very painful con­di­tions.

In­for­ma­tion from the Past is Valuable

In a fu­ture where agents that don’t care about hu­mans find the cry­op­re­served re­mains of hu­mans, they might be in­ter­ested in ex­tract­ing in­for­ma­tion from those brains. If it is not pos­si­ble to ex­tract this in­for­ma­tion with­out re­viv­ing the cry­op­re­served peo­ple, they might reusci­tate them and then in­ter­ro­gate these re­vived peo­ple for a very long time, with lit­tle re­gard for their well-be­ing.

Steps for Re­duc­ing the Risk from such Scenarios

b) When, in Al­cor’s best good faith judge­ment, it is de­ter­mined that at­tempt­ing re­vival is in the best in­ter­ests of the Mem­ber in cry­op­reser­va­tion, Al­cor shall at­tempt to re­vive and re­ha­bil­i­tate the Mem­ber. It is un­der­stood by the Mem­ber that a care­ful as­sess­ment of the risks ver­sus the benefits of a re­vival at­tempt will be ma­te­rial to de­ter­min­ing when to at­tempt re­vival. […]
d) Where it is pos­si­ble to do so, Al­cor rep­re­sents that it will be guided in re­vival of the cry­op­re­served Mem­ber by the Mem­ber’s own wishes and de­sires as they may have been ex­pressed in a writ­ten, au­dio, or video State­ment of Re­vival Prefer­ences and De­sires, which the Mem­ber may at his/​her dis­cre­tion at­tach to this Agree­ment.

Al­cor Life Ex­ten­sion Foun­da­tion, “Cry­op­reser­va­tion Agree­ment” p. 1516, 2012

Although not a failsafe mea­sure, steps can be taken to re­duce the risks from hellish sce­nar­ios above by mak­ing ar­range­ments with cry­on­ics or­gani­sa­tions. This may in­clude not want­ing cry­op­reser­va­tion to con­tinue in an as­cended econ­omy, ob­ject­ing to re­vival as an em­u­la­tion or re­vival af­ter more than a cer­tain num­ber of years (to pre­vent be­ing reusci­tated in an in­com­pre­hen­si­bly strange and alien world).

Other Thoughts

Many peo­ple ar­gue that the value of a year of life in the fu­ture might be much lower than in the pre­sent, be­cause friends and familiy are not around, and it is very likely that the fu­ture will be ex­tremely alien and un­fa­mil­iar.

Th­ese are valid con­sid­er­a­tions, but can be damp­ened a bit: Hu­mans have shown to adapt to very differ­ent and varied cir­cum­stances, and hu­mans to­day feel that mod­ern life in big cities with reg­u­lar cal­en­dars and highly struc­tured lives with­out any wor­ries about sur­vival is nor­mal, while for most hu­mans who ever lived, it would be any­thing but. One can spec­u­late that very similar facts will also hold for the fu­ture (be­com­ing in­creas­ingly un­likely the fur­ther reusci­ta­tion lies in the fu­ture). There would cer­tainly be a big cul­ture shock in the fu­ture, but it seems not qual­i­ta­tively differ­ent from the shock peo­ple have when they visit differ­ent coun­tries to­day. It is pos­si­ble that fu­ture so­cieties might try to help peo­ple with this kind of fu­ture shock, but that is of course far from cer­tain.

It is true that most cry­on­i­cists will not be able to con­vince their friends and fam­ily to sign up for it too, so they will be alone in the fu­ture at first. Peo­ple to­day some­times leave their friends and even fam­i­lies to move to other places, but those peo­ple seem to be the ex­cep­tion rather than the norm. How­ever, peo­ple nearly always move on with their life, even as they get di­vorced, get es­tranged from their friends or see their chil­dren less reg­u­larly – they don’t seem to pre­fer death to con­tin­u­ing their lives with­out spe­cific peo­ple. This con­sid­er­a­tion doesn’t seem to be a True Re­jec­tion.

After these con­sid­er­a­tions, I con­ser­va­tively set the value of a lifeyear in the fu­ture to $50000.

val_year=50000

Prob­a­bil­ity of Revival

Spe­cific equa­tions and val­ues have been pro­posed, usu­ally yield­ing prob­a­bil­ity of suc­cess 0 < x < 10%. For ex­am­ple, Steven Har­ris in 1989 es­ti­mated 0.2-15%, R. Mike Perry in the same ar­ti­cle runs a differ­ent anal­y­sis to ar­rive at 13-77%, Ralph Merkle sug­gests >85% (con­di­tional on things like good preser­va­tion, no dystopia, and nan­otech); Robin Han­son calcu­lated in 2009 a ~6% chance, Roko gave 23%; Mike Dar­win in 2011 (per­sonal com­mu­ni­ca­tion) put the odds at <10%; an in­for­mal sur­vey of >6 peo­ple (LW dis­cus­sion) av­er­aged ~17% suc­cess rate; Jeff Kauf­man in 2011 pro­vides a calcu­la­tor with sug­gested val­ues yield­ing 0.2%; The 2012 LessWrong sur­vey yields a mean es­ti­mate of cry­on­ics work­ing of 18% (n=1100) and among ‘vet­er­ans’ the es­ti­mate is a lower 12% (n=59) - but in­ter­est­ingly, they seem to be more likely to be signed up for cry­on­ics.

Gw­ern Bran­wen, “Plasti­na­tion ver­sus Cry­on­ics”, 2014

Be­sides these es­ti­mates, there ex­ist also two re­lated ques­tions on the pre­dic­tion web­site metac­u­lus. “Be­fore 1 Jan­uary 2050, will any hu­man cry­on­i­cally pre­served for at least 1 year be suc­cess­fully re­vived?” has a me­dian prob­a­bil­ity of 16% (n=117), “If you die to­day and get cry­on­i­cally frozen, will you “wake up”?” re­ceives 2% (n=407). I am not sure where the differ­ence comes from, per­haps ei­ther from wor­ries about the qual­ity of cur­rent preser­va­tion or from a low trust in the longevity of cry­on­ics or­gani­sa­tions. This google sheet con­tains 7 es­ti­mates of suc­cess: 0.04%, 0.223%, 29%, 6.71%, 14.86%, 0.23% and 22.8%, with var­i­ous differ­ent mod­els un­der­ly­ing these es­ti­mates.

Calcu­lat­ing the mean of these re­sults in a chance of ~13%:

It would cer­tainly be in­ter­est­ing to set up a pre­dic­tion mar­ket for this ques­tion, or get a team of su­perfore­cast­ers to es­ti­mate it, but ba­si­cally, it seems like for a young or mid­dle-aged per­son, the es­ti­mated prob­a­bil­ity is around 10%. How­ever, the peo­ple sur­veyed are of­ten sym­pa­thetic to cry­on­ics or even signed up, and peo­ple in gen­eral are over­con­fi­dent, so be­ing con­ser­va­tive by halv­ing the es­ti­mate seems like a good idea.

prob_succ=0.05

Years Gained

Con­di­tional on be­ing re­vived, what is the av­er­age life ex­pec­tancy?

If re­vival is achieved, it seems highly likely that ag­ing and most de­gen­er­a­tive dis­eases have been erad­i­cated (it makes lit­tle sense to re­vive a per­son that is go­ing to die again in 10 years). Also, most re­vival sce­nar­ios hinge upon ei­ther the fea­si­bil­ity of very ad­vanced nan­otech­nol­ogy, which seems to be highly con­ducive to fix­ing ag­ing, or on whole brain em­u­la­tion sce­nar­ios, which would likely make ag­ing un­nec­es­sary (why on pur­pose de­grade a digi­tal brain?).

If re­vival hap­pens, there are still risks from ac­ci­dents and homi­cide or suicide that can kill the reusci­tated cry­on­i­cist, as well as ex­is­ten­tial risks that face all of hu­man­ity.

The web­site Pols­tats illus­trates the risks purely from ac­ci­dents and homi­cides us­ing data from the In­for­ma­tion In­surance In­sti­tute. They ar­rive at “a much more im­pres­sive 8,938 years” av­er­age life ex­pec­tancy. An an­swer on Math­e­mat­ics Stack­Ex­change to the ques­tion “What’s the av­er­age life ex­pec­tancy if only dy­ing from ac­ci­dents?” ar­rives at 2850 years.

Tak­ing ex­is­ten­tial risks into ac­count is a bit harder. It is un­clear whether most of the prob­a­bil­ity mass for ex­is­ten­tial risks should be placed be­fore reusci­ta­tion of cry­on­ics pa­tients be­comes fea­si­ble, or af­ter it. It is also un­clear how high the ex­is­ten­tial risk for hu­man­ity is over­all. As­sum­ing that the ex­is­ten­tial risk for hu­man­ity over the next 10000 years is ~40% (this num­ber is pretty much a guess), and half of that risk is placed be­fore reusci­ta­tion, then the life ex­pec­tancy of cry­on­ics is .

That num­ber should be qual­ified fur­ther in an “Age of Em” sce­nario: that sce­nario will con­tain less nat­u­ral risks (em­u­la­tion can be backed up, they live in a simu­lated world where homi­cide risks and care ac­ci­dents make no sense), but an em also suffers from the risk of not hav­ing enough money to con­tinue be­ing run, and from the fact that the em era might not last sev­eral sub­jec­tive mil­len­nia. This sce­nario de­serves fur­ther con­sid­er­a­tion (see also Han­son 1994).

To con­clude, it seems like reusci­tated cry­on­i­cists will on av­er­age live around 4500 years, al­though there is room for de­bate on this num­ber.

years_gain=4500

Prob­a­bil­ity of Be­ing Preserved

It seems like not all peo­ple who sign up for cry­on­ics re­main cry­on­i­cists un­til their death, and not all cry­on­i­cists who die as mem­bers ac­tu­ally get pre­served.

There seems to be very lit­tle data about this ques­tion, but as an ex­tremely con­ser­va­tive es­ti­mate I would put the ra­tio of mem­bers of cry­on­ics or­ga­ni­za­tions who ac­tu­ally get pre­served at 60% (it seems likely that the ac­tual num­ber is higher). For­tu­nately, a cry­on­ics mem­ber can in­crease this num­ber by be­ing dili­gent about their cry­on­ics ar­range­ment, liv­ing near the preser­va­tion fa­cil­ity be­fore death, in­form­ing fam­ily mem­bers about their ar­range­ment, try­ing to lead a safe life and keep­ing con­tact to their cry­on­ics or­gani­sa­tion.

prob_pres=0.6

Sur­viv­ing Un­til LEV

The benefit of cry­on­ics is only re­al­ized in one case: One lives to the planned year of sign­ing up, but then dies be­fore LEV. Both dy­ing be­fore sign­ing up or liv­ing un­til LEV make the value of cry­on­ics $0. One can calcu­late the prob­a­bil­ity of this sce­nario by mul­ti­ply­ing the prob­a­bil­ities of liv­ing un­til signup with the prob­a­bil­ity of then dy­ing be­fore LEV.

To calcu­late the prob­a­bil­ity of liv­ing to a given age, we can use the gom­pertz dis­tri­bu­tion again:

func­tion prob_liveto(age)
	re­turn gom­pertz(age)/​gom­pertz(curage)
end

The prob­a­bil­ity of dy­ing be­fore LEV is 0 if LEV has already oc­curred:

if curyear+(age-curage)>levyear then
	re­turn 0

Othe­wise, we as­sume that one has signed up for cry­on­ics at age and now wants to know the prob­a­bil­ity of dy­ing un­til LEV. That is the same as , or the prob­a­bil­ity of liv­ing un­til curage+(levyear-curyear) given one has already lived un­til age.

else
	re­turn 1-(gom­pertz(curage+(levyear-curyear))/​gom­pertz(age))
end

Conclusion

The com­plete code for the model can be found here.

Stan­dard Parameters

With the pa­ram­e­ters pre­sented above, it turns out that it is op­ti­mal to sign up for cry­on­ics right away, mainly be­cause the mo­ti­va­tion drift pun­ishes the pro­cras­ti­na­tion quite heav­ily.

Cur­rently 20 years old

At the age of 20 years, the value of sign­ing up for cry­on­ics the same year is $2797894 () ac­cord­ing to this model, pro­long­ing the de­ci­sion un­til one is 30 re­duces this num­ber to $1666580 (), and wait­ing un­til 40, 50 and 60 years yields a value of $982100 (), $559610 () and $287758 (), re­spec­tively.

.l(“nplot”)

data::.r()

grid([0],(#data),[10];[0],(|/​data),[1000000])

xti­tle(“Years from now”)
yti­tle(“Dol­lar value of sign­ing up for cry­on­ics”)
barplot(data)
draw()

Value of signing up for cryonics in n years at age 20, standard parameters.

Cur­rently 40 years old

The val­ues of sign­ing up for cry­on­ics look very similar to the val­ues for a 20 year old. Perform­ing the signup im­me­di­ately at age 40 is worth $6590556 ($~$6.6*10^6$) at age 40 and is the best time to do it.

.l(“nplot”)

data::.r()

grid([0],(#data),[10];[0],(|/​data),[1000000])

xti­tle(“Years from now”)
yti­tle(“Dol­lar value of sign­ing up for cry­on­ics”)
barplot(data)
draw()

Value of signing up for cryonics in n years at age 40, standard parameters.

Without Mo­ti­va­tion Drift

Since many peo­ple ques­tion the idea of mo­ti­va­tion drift and trust them­selves in the fu­ture a lot, one can simu­late this trust by set­ting the de­cay pa­ram­e­ter to 1.

In this model, a very differ­ent pic­ture emerges:

.l(“nplot”)

data::.r()

grid([0],(#data),[10];[0],(|/​data),[1000000])

xti­tle(“Years from now”)
yti­tle(“Dol­lar value of sign­ing up for cry­on­ics”)
barplot(data)
draw()

Value of signing up for cryonics in n years at age 20, no motivation drift.

It is still op­ti­mal to sign up with­out hes­i­ta­tion, but now the differ­ence is much lower.

$ lua cry­oyear.lua 20 50000 0.05 0.6 4500 1 | sort -n | tail −10
2785676.2860511: 29
2787605.1801168: 28
2789611.0731771: 27
2791107.7280825: 26
2792420.5648782: 25
2793783.1701729: 24
2794997.5035013: 23
2796078.6567918: 22
2797040.1939684: 21
2797894.3040717: 20

This means that cry­ocras­ti­na­tion is not that much of a sin even with a lot of self trust.

The Critic’s Scenario

Some­body who is very crit­i­cal might ob­ject and ar­gue that the prob­a­bil­ity of suc­cess is much lower, and even if cry­on­ics suc­ceeds, it will not lead to ex­tremely long lifes­pans. Let’s say they also don’t be­lieve in value drift. Such a per­son might pro­pose the fol­low­ing as­sign­ment of vari­ables:

curage=20
val_year=50000
prob_succ=0.01
years_gain=50
prob_pres=0.6
de­cay=1

In this case, sign­ing up for cry­on­ics has nega­tive value that con­verges to 0 the older one gets:

$ lua cry­oyear.lua 20 50000 0.01 0.6 50 1 | sort -n | tail −10
−80320.313507659: 69
−78526.595695932: 70
−77042.774290053: 71
−75002.570281634: 72
−72832.328023689: 73
−70916.116822976: 74
−68452.980090227: 75
−65840.832675399: 76
−63425.293847013: 77
−60490.80006618: 78

Please note that the fol­low­ing graph should have nega­tive val­ues on the y-axis. This should get fixed some­time in the fu­ture.

.l(“nplot”)

data::-.r()

grid([0],(#data),[10];0,(|/​data),[10000])

xti­tle(“Years from now”)
yti­tle(“Dol­lar value of sign­ing up for cry­on­ics”)
fillrgb(0.4;0.4;1)
barplot(data)
draw()

Critical perspective on cryonics

Other Modifications

It is pos­si­ble to think of many other mod­ifi­ca­tions to the pa­ram­e­ters in the script, in­clud­ing the prob­a­bil­ity of cry­on­ics suc­cess, the value of a lifeyear, the amount of years gained, or even big­ger mod­ifi­ca­tions such as adding mod­els for the prob­a­bil­ity of the de­vel­op­ment of life ex­ten­sion tech­nol­ogy in the near fu­ture.

The reader is en­couraged to en­ter their own value and ex­e­cute the script to de­ter­mine whether it is ad­van­ta­geous for them to sign up for cry­on­ics, and if yes, whether cry­ocras­ti­na­tion would be a good idea.

Ap­pendix A: A Guessti­mate Model

The web­site Guessti­mate de­scribes it­self as “A spread­sheet for things that aren’t cer­tain”. It pro­vides Monte-Carlo simu­la­tions in a spread­sheet-like in­ter­face.

I used Guessti­mate to calcu­late the un­cer­tainty in the value pro­vided by sign­ing up for cry­on­ics as a 20 year old. The model is available here.

Variables

Most of the pa­ram­e­ters were sim­ply taken from this text, but some de­serve more ex­pla­na­tion.

Year for Longevity Es­cape Velocity

When I give any kind of timeframes, the only real care I have to take is to em­pha­size the var­i­ance. In this case I think we have got a 50-50 chance of get­ting to that tip­ping point in mice within five years from now, cer­tainly it could be 10 or 15 years if we get un­lucky. Similarly, for hu­mans, a 50-50 chance would be twenty years at this point, and there’s a 10 per­cent chance that we won’t get there for a hun­dred years.

Aubrey de Grey, “Aubrey de Grey on Progress and Timescales in Re­ju­ve­na­tion Re­search”, 2018

The 90% con­fi­dence in­ter­val for this vari­able lies in : Aubrey de Grey gives a mean of 2038, I be­lieve that num­ber to be quite op­ti­mistic, but not com­pletely so. He doesn’t give a lower bound, but judg­ing from the rea­son­able as­sump­tion that longevity es­cape ve­loc­ity is likely not 2 years away, this seems like a log-nor­mal dis­tri­bu­tion-ish, which is also what I used in the spread­sheet, with a 90% con­fi­dence in­ter­val in .

Age at Death

Un­for­tu­nately, Guessti­mate doesn’t sup­port Gom­pertz dis­tri­bu­tions, so I had to ap­prox­i­mate the age of death by as­sum­ing that it was a log-nor­mal dis­tri­bu­tion with the 90% con­fi­dence in­ter­val in , but mir­rored along the y-axis. The data by Wolfram Alpha looks similar to the end re­sult, and both have a mean age of death of ~83 years.

Years Lived After Revival

This was an­other log-nor­mal dis­tri­bu­tion, with a 90% con­fi­dence in­ter­val of years. Why the huge range? On the one hand, re­vival with­out suffi­cient re­ju­ve­na­tion tech­nol­ogy seems un­likely, but pos­si­ble; an­other pos­si­bil­ity is be­ing re­vived and then dy­ing in an ac­ci­dent or war. The high up­per range ac­counts for a very sta­ble fu­ture with re­ju­ve­na­tion tech­nol­ogy. Although the dis­tri­bu­tion is log-nor­mal, the mean is still 32000 years, and the 50th per­centile is around 1300 years.

Value of Lifeyears After Revival

Here, I as­sumed that both nega­tive and pos­i­tive de­vel­op­ment of the fu­ture is equally pos­si­ble, re­sult­ing in a nor­mal dis­tri­bu­tion with a 90% con­fi­dence in­ter­val in . I per­son­ally be­lieve that be­ing re­vived in a fu­ture with nega­tive value is quite un­likely, as out­lined in this sec­tion, but it’s always the thing that peo­ple bring up and want to ar­gue about end­lessly (per­haps try­ing to con­vince me of their val­ues or test whether mine are ac­cept­able), so I in­cluded the pos­si­bil­ity of sub­stan­tial nega­tive de­vel­op­ment.

Provider Cost per Year

Im­ple­ment­ing the whole mem­ber­ship_fees in Guessti­mate seems pos­si­ble, but in­cred­ibly bur­den­some. I ap­prox­i­mated it us­ing a nor­mal dis­tri­bu­tion with a 90% con­fi­dence in­ter­val of .

Value

The re­sult is cer­tainly in­ter­est­ing: in this model, sign­ing up for cry­on­ics has a mean value of $35m and a me­dian of ≈-$100k (per­haps be­cause of longevity es­cape ve­loc­ity ar­riv­ing and mak­ing the value sim­ply the cost for sign­ing up), but with very long tails, es­pe­cially on the pos­i­tive side: a fifth per­centile of -$1.59m, and a 95th per­centile of squints $63.2m – quite a range!

The min­i­mum and max­i­mum of the simu­la­tion are even more ex­treme: -$1b for the min­i­mum and $11.3b for the max­i­mum.

Be­cause of these huge num­bers, per­haps it makes sense to try to vi­su­al­ize them log­a­r­ith­mi­cally. I ex­ported the num­bers for the vari­able ‘Value’ from Guessti­mate and con­verted them into a Klong ar­ray.

.l(“math”)
.l(“nplot”)

.l(“https://​​ni­plav.github.io/​​val­ues.kg″)

log­val­ues::_{:[x<0;-ln(-x):|x=0;x;ln(x)]}‘val­ues
log­val­ues::log­val­ues@<log­val­ues
in­ci­dence::{(log­val­ues@*x),#x}‘=log­val­ues

grid((*log­val­ues),(*|log­val­ues),[5];[0],(|/​​#’=log­val­ues),[100])

sc­plot2(in­ci­dence)
draw()

Distribution of value of signing up for cryonics, logarithmically

Note that the scale is log­a­r­ith­mic to the nat­u­ral log­a­r­ithm (sym­met­ri­cally for both nega­tive and pos­i­tive val­ues), not the log­a­r­ithm to base 10, be­cause this makes the data more gran­u­lar and there­fore eas­ier to un­der­stand.

As one can see, the dis­tri­bu­tion has turned out sort-of bi­modal: Most cases of sign­ing up for cry­on­ics have a value of -$100k (pre­sum­ably be­cause longevity es­cape ve­loc­ity ar­rives first), the rest is ei­ther very nega­tive of very pos­i­tive. To be ex­act, (+/​{*|x}‘flr({*x<0};in­ci­dence))%#log­val­ues of cases have nega­tive value, and (+/​{*|x}’flr({*x>0};in­ci­dence))%#log­val­ues of cases have pos­i­tive value. Of the ones with nega­tive value, most are sim­ply flukes where longevity es­cape ve­loc­ity ar­rives first: 2286%#log­val­ues .

Conclusion

In this model, sign­ing up for cry­on­ics is still a good idea from a strict ex­pected-value per­spec­tive. But de­ci­sion pro­cesses with a pre­cau­tion­ary prin­ci­ple might be much more wary of do­ing any­thing rash be­fore fu­tures with nega­tive value can be ruled out.

Discussions