Scots Philosophical Association

University of St. Andrews

Are We Living in a Computer Simulation?
Author(s): Nick Bostrom
Source: The Philosophical Quarterly, Vol. 53, No. 211 (Apr., 2003), pp. 243-255
Published by: Oxford University Press on behalf of the Scots Philosophical Association and the
University of St. Andrews
Stable URL: http://www.jstor.org/stable/3542867
Accessed: 22-06-2015 04:23 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/3542867?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

Oxford University Press, Scots Philosophical Association and University of St. Andrews are collaborating with JSTOR
to digitize, preserve and extend access to The Philosophical Quarterly.

http://www.jstor.org

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ThePhilosophical
Vol.53,No.2
Quarterly,

April
2003

ISSNoo3-8g94

ARE WE LIVING IN A COMPUTER SIMULATION?

ByNICK BOSTROM
is very
to
I argue
thatat leastoneofthe
is true:
(i) thehuman
likely
species
following
propositions
is extremely
extinct
a 'posthuman'
become
civilization
before
reaching
stage;(2) anyposthuman
toruna sgnficant
number
(orvariations
unlikely
ofsimulations
ofitsevolutionary
thereof);
history
thatthere
is a
ina computer
Itfollows
thatthebelief
simulation.
certainly
living
(3) wearealmost
chance
whorunancestor-simulations
isfalse,
thatweshallonedaybecome
signficant
posthumans
I discusssomeconsequences
unlesswearecurrently
ofthisresult.
livingina simulation.

I. INTRODUCTION
Many worksof sciencefictionas well as some forecastsby serioustechnologistsand futurologists
predictthatenormousamountsofcomputingpower
will be available in the future.Let us suppose for a momentthat these
are correct.One thingthatlatergenerationsmightdo withtheir
predictions
oftheirforebearsor of
super-powerful
computersis rundetailedsimulations
people like theirforebears.Because theircomputerswould be so powerful,
theycould runa greatmanysuch simulations.
Suppose thatthesesimulated
people are conscious(as theywould be if the simulationswere sufficiently
and ifa certainquitewidelyacceptedpositionin thephilosophy
fine-grained
of mindis correct).Then itcould be thecase thatthevastmajorityofminds
likeoursdo notbelongto theoriginalrace but ratherto people simulatedby
the advanced descendantsof an originalrace. It is thenpossibleto argue
thatifthiswerethecase, we wouldbe rationalto thinkthatwe are likelyto
be among the simulatedmindsratherthan among the originalbiological
ones. Thereforeifwe do notthinkthatwe are currently
livingin a computer
simulation,we are not entitledto believe thatwe shall have descendants
who will run lots of simulationsof theirforebears.That is the basic idea.
The restofthispaperwillspellitout morecarefully.
thisthesismayholdforthoseengagedin futuristic
Apartfromtheinterest
there
are
also
morepurelytheoretical
rewards.The argumentis
speculation,
a stimulusforformulating
some methodological
and metaphysical
questions,
? The Editorsof ThePhilosophical
Road, Oxfordox4 2DQUK,
2003. Publishedby BlackwellPublishing,
Quarterly,
9600 Garsington
and 350 Main Street,Maiden,MA02148,USA.

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

NICK BOSTROM

244

and it suggestsnaturalistic
analogiesof certaintraditionalreligiousconceptions,whichsomemayfindamusingor thought-provoking.
The structure
of thepaper is as follows.First,I formulatean assumption
which I need to importfromthe philosophyof mind in order to get the
argumentstarted.Secondly,I considersome empiricalreasonsforthinking
thatrunningvastlymanysimulationsof humanmindswould be withinthe
capabilityof a futurecivilizationthathas developedmanyof thosetechnologiesthatcan alreadybe shownto be compatiblewithknownphysicallaws
and engineering
constraints.
This partis notphilosophically
butit
necessary,
an
incentive
for
attention
to
the
Then
rest.
follows
the
core
provides
paying
of the argument,whichmakesuse of some simpleprobability
and
theory, a
sectionprovidingsupportfora weak indifference
the
principle
argument
of thedisjunctionmentioned
employs.Lastly,I discusssome interpretations
in theabstract,whichformstheconclusionofthesimulationargument.
II. THE ASSUMPTION OF SUBSTRATE-INDEPENDENCE
A common assumptionin the philosophyof mind is that of substrateThe idea is thatmentalstatescan superveneon any ofa broad
independence.
class of physicalsubstrates.Provideda systemimplementsthe rightsortof
and processes,it can be associatedwithconscious
computationalstructures
experiences.It is not an essentialpropertyof consciousnessthatit is implemented on carbon-basedbiological neural networksinside a cranium:
silicon-based
processorsin a computercould in principledo thetricktoo.
forthisthesishave been givenin theliterature,
and although
Arguments
itis notentirely
I shallheretakeitas given.
uncontroversial,
The argumentI shall presentdoes not, however,depend on any very
or computationalism.
For example,I need
strongversionof functionalism
not assumethatthethesisof substrate-independence
is necessarily
true(either
or
a
that
analytically metaphysically)merely
computerrunninga suitable
in
would
fact
be
conscious.
I need not assume thatin
Moreover,
program
orderto createa mindon a computeritwouldbe necessaryto programitin
such a way thatit behaveslike a human in all situations,
includingpassing
theTuringtest,etc. I need onlytheweakerassumptionthatit would suffice
forthegenerationofsubjectiveexperiencesthatthecomputational
processes
of a human brain are structurally
detail,
replicatedin suitablyfine-grained
such as on the level of individualsynapses.This attenuatedversion of
is quitewidelyaccepted.
substrate-independence
nerve
Neurotransmitters,
growthfactorsand other chemicalsthat are
smallerthana synapseclearlyplaya role in humancognitionand learning.
? The EditorsofThePhilosophical
Qjuarterly,
200oo3

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ARE WE LIVING IN A COMPUTER SIMULATION?

245

of thesechemicals
The substrate-independence
thesisis not thatthe effects
affect
but
that
are smallor irrelevant, rather
subjectiveexperienceonly
they
via theirdirector indirectinfluenceon computationalactivities.For exin subjectiveexperiencewithoutthere
ample,iftherecan be no difference
in synapticdischarges,then the requisitedetail of
also being a difference
simulationis at thesynapticlevel(or higher).
III. THE TECHNOLOGICAL

LIMITS OF COMPUTATION

At our currentstage of technologicaldevelopment,we have neither

to createconscious
sufficiently
powerfulhardwarenor therequisitesoftware
have been givento theeffect
mindsin computers.But persuasivearguments
thatif technologicalprogresscontinuesunabated,thenthesetechnological
will eventuallybe overcome.Some authorsargue that this
shortcomings
be
stage may
only a fewdecades away.' Yet presentpurposesrequireno
assumptionsabout the time-scale.The simulationargumentworksequally
well forthosewho thinkthatit willtake hundredsof thousandsofyearsto
reach a 'posthuman'stage of civilization,where humankindhas acquired
showto be conmostofthetechnologicalcapabilitiesthatone can currently
sistentwithphysicallaws and withmaterialand energyconstraints.
willmakeitpossibleto
Such a maturestageoftechnologicaldevelopment
convertplanetsand otherastronomicalresourcesintoenormously
powerful
hard to be confidentin any upperbound on the
computers.It is currently
As we
computingpower thatmay be available to posthumancivilizations.
we cannot rule out the possibility
are stilllackinga 'theoryof everything',
thatnovelphysicalphenomena,not allowedforin currentphysicaltheories,
thatin our currentundermay be utilizedto transcendthose constraints
theoretical
limits
on
the
information
standingimpose
processingattainable
in a givenlump of matter.2
We can withmuch greaterconfidenceestablish
lowerbounds on posthumancomputation,by assumingonly mechanisms
that are already understood.For example, Eric Drexler has outlineda
1 See, e.g., K.E. Drexler,Engines
theComing
Era ofNanotechnology
ofCreation:
(London: Fourth
International
Estate,1985);N. Bostrom,'How Long BeforeSuperintelligence?',
ofFutures
Journal
Machines:WhenComputers
ExceedHuman
Studies,2 (1998); R. Kurzweil, The Age of Spiritual
Mind
(New York: Viking,1999); H. Moravec, Robot:MereMachineto Transcendent
Intelligence
(OxfordUP, 1999).
2 I.e., constraintssuch as the Bremermann-Bekenstein
bound and the black hole limit:
HJ. Bremermann,'Minimum Energy Requirementsof InformationTransferand Com21 (1982),pp. 203-17;J.D. Bekenstein,'Entropy
puting',International
ofTheoretical
Physics,
Journal
Contentand Information
Flow in SystemswithLimitedEnergy',Physical
D 30 (1984),
Review,
pp. 1669-79; A. Sandberg,'The Physicsof InformationProcessingSuperobjects:the Daily
LifeamongtheJupiterBrains',Journal
andTechnology,
ofEvolution
5 (1999)C The EditorsofThePhilosophical
2003
Quarterly,

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

NICK BOSTROM

246

designfora systemthe size of a sugar cube (excludingcoolingand power

per second.3Anotherauthor
supply)thatwould performIo21instructions
givesa roughestimateof I042operationsper second fora computerwitha
massoftheorderofa largeplanet.4(Ifwe could createquantumcomputers,
or learnto buildcomputersout ofnuclearmatteror plasma,we could push
closerto the theoreticallimits.Seth Lloyd calculatesan upper bound fora
I kgcomputerof5 x 1050 logicaloperationsper secondcarriedout on -I031
bits.5However, it sufficesfor my purposesto use the more conservative
estimatethatpresupposesonlycurrently
knowndesignprinciples.)
The amountof computingpowerneeded to emulatea humanmindcan
likewisebe roughlyestimated.One estimate,based on how computationally
ofa piece ofnervoustissuewhich
expensiveitis to replicatethefunctionality
we have alreadyunderstoodand whose functionality
has been replicatedin
silice,namely,contrastenhancementin the retina,yieldsa figureof - IO14
estimate,
operationsper secondforthe entirehumanbrain.6An alternative
based on the numberof synapsesin the brain and theirfiringfrequency,
givesa figureof-IO16-IO17 operationsper second.7Conceivably,evenmore
could be requiredifwe want to simulatein detailthe internalworkingsof
synapsesand dendritictrees.However,it is likelythat the human central
nervoussystemhas a highdegreeof redundancyon themicroscaleto comand noisinessof itsneuronalcomponents.One
pensateforthe unreliability
would thereforeexpect a substantialefficiencygain when using more
reliableand versatilenon-biological
processors.
Memory seems to be no more stringenta constraintthan processing
power.8Moreover,since the maximumhuman sensorybandwidthis -Io8
bitsper second,simulatingall sensoryeventsincursa negligiblecost comthecorticalactivity.
We can therefore
use theprocessing
pared to simulating
to
simulate
the
central
nervous
as
powerrequired
system an estimateofthe
totalcomputationalcostofsimulating
a humanmind.
If the environmentis included in the simulation,this will require
additionalcomputingpower- how much,dependson thescope and granularityofthesimulation.Simulatingtheentireuniversedownto thequantum
levelis obviouslyinfeasible,
unlessradicallynewphysicsis discovered.But in
orderto geta realisticsimulationofhumanexperience,muchless is needed
- onlywhateveris requiredto ensurethatthesimulatedhumans,interacting
3 K.E. Drexler,Nanosystems
(New York:JohnWiley& Sons, 1992).
4 RJ. Bradbury,'MatrioshkaBrains',working
(2002), http://www.aeiveos.com/
manuscript
-bradbury/MatrioshkaBrains/MatrioshkaBrains.html.
5 S. Lloyd, 'Ultimate Physical Limits to Computation',Nature,406 (31 August 2000),
pp. 1047-54.
6 H. Moravec,MindChildren
(HarvardUP, 1989).
7 See my'How Long beforeSuperintelligence?'.
8 See references
in foregoing
footnotes.
2003
? The Editorsof ThePhilosophical
Quarterly,

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ARE WE LIVING IN A COMPUTER SIMULATION?

247

do notnoticeany
in normalhumanwayswiththeirsimulatedenvironment,
The microscopicstructure
of the insideof the Earth can be
irregularities.
safelyomitted.Distant astronomicalobjects can have highlycompressed
need onlyextendto thenarrowband ofproverisimilitude
representations:
pertiesthatwe can observefromour planetor solar systemspacecraft.On
thesurfaceof Earth,macroscopicobjectsin inhabitedareas mayneed to be
continuouslysimulated,but microscopicphenomena could probablybe
filledin ad hoc.What you see throughan electronmicroscopeneeds to look
itscoherencewith
but you usuallyhave no way of confirming
unsuspicious,
unobservedpartsofthemicroscopicworld.Exceptionsarisewhenwe deliberatelydesignsystemsto harnessunobservedmicroscopicphenomenathat
operatein accordancewithknownprinciplesto get resultswe are able to
The paradigmcase of thisis a computer.The simulaverifyindependently.
tionmay therefore
need to includecontinuousrepresentation
of computers
down to the level of individuallogic elements.This presentsno problem,
sinceour currentcomputingpoweris negligiblebyposthumanstandards.
Moreover,a posthumansimulatorwould have enoughcomputingpower
in all human brainsat all times.
to keep trackof the detailedbelief-states
when
it
saw
a
that
human
was
about to make an observationof
Therefore,
themicroscopicworld,itcould fillin sufficient
detailin thesimulationin the
domain
as
and
where
needed.
Should
appropriate
any erroroccur, the
directorcould edit the statesof any brainsthathave become aware of an
the directorcould
anomalybeforethisspoilsthe simulation.Alternatively,
skipback a fewsecondsand rerunthesimulationso as to avoid theproblem.
It thus seems plausible that the main computationalcost in creating
simulations
thatare indistinguishable
fromphysicalrealityforhumanminds
in the simulationresidesin simulating
organicbrainsdown to the neuronal
or sub-neuronallevel. As we build more and fastercomputers,the cost of
our machinesmighteventually
come to dominatethecostofsimsimulating
While it is notpossibleto get a veryexact estimate
ulatingnervoussystems.
of thecostof a realisticsimulationof humanhistory,
we can use
-oI33-IO36
as
a
estimate.9
As
we
more
operations
rough
gain
experiencewithvirtual
for
reality,we shall get a bettergrasp of the computationalrequirements
such
worlds
realistic
to
in
their
visitors.
But
if
even
case,
making
appear
any
the estimateis inaccurateby several ordersof magnitude,this does not
mattermuch formy argument.I notedthata roughapproximationof the
computationalpower of a planetary-mass
computeris lo42operationsper
second, and that assumesonly alreadyknownnanotechnologicaldesigns,
whichare probablyfarfromoptimal.A singlesuchcomputercould simulate
9 IOO billionhumansx 50 years/humanx 30 millionsecs/yearx
each humanbrainper second= [1033,Io36]operations.

[o014,

Iol7

operationsin

C The EditorsofThePhilosophical
2003
Quarterly,

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

NICK BOSTROM

248

theentirementalhistoryofhumankind(I shallcall thisan ancestor-simulation)

by usingless than one millionthof itsprocessingpower forone second.A
posthumancivilizationmay eventuallybuild an astronomicalnumberof
such computers.I can conclude that the computingpower available to a
to run a huge number of ancestorposthumancivilizationis sufficient
simulationseven ifit allocatesonlya veryminutefractionofitsresourcesto
that purpose. I can draw this conclusion even while leaving a huge
substantial
marginoferrorin all our estimates:
Post-humancivilizationswould have enough computingpower to run
even whileusingonlya tinyfraction
hugelymanyancestor-simulations
oftheirresourcesforthatpurpose.

IV. THE CORE OF THE SIMULATION ARGUMENT

The basic idea of thispaper can be expressedroughlyas follows:if there
were a substantialchance that our civilizationwill get to the posthuman
thenhow come we are notliving
stageand runmanyancestor-simulations,
in sucha simulation?
I shalldevelopthisidea intoa rigorousargument.I need to introducethe
notation:
following
fp: Fractionof all human-leveltechnologicalcivilizationsthat surviveto
reacha posthumanstage
run by a posthuman
N: Average number of ancestor-simulations
civilization
H: Averagenumberof individualsthathave livedin a civilizationbeforeit
reachesa posthumanstage.
The actualfractionofall observerswithhuman-type
experiencesthatlivein
simulations
is then
sim=

+H
(f)VH)

Writing
fiforthe fractionof posthumancivilizationsthatare interestedin
runningancestor-simulations
(or thatcontainat least some individualswho
are interestedin them and have sufficient
resourcesto run a significant
numberof such simulations),
and NI forthe average numberof ancestorsimulations
runbysuchinterested
we have
civilizations,
?Th
Editors
of Th Philosophicalarter,
C The Editors of The PhilosophicalQuarterly,
2oo3

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ARE WE LIVING IN A COMPUTER SIMULATION?

249

and thus

F. fm=

(fif N) +i

Because of the immensecomputingpower ofposthumancivilizations,

XNis
extremely
large,as I pointedout in the previoussection.What (F) showsis
thatat leastoneofthefollowing
threepropositions
mustbe true:
I. fp=O
2. fj-O
fsizr

3.

V. A BLAND INDIFFERENCE PRINCIPLE

I can takea further
stepand concludethatgiventhetruthof(3),one'scredence
in the hypothesis
thatone is in a simulationshouldbe close to unity.More
generally,if we knew that a fractionx of all observerswith human-type
and we have no information
to indicatethat
experienceslivein simulations,
our own particularexperiencesare any more or less likelythan other
human-typeexperiencesto have been implementedin vivoratherthan in
thenour credencethatwe are in a simulationshouldequal x:
machina,
S. Cr(sIMI fm = x) = x.
This step is sanctionedby a veryweak indifference
principle.Two cases
need to be distinguished.
The firstcase, whichis theeasiest,is whereall the
minds in question are like our own in the sense that they are exactly
identicalwithours:theyhave exactlythesame information
and
qualitatively
the same experiencesas we have. The second case is wherethe mindsare
'like' each otheronly in the loose sense of being the sort of minds that
are typicalof human creatures,but where theyare qualitativelydistinct
fromone anotherand each has a distinctsetof experiences.I maintainthat
even in the lattercase, where the minds are qualitativelydifferent,
the
simulationargumentstillworks,providedthatwe have no information
bearing on the questionofwhichof thevariousmindsare simulatedand which
are implemented
biologically.
A detaileddefenceofa stronger
principle,whichimpliestheabove stance
forboth cases as trivialspecialinstances,has been givenin the literature.10
ofthatdefencehere,but I can bring
Space does notpermita recapitulation
10

In, e.g., N. Bostrom,'The Doomsday Argument,Adam and Eve, UN++, and Quantum
Bias: Observation
Joe', Synthese,
127 (2001), pp. 359-87; and most fullyin my book Anthropic
Selection
andPhilosophy
EfectsinScience
(New York: Routledge,2002).
Quarterly,
2003
? The Editorsof ThePhilosophical

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

NICK BOSTROM

250

out one of the underlyingideas by rehearsingan analogous situationof a

morefamiliarkind.Suppose thatx% ofthepopulationhas a certaingenetic
sequence S withinthe part of theirDNA commonlydesignatedas 'junk
of S (shortofwhat
thatthereare no manifestations
DNA'. Suppose further
bethere
are
no
known
correlations
would turnup in a gene assay)and that
unless
Then
tweenhavingS and any observablecharacteristic.
quiteclearly,
one has had one's DNA sequenced,it is rationalto assigna credenceofx%
of thefact
thatone has S. And thisis so quite irrespective
to the hypothesis
mindsand experithatthe people who have S have qualitatively
different
ences fromthe people who do not have S. (They are different
simplybecause all humanshave different
experiencesfromone another,not because
ofanyknownlinkbetweenS and whatkindofexperiencesone has.)
The same reasoningholds if S is not the propertyof havinga certain
ofbeingin a simulation,
assuming
geneticsequencebut insteadtheproperty
us
to
that
enables
we
have
no
information
that
differences
predictany
only
between the experiencesof simulatedminds and those of the original
biologicalminds.
It shouldbe stressedthatthebland indifference
principleexpressedby (S)
onlybetweenhypothesesabout whichobserverone
prescribesindifference
about whichof theseobserversone is. It
is, when one has no information
betweenhypotheseswhen one
does not in generalprescribeindifference
about whichof the hypothesesis true.In contrast
lacksspecificinformation
it is
with Laplacean and othermore ambitiousprinciplesof indifference,
thattend
therefore
immuneto Bertrand'sparadox and similarpredicaments
to plagueindifference
scope.
principlesofunrestricted
familiar
with
the doomsday argument"may worrythat the
Readers
invokedhere is the same assumptionas is
bland principleof indifference
responsibleforgettingthedoomsdayargumentofftheground,and thatthe
counter-intuitive
natureofsome oftheimplications
ofthelatterincriminates
or castsdoubt on the validityof the former.This is not so. The doomsday
and morecontroversial
argumentrestson a muchstronger
premise,namely,
thatone shouldreason as ifone were a randomsamplefromthe set of all
weknow
people who willeverhave lived(past,present,and future)eventhough
thatweareliving
in theearlytwenty-first
ratherthanat some pointin the
century
distantpast or the future.The bland indifference
principle,by contrast,
have
to
cases
where
we
no
information
about whichgroupof
appliesonly
we
to.
people belong
If bettingodds providesome guidanceto rationalbelief,it may also be
worthponderingthatifeverybody
were to place a bet on whethertheyare
" See, e.g.,J. Leslie, 'Is the End of the World Nigh?', ThePhilosophical
Quarter{y,
40 (I990),
pp. 65-72.
2003
Quarterly,
? The EditorsofThePhilosophical

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ARE WE LIVING IN A COMPUTER SIMULATION?

251

in a simulationor not,thenifpeople use thebland principleofindifference,

and consequentlyplace theirmoneyon being in a simulationiftheyknow
thatthatis wherealmostall people are, thenalmosteveryonewillwin their
bets.If theybet on notbeingin a simulation,
thenalmosteveryonewilllose.
It seemsbetterthatthebland indifference
principleshouldbe heeded.
Further,one can considera sequence of possiblesituationsin whichan
increasingfractionof all people live in simulations:98%, 99%, 99-9%,
99-9999%,and so on. As one approachesthelimitingcase in whicheverybody
is in a simulation(fromwhich one can deductively
inferthat one is in a
simulationoneself),it is plausibleto requirethatthecredenceone assignsto
being in a simulationshouldgraduallyapproach the limitingcase of comin a matchingmanner.
pletecertainty
VI. INTERPRETATION
The possibility
If (I) is
represented
byproposition(I) is fairlystraightforward.
then
humankind
will
almost
fail
to
reach
a
true,
certainly
posthumanlevel;
forvirtually
no speciesat our levelofdevelopmentbecome posthuman,and
it is hard to see anyjustification
forthinkingthatour own specieswill be
or
from
futuredisasters.Conditionallyon (i),
especiallyprivileged protected
we
must
a
credence
to DOOM,the hypothesisthat
therefore,
give high
humankindwillgo extinctbeforereachinga posthumanlevel:
Cr(DOOM fp= I)= I

One can imaginehypothetical

situations
wherewe have such evidenceas
would trumpknowledgeoffp.For example,ifwe discoveredthatwe were
about to be hit by a giant asteroid,thismightsuggestthatwe had been
exceptionally
unlucky.We could thenassigna credenceto DOOMlargerthan
our expectationofthefractionofhuman-levelcivilizations
thatfailto reach
In the actual case, however,we seem to lack evidencefor
posthumanity.
thatwe are specialin thisregard,forbetteror worse.
thinking
Proposition(I) does not by itselfimplythatwe are likelyto go extinct
soon,onlythatwe are unlikelyto reach a posthumanstage.This possibility
is compatiblewithour remainingat, or somewhatabove, our currentlevel
of technologicaldevelopmentfora long timebeforegoingextinct.Another
way for (I) to be true is if it is likelythat technologicalcivilizationwill
humansocietiesmightthenremainon Earthindefinitely.
collapse.Primitive
There are manyways in which humanitycould become extinctbefore
of (I) is that
reachingposthumanity.
Perhapsthemostnaturalinterpretation
we are likelyto go extinctas a resultof the developmentof some powerful
? The Editors of The PhilosophicalQuarterly,
200oo3

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

NICK BOSTROM

252

but dangeroustechnology.'2One candidateis molecularnanotechnology,

of self-replicating
whichin its maturestagewould enable the construction
nanobots capable of feedingon dirt and organic matter- a kind of
mechanicalbacteria. Such nanobots,designedfor malicious ends, could
cause theextinction
ofall lifeon our planet.13
in thesimulationargument'sconclusionis thatthe
The secondalternative
in runningancestorfractionof posthumancivilizationsthatare interested
simulationsis negligiblysmall.In orderfor(2) to be true,theremustbe a
If the number
strongconvergence
amongthecoursesofadvancedcivilizations.
of ancestor-simulations
createdby the interestedcivilizationsis extremely
extreme.
large, the rarityof such civilizationsmust be correspondingly
no
civilizations
decide
to
their
use
resources
to run
Virtually posthuman
numbers
of
ancestor-simulations.
all
Furthermore,
large
virtually posthuman
civilizationslack individualswho have sufficient
resourcesand interestto
run ancestor-simulations;
or else they have reliablyenforcedlaws that
such
individuals
from
prevent
actingon theirdesires.
What forcecould bringabout such a convergence?One mightspeculate
thatadvancedcivilizations
all developalong a trajectory
thatleads to recognitionofan ethicalprohibition
ancestor-simulations
because
againstrunning
of the suffering
thatis inflicted
on the inhabitantsof the simulation.However,fromour presentpointof view,it is not clear thatcreatinga human
race is immoral.On thecontrary,
we tendto view theexistenceof our race
as constituting
a greatethicalvalue. Moreover,convergenceon an ethical
view of the immoralityof runningancestor-simulations
is not enough: it
mustbe combinedwithconvergenceon a civilization-wide
social structure
thatenablesactivities
consideredimmoralto be effectively
banned.
Anotherpossible convergencepoint is that almost all individualposthumansin virtually
all posthumancivilizations
developin a directionwhere
This would requiresignitheylose theirdesireto run ancestor-simulations.
ficantchangesto themotivesdrivingtheirhumanpredecessors,
forthereare
ifthey
certainlymanyhumanswho would like to run ancestor-simulations
could affordto do so. But perhaps many of our human desireswill be
regarded as silly by anyone who becomes a posthuman.Maybe the
scientificvalue of ancestor-simulations
to a posthuman civilizationis
is
not
too
negligible(which
implausiblegivenits unfathomableintellectual
and
superiority), maybeposthumansregardrecreationalactivitiesas merely
12 See

my'ExistentialRisks:AnalyzingHuman ExtinctionScenariosand Related Hazards',

and
Journalof Evolution
9 (2001), for a surveyand analysis of the presentand
Technology,
anticipatedfuturethreatsto humansurvival.
13 See, e.g., Drexler;and R.A. FreitasJr,'Some Limitsto Global
Ecophagyby Biovorous
Nanoreplicators, with Public Policy Recommendations', Zyvexpreprint,
April (2000),
http://www.foresight.org/NanoRev/Ecophagy.html.
Quarterly,
2003
? The Editorsof ThePhilosophical

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ARE WE LIVING IN A COMPUTER SIMULATION?

253

a veryinefficient
way of gettingpleasure- which can be obtainedmuch
more cheaply by direct stimulationof the brain's reward centres.One
conclusionthat followsfrom(2) is that posthumansocietieswill be very
different
fromhuman societies:they will not contain relativelywealthy
independentagentswho have the fullgamutof human-likedesiresand are
freeto act on them.
The possibilityexpressedby alternative(3) is conceptuallythe most
one. If we are livingin a simulation,thenthecosmoswe are obintriguing
ofphysicalexistence.The physicsin
servingisjust a tinypiece ofthetotality
the universewherethe computerrunningthe simulationis situatedmay or
may not resemblethephysicsof theworldwe observe.While theworldwe
see is in some sense'real',itis notlocatedat thefundamental
levelofreality.
It maybe possibleforsimulatedcivilizations
to becomeposthuman.They
on powerfulcomputersthey
may thenrun theirown ancestor-simulations
build in their simulated universe. Such computerswould be 'virtual
machines',a familiarconceptin computerscience.(Javascript
web-applets,
forinstance,run on a virtualmachine- a simulatedcomputer- insidea
desktop.)Virtual machinescan be stacked:it is possible to simulateone
machinesimulating
anothermachine,and so on, in arbitrarily
manystepsof
iteration.If we do go on to createour own ancestor-simulations,
thensince
thiswouldbe strongevidenceagainst(I) and (2),we wouldtherefore
have to
concludethatwe live in a simulation.Moreover,we would have to suspect
that the posthumansrunningour simulationare themselvessimulated
beings;and theircreatorsin turnmayalso be simulatedbeings.
Realitymaythuscontainmanylevels.Even ifit is necessaryforthehierstatusof thisclaim is
archyto bottomout at some stage- themetaphysical
somewhatobscure- theremay be room fora large numberof levels of
reality,and the numbercould be increasingover time.(One consideration
thatcountsagainstthe multi-level
is thatthe computationalcost
hypothesis
forthe basement-level
simulatorswould be verygreat.Simulatingeven a
singleposthumancivilizationmightbe prohibitively
expensive.If so, then
we should expect our simulationto be terminatedwhen we are about to
becomeposthuman.)
Althoughall the elementsof such a systemcan be naturalistic,even
physical,it is possibleto draw some loose analogieswithreligiousconceptionsof theworld.In some ways,theposthumansrunninga simulationare
like gods in relationto the people inhabitingthe simulation:the posthumanscreatedtheworldwe see; theyare of superiorintelligence;
theyare
in
the
sense
that
can
interfere
in
the
'omnipotent'
they
workingsof our
worldevenin waysthatviolateitsphysicallaws; and theyare 'omniscient'in
the sense thattheycan monitoreverything
thathappens.However,all the
? The Editors of The PhilosophicalQuarterly,
2003

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

NICK BOSTROM

254

demigodsexcept those at the fundamentallevel of realityare subjectto

sanctionsbythemorepowerfulgodslivingat lowerlevels.
Furtherruminationon thesethemescould climaxin a naturalistic
theogony
and theconstraints
ofthishierarchy,
thatwouldstudythestructure
imposed
on its inhabitantsby the possibilitythat theiractionson theirown level
may affectthe treatmenttheyreceivefromdwellersof deeper levels.For
then
example,if nobody can be sure thattheyare at the basement-level,
thattheiractionswill be
everybodywould have to considerthe possibility
An
rewardedor punished,perhapsusingmoralcriteria,by theirsimulators.
Because ofthisfundamental
afterlife
wouldbe a realpossibility.
uncertainty,
even the basementcivilizationmay have a reasonto behave ethically.The
factthatit has such a reason formoral behaviourwould of course add to
everybodyelse's reasonforbehavingmorally,and so on, in a trulyvirtuous
whichit would
circle.One mightget a kindof universalethicalimperative,
be in everybody's
to obey,as itwere,'fromnowhere'.
self-interest
In additionto ancestor-simulations,
one may also considerthepossibility
ofmoreselectivesimulations
thatincludeonlya smallgroupofhumansor a
singleindividual.The restof humanitywould thenbe zombiesor 'shadowforthe fullysimulated
people' - humanssimulatedonlyat a level sufficient
ones not to noticeanythingsuspicious.It is not clear how much cheaper
shadow-peoplewouldbe to simulatethanreal people. It is noteven obvious
that it is possible for an entityto behave indistinguishably
froma real
human and yetlack consciousexperience.Even ifthereare such selective
we shouldnotthinkthatwe are in one of themunlesswe think
simulations,
theyare much more numerousthan completesimulations.There would
have to be about Ioo billiontimesas many'me-simulations'
of
(simulations
the lifeof onlya singlemind)as thereare ancestor-simulations
in orderfor
mostsimulatedpersonsto be in me-simulations.
There is also the possibility
of simulatorsabridgingcertainpartsof the
mentallivesof simulatedbeingsand givingthemfalsememoriesof the sort
of experiencesthattheywould typically
have had duringthe omittedinterval. If so, one can consider the following(far-fetched)
solutionto the
in the worldand all memoriesof
problemof evil: thatthereis no suffering
are illusions.Of coursethishypothesis
can be seriouslyentertained
suffering
at
those
times
when
one
is
not
only
currently
suffering.
whatare the implications
forus? The
Supposingwe live in a simulation,
remarks
the
are
not
all
that
radical.
foregoing
notwithstanding, implications
Our best guide to how our posthumancreatorshave chosen to set up our
worldis thestandardempiricalstudyoftheuniversewe see. The revisionsto
most parts of our beliefnetworkswould be ratherslightand subtle- in
proportionto our lack ofconfidencein our abilityto understandthewaysof
? The EditorsofThePhilosophical
Quarter~y,
2003

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions

ARE WE LIVING IN A COMPUTER SIMULATION?

255

thetruthof (3) shouldhave no

posthumans.Properlyunderstood,therefore,
tendencyto make us 'go crazy'or to preventus fromgoingabout our business and makingplans and predictionsfortomorrow.The chiefempirical
importanceof (3) at the presenttimeseemsto lie in itsrole in thetripartite
conclusionestablishedabove.14We may hope that (3) is true,since that
would decreasethe probabilityof (I), althoughifcomputationalconstraints
a simulationbeforeit reaches
would terminate
make itlikelythatsimulators
a posthumanlevel,thenour besthope wouldbe that(2) is true.
If we learn more about posthumanmotivesand resource-constraints,
maybe as a resultof developingtowardsbecomingposthumansourselves,
thatwe are simulatedwillcome to have a much richer
thenthe hypothesis
setofempiricalimplications.
VII. CONCLUSION
A technologically
mature 'posthuman'civilizationwould have enormous
computingpower.Giventhisempiricalfact,thesimulationargumentshows
is true:(I) thefractionofhumanthatat leastoneofthefollowing
propositions
levelcivilizations
thatreach a posthumanstageis veryclose to zero; (2) the
in runningancestorfractionof posthumancivilizationsthatare interested
simulations
is veryclose to zero; (3) thefraction
ofall people withour kindof
experienceswho are livingin a simulationis veryclose to one.
If (I) is true,then we will almostcertainlygo extinctbeforereaching
If (2) is true,thentheremustbe a strongconvergenceamong
posthumanity.
the courses of advanced civilizationsso that virtuallynone containsany
and are
relatively
wealthyindividualswho desireto runancestor-simulations
freeto do so. If (3) is true,thenwe almostcertainlylive in a simulation.In
the darkforestofour currentignorance,it seemssensibleto apportionone's
credenceroughlyevenlybetween (2),and (3).
(I),
Unless we are now livingin a simulation,our descendantswill almost
neverrunan ancestor-simulation.'5
certainly
University
Oxford
14For some reflections
by anotherauthoron the consequencesof (3), whichwere sparked
by a privatelycirculatedearlierversionof thispaper, see R. Hanson, 'How to Live in a
andTechnology,
Simulation',Journal
ofEvolution
7 (2001).
to manypeople forcomments,and especiallyto Amara Angelica,Robert
15 I am grateful
Bradbury,Milan Cirkovie,Robin Hanson, Hal Finney,Robert A. FreitasJr,John Leslie,
Mitch Porter,Keith DeRose, Mike Treder, Mark Walker, Eliezer Yudkowsky,and the
anonymousreferees.
C The Editorsof ThePhilosophical
Quarterly,
2o003

This content downloaded from 128.148.252.35 on Mon, 22 Jun 2015 04:23:51 UTC
All use subject to JSTOR Terms and Conditions