collaboratedgoogleadsense

MNSES9100-Science,ethicsandsociety-Autumn2017|1WillArtificialIntelligenceBrightenorThreatentheFutureHeminAliQadirDepartmentofInformaticsUniversityofOslo1.
IntroductionArtificialIntelligence(AI)isauniqueinnovation.
Itisdifferentinthewaythatithasabilitytothink,reason,andsolveproblems.
TheeffortsnowistowardsmakinggeneralAIwhichmimicshumanbraincapability.
ItisexpectedthatAIswillaffectalmosteverythingconnectedtohumanlifee.
g.
ethics,privacy,security,employment,economy,healthcareetc.
Creatingintelligentmachinesisnotwithoutconcernsandethicalissueswhenitcomestohowsafetheywillbe,andhowtoensurethattheywillnotharmhumansandothermorallyrelevantcreatures.
Additionally,howtheymightdifferfromhumansinconcerningethicalissuesassociatedwithprivacyandconfidentiality.
Moreover,AsAIprogressgetsaccelerated,morerobotsandautonomoussystemswillbecreatedandreplacethehumanlabor.
Incontrast,AIhasitsadvantages.
Ithasmanyapplicationsthatwouldleadhumanitytowardsmakingthisplanetabetterplacetolive.
Throughoutthisessay,IwillbeansweringwhetherweneedtomakethemornotbyarguingtheadvantagesanddisadvantagesofAI.
Foreverytechnologytosurvive,itsadvantageshavetooutweighitsdisadvantages.
MostofdisadvantagesofAIsareassociatedwithethicsandconcernsthatAIsbringtohumansociety.
Itmaylookscarytomanypeople.
Somesayitwillsavehumanity,andotherssayitdestroysus.
Eitherway,ifithappens,theworldwillbechangedforever.
2.
WhatisArtificialIntelligence(AI)Whenbabiesarebornthereisactuallynosomuchinformationaboutthisworldintheirheads,butgraduallytheylearnaboutthethingsaround.
Intheirveryearlyages,theylearntosuck,swallow,andgrasp.
Lately,theybecomeasophisticatedhumanbeingthatcanthink,reason,planahead,andsolveproblems.
Thatiscalledchilddevelopment.
Mosthumansgothroughaseriesoflifeperiodsandlearningprocessestoearninformation,recognizedifferentthings,speakdifferentlanguages,andevendesignadvancemachineslikearobot.
Thatisobviouslythepowerofthehumanbrainthatisincrediblyabletolearnquickly,thinkdeeply,andadaptremarkablywithnewstuff[2].
MNSES9100-Science,ethicsandsociety-Autumn2017|2Oneofthemostadvancedmachinesthatthemankindhaseverbuiltiscomputer.
AccordingtotheOxforddictionary,acomputeris"anelectronicdeviceforstoringandprocessingdata,typicallyinbinaryform,accordingtoinstructionsgiventoitinavariableprogram.
"[1]However,computerscanonlydowhattheyhavebeentaughtbyhumans.
Theycannotlearnbythemselvesandsolveproblemswithoutthehumans'support.
Aquestionwhich'sdrawnmostcomputerscientists'andengineers'attentionformanydecadesiswhetheritispossibletobuildacomputermindthatcanthink,learn,andbehavelikeahumanThisresearchdisciplineiscalledArtificialIntelligence(AI)nowadays.
ThetermofAIisfirstcoinedin1956bythegreatAmericancomputerscientist,alsocalledFatherofAI,JohnMcCarthyattheMassachusettsInstituteofTechnology(MIT).
Thisnewestfieldofstudyemphasizesthecreationofintelligentmachinesinawaythatmimicshumanbrainintelligence,acomputermindthatcanthinkandacthumanlyandrationally[3].
AIisactuallyabranchofcomputersciencewiththecontributionsfromotherscientificdisciplinessuchasBiology,Psychology,Linguistics,Mathematics,andEngineering.
ThemainmotivationofAIistodevelopcomputerfunctionscorrelatedwithhumanintelligence[29].
Figure1.
AIisbasedonvariousdisciplinesArtificialintelligenceisgenerallycategorizedintotwotypes;narrow(weak)AIandgeneral(strong)AI.
NarrowAIattemptstoperformonlyonetask.
NarrowAIsystemsarealreadypartofsystemssuchasAppleSiri,Facebookfriendrecommendations,inourcars,ourhomesandinourdailytrafficcontrol.
And,it'sbeenaroundforyearsfordoingaspecifictaskbetterthananyhuman[6].
However,thechallengenowistocreateageneralAIsystemthatattemptstoperformthefullrangeofhumancognitiveabilitiessuchashumanlanguage,perception,objectrecognition,attention,motorskills,etc.
,asystemthatcanlearnanddevelopitselfandbecomemoreintelligentwithobservingmoredata[25].
AccordingtoRayKurzweil,Google'sDirectorofEngineering,AIwillreachhumanintelligenceby2029,andby2045AIcomesintoitsownandbecomethemostintelligentspeciesontheplanet.
HethinksthattheprocesstowardsgeneralAIhasalreadybegunfordecades[27].
HisMNSES9100-Science,ethicsandsociety-Autumn2017|3claimisbasedonsomeavailableintelligentmachinesthathavealreadyoutperformedhumanintelligence.
Wehavemanyexamplesforsuchintelligentmachinesshowingbetterperformancethanhumaninsomespecifictasks.
Forinstance,in1997,IBM'sDeepBluebeatGarryKasparovatchess,becomingthefirstcomputertodefeatahumanworldchampion[31].
Justrecently,adeeplearningbasedalgorithmcalledResNetdesignedbyMicrosofthaspresentedbetterperformancethanhumaninimageclassification[30].
3.
ThemainconcernsofAIThemostcontroversialissueassociatedwithAIsistheintimidationsthattheywillprobablybringtohumansociety.
ThedevelopmentofAIishappeningmagicallyfast[10].
SomeresearchersanddevelopersthinkthatAIcouldgrowextremelystrong,anditwouldbecomeproblematicforhumanstocontrol.
AIsystemsaredevelopedbyintroducingintothemasmuchhumanintelligenceaspossible,suchascapacityforlogic,understanding,self-awareness,learning,emotionalknowledge,planning,creativity,andproblem-solving.
Ifithappens,thehumansthemselveswillseemtobethreatened,andthatmightbeabigthreattooursafety.
Theself-optimizingAIsystemscanbecomesuper-intelligentandoutperformhumanintelligence.
Then,itcouldbeverydifficulttostopthemfromrealizingtheirgoals,whichmayleadtounintentionalconsequences[3].
StephenHawking,whoisoneofthemostinfluentialscientistsalivetoday,believesthathumankindisindangerofself-destructionduetoAI.
InhisinterviewwithWIREDHawkingsaid,"Asuper-intelligentAIwillbeextremelygoodataccomplishingitsgoals,andifthosegoalsaren'talignedwithours,we'reintrouble.
"OtherbignameswithinscienceandtechnologyhavethesameanxietyasHawkingdoes[4].
MichaelVassar,anAIexpertandthechiefscienceofficerofMetaMedResearch,believesthatartificialsuper-intelligencewillannihilatehumanityifwedonotapproachAIwithsignificantcaution.
HealsoservedasthepresidentoftheMachineIntelligenceResearchInstitute.
InavideopostedbyBigThink,hestatesthat,"Ifgreater-than-humanartificialgeneralintelligenceisinventedwithoutduecaution,itisallbutcertainthatthehumanspecieswillbeextinctinveryshortorder"[6].
TheSwedishphilosopher,NickBostrom,Whoiswell-knownforhisworkonAI,thinksthattheproblemofcontrolhastobesolvedbeforesuperintelligenceisbroughtintobeing.
Otherwise,upcomingdevelopmentsinartificialintelligenceinvestigationmayposeasupremehazardtohumanity.
ThisthreathasmadeElonMusk,afamousbusinessmanwhoistheco-founderandCEOofmanylargecompaniessuchasTeslaInc.
,towarnusaboutthethreatsthatAIwillmakewithoutanyregulationandrestriction[8].
Forthatreason,inDecember2015,heannouncedOpenAIwhichisanon-profitableartificialintelligence(AI)researchorganization.
HeaimstosecurethesafetyinbuildingArtificialGeneralIntelligence(AGI),inawaythatisbeneficialtohumanity[8][9].
MNSES9100-Science,ethicsandsociety-Autumn2017|4Thisbigthreataside,anotherriskofAIisonourjobsinthefutureasitsubstitutesforhumanroles[25].
AIsystemshavealreadybegunreplacingthehumankindinfewindustries[3].
Inawidelyinvestigatedresearchbasedonstatistics,CarlBenediktandMichaelOsborneestimatedthatabout47percentofemployeesintheUSareatriskoflosingtheircareersintwenty-firstcenturybecauseofAI[10].
Equivalentestimationshavebeenshownforothercountries.
DeloittewhocollaboratedwithCarlBenediktandMichaelOsbornesuggeststhat35percentofjobsintheUKareathighriskofcomputerizationinthenext10to20years[11].
1Pakistan,Bangladesh,Vietnam,andIranarethelargestcountriesbypopulationnotincluded2France,Germany,Italy,Spain,andtheUnitedKingdom.
Note:NumbersmaynotsumduetoroundingSOURCE:OxfordEconomicForecasts;Emsidatabase;USBureauofLaborStatistics;McKinseyGlobalInstituteAnalysisFigure2ActivitiesthataremorevulnerabletoAIMNSES9100-Science,ethicsandsociety-Autumn2017|5Mckinseyglobalinstitutehasconductedastudytoinvestigatethenumberofautomatableactivitiesinlargecountriesbypopulation.
Globallyspeaking,theythinkthat1.
1billionemployeesequivalentto$15.
8trillioninwagesaretechnicallyautomatableactivities.
Theygavemoredetailsaboutthisinanexhibitthatisshowninfigure2.
Thisisactuallyabigshiftofincomefromlabortocapitalandraiseshighinequalityinmanycountries.
Lookatfigure2andseeifyourcountryisatriskofthismatter.
Asitcanbeseenthereissomevariationamongcountries.
Apparently,AIwillbringmoreriskstohumanity,butthetwomajorconcernsaredissuaded.
SomeotherissueswithAIthatmanyscholarsaretalkingaboutareitseffectonpeople'saffordancesandcapabilities.
People'sabilitytoaccomplishsometasksmayweakenasmachinesdeliversuperperformance.
Forexample,children'sabilityhasbeenreducedtodobasicarithmeticoperationsaftercalculatorsareintroducedtoclassrooms[25].
4.
EthicalIssuesofAIThetwomainconcernsdiscussedintheprevioussectioncanalsobeconsideredastwomajorethicalissuesthatAIwillintroduceintooursocietyinthefuture.
TheethicalimpactofAIonthesocialinequalityisextremelyworrying.
Thinkaboutthisgapwhenallrevenuesgotofewerpeopleoncecompaniesstoprelyingonthehumanworkforce.
EzekielEmanuel,abioethicistandformerhealthcareadvisertoBarackObama,says"Ifyoulookatwhatgivespeoplemeaningintheirlives,it'sthreethings:meaningfulrelationships,passionateinterests,andmeaningfulwork.
"Healsoaddsthat"Meaningfulworkisaveryimportantelementofsomeone'sidentity.
"Hesaysthattheriskofsuicide,substanceabuseanddepressionwillincreasewhentheindustriesreplacehumanbyrobots.
AIisnowtouchingmostaspectsofourlivesinonewayoranother.
Forexamples,wherethecancerouscellsarelocated,whetherwegetabankloan,whatbooksorflightstobuyonline,etc.
Allofthesecannowautomaticallybedonebyadvancesoftware.
ManyresearchersthinkthatthebiggestchallengefacingAIisanethicalone,includingprivacyconcerns.
[19].
NickBostromsays,"Machineintelligenceisthelastinventionthathumanitywilleverneedtomake.
"Andtherefore,theethicalissuesrelatedtoAIarequitedifferentfromanyethicalproblemsarisingintheexistingtechnology.
Inhispaper,hesurveyedsomeoftheuniqueethicalissuesincreatingAI[28].
Hebelievesthatbuildinggeneralsuperintelligenthastoberegulatedethically;otherwise,theywouldbecomeanautonomousagentduetothecapabilityofindependentcreativityandofmakingitsownplans[28].
AperfectexampleforthisiswhenFacebookunwillinglyshutdowntwoAIsaftertheystartedtalkingtoeachotherintheirownlanguage[20].
MNSES9100-Science,ethicsandsociety-Autumn2017|6MoralityisanimportantissuewithmakingAI.
BuildingAIsmoreintelligentthanhumanbringsupconcernsrelatedtosafetyandtrust.
PeterNorvig,directorofresearchatGoogle,statesthat"Thechallengenowistomakesureeveryonebenefitsfromthistechnology.
"Giantcompanies(likeGoogle,Facebook,Microsoft,IBM,Amazonetc.
)arecompetingagainsteachothertodevelopAImachines,butthereisalsoaquestiontowhatextentthesemachinesareusedforwarfarepurposesItisfeasibleforpeopletoutilizeAI-basedsystemsforharmfulinadditiontohelpfulpurposes[25].
PoliciesandprocessesofmakingAIsshouldworktoensurethatthebenefitsofAItechnologieswillbespreadwidelyandfairly.
NickBostromandEliezerYudkowskyintheirDraftforCambridgeHandbookofArtificialIntelligencedepeicedepictascenariothatanAI-basedsystemcouldleadtodiscriminationbasedonrace,sexualorientation,orotherfactors.
SinceAIsystemiscontinuouslylearningbasedondecisiontreesorBayesiannetworks,theythinkthatanAI-basedsystemcouldlearntodiscriminatequalifiedapplicantsbecauseoftheirrace,gender,etc.
,sincethisisthekindofinformationonwhichitcanlearnandfurtherdevelop.
ItispossibleforAIalgorithmtousetheaddressdataofapplicantswhowerebornorpreviouslylivedinareasoccupiedpredominantlybypoverty-strickenorblackpeople[24].
AndthathasbeenshowninarecentstudyconductedbyLatanyaSweeney,aprofessoratHarvardUniversity.
Basedon120,000Internetsearchads,shefoundrepeatedincidencesofracialbias.
ShedemonstratedthatapersonwhowasraciallyassociatedwiththeblackcommunitywouldmorelikelybeconnectedtoacriminalrecordbyGoogleAdSense[21].
Anotherbigethicalquestionisresponsibilityandassignmentofblame.
Forexample,manyethicalissuesariseforself-drivingvehicles.
Questionslike;whohastobeblamedforanaccidenthowacarisprogramedtoactinasituationinwhichthepriorityhastobegiventooneofthechoiceamountothers,especiallywhentherearesplit-seconddecisionstobemadeaboutwhomtoputatrisk[25].
ThispictureiswelldepictedinamoviecallediRobotwhenarobotistryingtorescueapoliceofficerfromacarcrashwhileleavingafour-yearchildinthemiddleoftheroad.
Insomecountries,suchasfourstatesintheUS(Nevada,Florida,California,andMichigan),OntarioinCanada,theUnitedKingdom,France,andSwitzerland,someruleshavebeenpassedfortestingtheself-drivingcarsonpublicroads.
Howevertheserulesdon'taddressissuesregardingresponsibility[25].
TherewasanongoingdebateonarecentaccidentoccurredaftertheTesla'sdriver-assistfailedtodetectanothervehicleinitspathastheresultitsownerdied[13].
Moreover,questionslikewhoisresponsibleforafailuremadebyanintelligentmedicaldevicethatputsapatientinadangerofdeath.
ThethreattoourprivacyisanadditionalethicalissueofAI.
SomanypeoplehaveaccountswiththeirprivateinformationontheInternet.
ItishighlyprobablethatanAI-basedsystemwillbeabletoMNSES9100-Science,ethicsandsociety-Autumn2017|7revealtheprivateinformationaboutindividualstothepublicthroughsomedecisionsandpredictions.
PrivacyconcernsrelatedtoAIinthetransportationsectorarealsowidespread.
Forexample,recordingofpersonalactivity,especiallywithpersonalsmartphoneshasincreasedaswell.
SinceanAIprogramcanrecognizespeechandunderstandlanguages,itcaninterpretconversationsviaemails,telephones,andonlinevideo/speakchatting[3].
Justaweekbeforewritingthisarticle,manywell-knownnewspapersreportedthatGoogleisabletotracktheAndroiduserseveniftheyhaveswitchedoffthelocationserviceortakeoutthesimcard[22][23].
5.
TheadvantagesofAIDespitetheethicalissuesandconcernsassociatedwithArtificialintelligence,itismandatorytotalkaboutthebenefitsthatAIsbringtohumansocietyinvariousfieldsoflife.
AnAIsystemthatisabletotransformmassiveamountofcomplexandambiguousdataintoinsightcouldpotentiallydiscloselong-heldsecretsandsupporthumanstosolvesomeoftheworld'smostlastingproblems[14].
Inourcurrentlife,AIsystemshavebeendeployedinawiderangeofapplications.
Everyday,weuseAIwithoutbeingawareof.
Whenweuseourcreditcardstobuygoods,itisanAIthatdecideswhetherthistransactionhastobeapprovedornot.
TheGPSsystemisbasedonanAIalgorithmtofindthebestroutetothedestinationenteredbytheusers.
ThewebsearchenginessuchasGoogleandMicrosoftBingrelyonAItogiveuswebpagesthataremostappropriatetoourinquiry.
ThemobilefacedetectionandiPhoneSiriarebasedonAIsystemsaswell.
Electronictranslateserviceswhichareabletotranslateandunderstand70languagesarebasedonstatisticalmachinelearning,whichispartofAI[14].
Ithasalsobeenusedinotherfieldssuchasmedicine,robotics,remotesensing,scientificdiscovery,stocktrading,etc.
[14].
ThetrendsforthefutureAIaretosimulatehumanintelligenceinordertobeabletosolveproblemsandmakedecisions.
Therefore,futureAIcanprovidemanypioneeringadvantagesoverhumans.
Itcanaddresshumanlimitationssuchascomputationallimitationsandfaultyheuristics[15,16].
Itwillhelphumanitytosolveproblemsandreachdecisionsinafasterspeedandwithlessuncertainty.
Inaddition,permanency,reliability,andcost-effectivenessaregreatadvantagesofthefutureAIaswell[16].
Thehumanbrainneedsgoodintuitioninmakingsenseofalotofdata.
Thatmakesourbraintoactslowlyintakingstepsandevaluatingallpossiblesequencesofactions.
Inthechessgame,forinstance,itisveryhard,evenforthebestplayerintheworld,thatallpossibilitiesofmovementwouldcometoourmindinaveryshorttime.
Incontrast,accordingtoMoore'slaw,computer'scomputingpowergetsdoubledevery18months.
Therefore,withbetterprocessorcapacity,AIMNSES9100-Science,ethicsandsociety-Autumn2017|8algorithmscanfacilitatefasterdecisionmaking,anddecreasecomputationaltimeinsolvingcomplexproblems[14,16].
Sometimesadirectmathematicalexpressioncan'tbeestablishedforcauseandeffect.
However,anAIcanmodelaprobabilityinferencebasedontheuncertaintybetweenreal-lifecauseandeffectscenarios.
ThisisauniquefeatureofAIthatisnotfoundinotherinventions[16].
Analgorithmwillwidelybeutilizedinmanydifferentapplicationswhenitsreliabilityhasbeenconfirmed.
OneofthefeaturesofAIis"reinforcementlearning"thatmakesanAIsystembecomeevenmorereliablebycontinuouslylearningfromreal-worldsuccessandfailure[16].
Oursocietyisenormouslyaffectedduetofailuresofhumanreasoning.
Bothphilosophers,StanovichandKahneman,agreethatbecauseofcomputationallimitationsthehumanmindfrequentlyreasoninsuboptimalorincorrectways.
Contrarilytoourbrain,anAIwouldreasonmorereliablywithitsself-modificationtoovercomesuchbiases[15].
EachyearintheUSalone,40000peopledieincaraccidents.
Mostofthecaraccidentsareduetohumanfault.
Abouthalfoftheliveswillbesavedwiththeadoptionofself-drivingcarsbasedonanAIsystem.
AIresearchershavebeeninvestigatingwaystodevelopanintelligentself-drivingcarthatisabletoexpectwhatothercarsandpedestriansareabouttodo,understandwhatishappingintheroads,andbeabletocopewithunexpectedsituations.
Obviously,suchanintelligentdrivingsystemwilloutperformhumandrivingperformance[14].
Currently,somespecificAItoolsareinuseinourtrafficsystem.
Thesetoolscanautomaticallygatherinformationfromthetrafficsensorsandconvertthemintointelligentagents.
Thathasimprovedthehumanefficiencytoreliablydetectandreporttrafficaccidentsorpredicttrafficconditions[16].
ItisestimatedthatourtransportationsystemswillbemoresecureandreliableinthefuturewithadvancedAIs.
Thefutureofreal-timetrafficmanagementandcontrolplanswillbemoreaccurateandwell-organizedwithwideradoptionofAItoolsindifferentareasoftransportation[16].
AIwillhavemajorimpactsonindustriesandautomation.
AnAIrobotcanbemadetofulfillitstasktirelesslyandwithoutbeingaffectedbyemotions.
WithanAIthatdoesnotneedtotakearestandsleep,theprobabilitiesoferrorarealmostnilandthatleadstoperfectionandfasterproduction.
Inotherwords,AIwillminimizethecostoflivingbecauseitreducestheneedforhumanmanpower,thusreducingoperationalcosts[16,18].
AnotheradvantageofAIisthattheycouldbeusedtohelpusunderstandourworldandexploretheuniverse.
Weashumanarevulnerableandunabletoendurethehostileandunfriendlyenvironmentofthespace.
However,anintelligentAIcanbedesignedforsuchanaggressiveenvironmentwithoutbeingaffectedphysicallyandnotfunctioningproperly[17].
MNSES9100-Science,ethicsandsociety-Autumn2017|96.
ConclusionArtificialIntelligencecomprisesauniquesetoftechnologiesthatarequitedifferentfromotherinventionsthathumanityhaseverdeveloped.
Itisexpectedthatitwillchangetheworldforever.
Whenitcomestoansweringthequestionwhetheritwillbrightenorthreatenourfuture,Iwouldsaythatisadouble-sidedquestion.
Itactuallydependsonhowitwillbeintroduced.
Amachinethatcouldbesmarterandwillnotgetboredandtiredwillhaveclearadvantagesoverhumans.
Inthiswayitmayhelpmankindinvariousaspectsoflifesuchasinmedicine,self-drivingcars,fasterproductionlines,solvingproblemseasier,protectingusfromalien,etc.
Incontrast,iftherearenoworldwideregulationsandstandardsthatensuretoalleviatethechallengesrelatedtoethicsandriskofAIs,theycouldbethemostdestructiveinnovation,andseriouslythreatenthehumansociety.
Therefore,theymustbedesignedinaccordancewithstandardrulesandregulations.
Inwaysthattheyformtrustandunderstanding,andrespectallkindoflivingspeciesandcivilrights.
Inshort-term,theultimateeffectofAIisonincomelevels.
However,thatdependssubstantiallyongovernmentpolicies.
ItisverylikelythatAIwillbeusedformanytypesofjobsandreplacehumanworkforce,butalsooffersnewformsofjobs.
PeoplewhofindtheiremploymentalteredorendedastheconsequenceofAImaystarttoinvestinlearningnewskillsandseekingnewformsofworkandincomeopportunities.
References1.
Oxford(2017)'computer',inOxfordDictionary.
Availableat:https://en.
oxforddictionaries.
com/definition/computer2.
Russell,StuartJ.
,PeterNorvig,andErnestDavis.
Artificialintelligence:amodernapproach.
Boston:Pearson,2010.
3.
ArtificialIntelligence;intelligentsystems,TutorialsPoint(I)Pvt.
Ltd,2015.
https://www.
tutorialspoint.
com/artificial_intelligence/artificial_intelligence_tutorial.
pdf4.
"StephenHawkingbelieveshumankindisindangerofself-destructionduetoAI.
"Futurism.
November20,2017.
AccessedDecember01,2017.
https://futurism.
com/stephen-hawking-believes-humankind-danger-self-destruction-ai/.
5.
"AnotherExpertJoinsStephenHawkingandElonMuskinWarningAbouttheDangersofAI.
"Futurism.
February14,2017.
AccessedNovember23,2017.
https://futurism.
com/2-expert-thinks-ai-will-undoubtably-wipe-out-humanity/.
6.
BBCNews.
AccessedNovember25,2017.
http://www.
bbc.
com/news/av/technology-34224406/what-is-artificial-intelligence.
7.
Bostrom,Nick.
Superintelligence:paths,dangers,strategies.
Oxford,UnitedKingdom:OxfordUniversityPress,2016.
8.
Futurism.
August14,2017.
AccessedNovember23,2017.
https://futurism.
com/elon-musk-reminds-us-of-the-possible-dangers-of-unregulated-ai/.
9.
"ElonMusk.
"Wikipedia.
December07,2017.
AccessedNovember26,2017.
https://en.
wikipedia.
org/wiki/Elon_Musk.
10.
Frey,CarlBenedikt,andMichaelA.
Osborne.
"Thefutureofemployment:Howsusceptiblearejobstocomputerisation"TechnologicalForecastingandSocialChange114(2017).
11.
"Frombrawntobrains:theimpactoftechnologyonjobsintheUK|DeloitteUK.
"DeloitteUnitedKingdom.
December04,2015.
AccessedNovember28,2017.
MNSES9100-Science,ethicsandsociety-Autumn2017|10https://www2.
deloitte.
com/uk/en/pages/growth/articles/from-brawn-to-brains--the-impact-of-technology-on-jobs-in-the-u.
html.
12.
Manyika,J.
,MichaelChui,MehdiMiremadi,JacquesBughin,KatyGeorge,PaulWilmott,andMartinDewhurst.
Afuturethatworks:automation,employment,andproductivity.
Placeofpublicationnotidentified:McKinseyGlobalInstitutte,2017.
13.
Kelion,Leo.
"Teslasaysautopilotinvolvedinsecondcarcrash.
"BBCNews.
July13,2016.
AccessedDecember05,2017.
http://www.
bbc.
com/news/technology-36783345.
14.
FrancescaRossi-EuropeanParliament,PE571.
380.
"ArtificialIntelligence:PotentialBenefitsandEthicalConsiderations"15.
Sotala,Kaj.
"AdvantagesofArtificialIntelligences,Uploads,AndDigitalMinds.
"InternationalJournalofMachineConsciousness04,no.
01(2012)16.
Chowdhury,Mashrur,andAdelW.
Sadek.
"ArtificialIntelligenceApplicationstoCriticalTransportationIssues.
"AdvantagesandLimitationsofArtificialIntelligence,2012,6-10.
17.
ShuklaShubhenduS.
,JaiswalVijay.
"ApplicabilityofArtificialIntelligenceinDifferentFieldsofLife".
InternationalJournalofScientificEngineeringandResearch(IJSER).
www.
ijser.
inISSN(Online):2347-3878,Volume1Issue1,September2013.
18.
Saranya,D.
"AStudyonArtificialIntelligenceandItsApplications".
InternationalJournalofAdvancedResearchinComputerandCommunicationEngineeringVol.
5,Issue4,April2016.
19.
Lufkin,Bryan.
"Future-WhythebiggestchallengefacingAIisanethicalone.
"BBC.
March07,2017.
AccessedDecember10,2017.
http://www.
bbc.
com/future/story/20170307-the-ethical-challenge-facing-artificial-intelligence.
20.
Griffin,Andrew.
"Facebook'sartificialintelligencerobotsshutdownaftertheystarttalkingtoeachotherintheirownlanguage.
"TheIndependent.
November24,2017.
AccessedDecember10,2017.
http://www.
independent.
co.
uk/life-style/gadgets-and-tech/news/facebook-artificial-intelligence-ai-chatbot-new-language-research-openai-google-a7869706.
html.
21.
Sweeney,Latanya.
"DiscriminationinOnlineAdDelivery.
"SSRNElectronicJournal,2013.
22.
Mailonline,TimCollinsFor.
"AndroiddevicescovertlysendlocationdatatoGoogleevenwhenusersDISABLEthesetting.
"DailyMailOnline.
November22,2017.
AccessedDecember10,2017.
http://www.
dailymail.
co.
uk/sciencetech/article-5104685/Google-secretly-collects-Android-users-locations.
html.
23.
Gibbs,Samuel.
"GooglehasbeentrackingAndroidusersevenwithlocationservicesturnedoff.
"TheGuardian.
November22,2017.
AccessedDecember10,2017.
https://www.
theguardian.
com/technology/2017/nov/22/google-track-android-users-location-services-turned-off-sim.
24.
Bostrom,Nick,andEliezerYudkowsky.
"Theethicsofartificialintelligence.
"TheCambridgeHandbookofArtificialIntelligence:316-3425.
Tyagi,Amit.
"ArtificialIntelligence:BoonorBane"SSRNElectronicJournal,2016.
26.
"OneHundredYearStudyonArtificialIntelligence(AI100),"StanfordUniversity,accessedAugust1,2016,https://ai100.
stanford.
edu.
27.
"KurzweilClaimsThattheSingularityWillHappenby2045.
"Futurism.
October05,2017.
AccessedJanuary18,2018.
https://futurism.
com/kurzweil-claims-that-the-singularity-will-happen-by-2045/28.
NickBostrom,"EthicalIssuesinAdvancedArtificialIntelligence"-postedinAI&Singularity:Oxfordphilosopher,Nov18,200329.
Flasinski,Mariusz.
"Introductiontoartificialintelligence".
Switzerland:Springer,2016.
30.
He,Kaiming,XiangyuZhang,ShaoqingRen,andJianSun.
"DeepResidualLearningforImageRecognition.
"2016IEEEConferenceonComputerVisionandPatternRecognition(CVPR).
31.
DeepBlue(chesscomputer).
"Wikipedia.
January15,2018.
AccessedJanuary18,2018.
https://en.
wikipedia.
org/wiki/Deep_Blue_(chess_computer).