TransformingtheAssetEcosystem:

TheRoleofAIin

AssetManagementandServicing

WhitePaper

2

TransformingtheAssetEcosystem:

TheRoleofAIin

AssetManagementandServicing

TableofContent

ExecutiveSummary 4

ResearchMethodology 6

TheNeedforAIinAssetManagementandAssetServicing

8

CurrentStateofAssetManagement 10

AIasaCompanion:CurrentApplicationinAssetManagement 14

BeneitsandChallengesofAIinAssetManagementIndustry 20

FromAmbitiontoImpact:

BuildingtheAugmentedOrganization 24

Contacts 29

3

4

ExecutiveSummary

Whatdoesthefuturelooklikeforassetmanagementandservicingfirms?

Theassetmanagementandservicinglandscapeisbecomingincreasinglycompetitiveastraditionalplayersfacegrowingpressurefrominnovativeintechstartupsthatareredeininghowinancialservicesaredelivered.Facedwithsustainedmarginpressure,complexandevolvingregulations(e.g.,EURetailInvestmentStrategy,MiFID

II,DORA,EUAIAct,…),aswellasrisingclientexpectationsfortransparency,customizedservicesanddigitalinteractions,theseorganizationsmustadaptstructurally,operationally,andtechnologically.

Inthiscontext,ArtiicialIntelligence(AI)cannolongerbeconsideredabreakthroughconcept.Itisbecominganessentialleverfortransformationacrosstheassetmanagementvaluechain—fromportfoliodecisionsupporttooperationalefficiencyandcompliance.AI,asdeinedinthispaper,referstosystemscapableofperformingtasksthattypicallyrequirehumanintelligence,suchasreasoning,learning,decision-making,andnaturallanguage

processing.

Yet,whilethepotentialofAIiswidelyrecognized—92%oforganizationsacknowledgethestrategicimportanceofdigitizationandnewtechnologyintegration—only52%reportpartialAIintegration,andmostremainatthepilotstage.Thegapbetweenambitionandrealityisnotjusttechnical:itreflectsdeeperchallengesrelatedtodatareadiness,regulatorycomplexity,andorganizationaladaptation.Thisobservationisfurtherconirmedbythe2024AIthematicreportfromtheCSSF,whichhighlightsthatalthoughAIadoptionisprogressing,asigniicantportionofLuxembourginancialinstitutionshaveyettoinvestinthesetechnologies.However,thereportalsoanticipatesarisingtrendinlocalinvestments,particularlyingenerativeAI,overthecomingyears—signalingashiftfrom

experimentationtomorestructureddeployment.

OurindingsrevealthatAIisprimarilyusedtodaytoaugmentratherthanreplacehumandecision-making.78%ofirmsdonotuseAIforreal-timedecisions,andinsteadfocusonsupportfunctionsthatenhancejudgment,reduceprocessingtime,andimproveinformationaccess.

Despitethisconservativedeployment,earlyresultsarepromising:

?69%ofrespondentsreportreducedmanualprocessingtimeinpre-tradeactivities;

?55%noteimproveddecisionquality;

?40%reportatleastpartialAIadoptionininvestment-relatedareas.

5

However,theroadtobroaderimplementationisstillmarkedbythreemainconstraints:

1.Dataqualityandinfrastructurelimitations(only9%ofirmshavefullystandardizedandaccessibledata);

2.Limitedinternalexpertise;

3.Budgetaryconstraints,particularlyamongirmsinearlystagesofintegration.

Manyleadingplayersarerespondingbyadoptinghybridmodels,owningthestrategyandexpertisewhile

outsourcingtechnologycomponents,particularlythroughpartnershipswithAIvendors.Commonlyoutsourcedelementsincludepre-trainedAImodelsfordocumentanalysis,cloud-basedinfrastructureformodeldeployment,APIsforgenerativeAI,orintelligentautomationsolutionsforprocessingunstructureddata.Thisallows

organizationstoremainfocusedonvaluecreationwhilebeneitingfromthespeedanddepthofexternalcapabilities.

Maintainingin-houseunderstandingofAIcapabilitiesandtheirimplicationsallowsirmstosteertechnology

adoptioninlinewiththeiroperationalmodels,regulatoryconstraints(notablyEUAIAct),andlong-termstrategicgoals.

Lookingahead,thesuccessofAItransformationwilldependlessontechnologyitselfthanontheabilityto

integrateitmeaningfullywithintheorganization.Researchconirmsthat60%oftransformationsuccessdependsonpeople,versusonly10%ontoolsand30%onprocesses.

FirmsthatsucceedwillbethosethatpositionAIasacollaborativeco-pilot—onethatampliies,ratherthan

replaces,humanexpertise—andembeditintotheiroperatingmodelsinawaythatisscalable,explainable,andalignedwithregulatoryandethicalexpectations.Thisis,however,onlyairststep.Theinalstageistoevolvefroman“augmentedemployee”toan“augmentedorganization”wheretheoperatingmodelistransformedbyputtingdataandAIatitscore.Assuchcapabilitiesandoperationalefficiencyisincreasedandemployeeandcustomer

experiencesarereinvented.Changingthevaluechaininthiswaywillbecomemoreandmoremandatorytostayrelevantinthemarket.

AssetManagement-Inthecontextofthispaper,AssetManagementencompassesbothcoreinvestmentactivities(suchasportfolioconstructionandfundselection)andtheoperationalandadministrativefunctionscommonly

referredtoasassetservicing.Theseincludefundadministration,custody,andrelatedbankingfunctions.

6

Research

Methodology

ThiswhitepaperexaminesAIadoptioninassetmanagementbyassessingorganizations’maturity,collectingexecutives’perspectivesoftheEuropeanmarketandidentifyingtheprerequisitesforbecominganaugmented

organizationwithintheassetmanagementandservicingvaluechain.

DataCollection

Theprimarydataisgatheredthroughaseriesofsemi-structuredinterviewcallsconductedbetweenOctoberandMarch2025.Thesampleincludesmorethan40executivesholdingkeypositions(CEO,COO,CTO,HeadofInnovation,Transformation,PortfolioManagers

andSales)withintheirorganizations.aimingtoreflectstrategicprioritiesandoperationalrealities.

Theparticipatingorganizationscoverarepresentativecross-sectionoftheindustry:

?62%fromassetmanagement-relatedirms,

?19%fromsolutionproviders,oferingperspectiveontechnologydevelopment

?19%frombanks.

Thisvarietyinorganizationensuresperspectiveswerecapturedacrossdiferentdimensionsoftheasset

managementvaluechain.

Theparticipatingorganizationsalsodifersigniicantlyinsize,fromspecializedirmswithfewerthan100

employeestolargeglobalinstitutionswithmore

than10,000.Thisdiversityofersacomprehensive

perspectiveonvariousoperationalcontextsandhelpsidentifytrendsanddiferencesinAIadoptionbasedonorganizationalscale.

TheinterviewsfocusonorganizationsoperatingwithinEurope.withamajoritybasedinLuxembourg,reflectingitsroleasakeyhubforassetmanagementand

servicing.InsightsarealsogatheredfromparticipantsinothermajorEuropeaninancialcenters.

Asemi-structuredapproachisused,combininga

standardizedquestionnairetoensurecomparability

acrossresponses,withopen-endedquestionstoexplorenuances,speciicchallengesandapproachesingreaterdepth.

7

Theinterviewssystematicallyinvestigateseveralcoreareas:

?Perceivedimportanceandcurrentlevelofdigitalization.

?InvestmentandresourceallocationfordigitaltransformationandAI.

?AIimplementationinpre-tradeandpost-tradeactivities.

?Achievedlevelsofautomation.

?Maturityofdatainfrastructure,governance,andcloudadoption.

?FutureinvestmentplansandbarrierstoAIadoption.

?MeasuredandperceivedimpactofAIonefficiencyanddecision-making.

?Qualitativeperspectivesonchallenges,

advantages,andtherolesofserviceprovidersandregulatorsintheAItransformationinasset

management.

Toenrichtheprimaryqualitativedata,theinsightsgatheredfrominterviewswerecross-referencedwithindingsfromleadingindustryreportsandmarketresearchfocusedonAIadoptionwithintheasset

managementsector.

DataAnalysis

Thedatacollectedfollowedamulti-stageanalysisprocess:

?QuantitativeAnalysis:Responsestoscaled

questions(e.g.,maturityratings,budget

allocation,FTEs,impactpercentages)were

aggregatedandanalyzedtoidentifytrends,

levelsofadoption,andperceivedimpactsacrossthesample.

?QualitativeAnalysis:Responsestoopen-endedquestionsanddiscussionpointswereanalyzedtoidentifykeythemes,recurringchallenges,

beneits,usecases,andcommonbarriersrelatedtoAIimplementation.

?ComparativeAnalysis:Resultsweresegmentedbyorganizationtype(AssetManager,Bank,

ServiceProvider,AssetServicer)andsizeto

identifydiferencesinpriorities,maturity,andchallenges.

?Synthesis:Quantitativeandqualitativeindingswerecombinedtodevelopacomprehensive

understandingofthecurrentstateofAIadoptioninassetmanagementandasset

servicinghighlightingrecurringpatterns,

explanatoryinsights,andcommonfactorsforsuccessfulimplementation.

Thisapproachensuresabalanced,data-driven

understandingofthestateofAIinassetmanagementvaluechain,integratingbothnumericalindicators,

executiveperspectivesandthestepstheyaretakingtowardbecomingaugmentedorganizations.

8

TheNeedforAIinAssetManagementandAssetServicing

AIasaDriverofTransformation

AItechnologies—spanningmachinelearning,deep

learning,andgenerativeAI—oferassetmanagementandservicingirms,opportunitiestostreamlineprocesses,optimizedecision-making,anddevelopnewvalue-addedservices.Theconvergenceofgrowingdataavailability,

rapidtechnologicaladvancement,andpressureonproitabilityisacceleratinginterestinAIadoption.

Thesurveyconirmsthisdynamic:83%ofrespondentsexpressinterestinintegratingAIsolutionsintotheir

processes,andresultsrevealaclearcorrelationbetweenAIdeploymentandprogressindigitaltransformation.

TheDigitalMaturityGap

WhileenthusiasmforAIanddigitaltransformationrunshigh,asigniicantgapexistsbetweenambitionandexecution.While92%oforganizationsrecognizetheimportanceofdigitaltransformation,onlya

minorityhavemovedbeyondpilotorpartialAI

implementations.Mostorganizations(52%)remainatthe“partialintegration”stage,while24%arestillinthe“experimentalphase”,testingAI’spotentialincontrolledenvironments.

Thisimplementationgaphighlightspersistent

challengesacrosstheAssetManagementlandscape.

Integrationdifficultieswithlegacysystemscontinue

tohamperprogress.Dataquality,securityand

standardizationissuespreventorganizationsfrom

fullyleveragingtheirinformationassets.Digital

talentacquisitionandretentionremainsigniicant

hurdlesformanyirmsattemptingtoacceleratetheirtransformationjourney.Navigatingtheseobstacles

requiresaclear,strategicapproach–anissuethiswhitepaperseekstoexplore.

9

PurposeofThisWhitePaper

ThiswhitepaperaimstoclarifythestateofAI

adoptionintheassetmanagementsector.Basedon

industrysurveydata,itidentiiescurrenttrends,barrierstoimplementation,andkeystrategicpriorities.ThegoalistoprovideAssetManagementirmswithapracticalroadmapfornavigatingthedigitaltransformation

journey,leveragingAIcapabilities,andpositioning

themselvesforlong-termsuccessinacompetitiveandevolvinginancialecosystem.

10

CurrentStateof

AssetManagement

TechnologicalTransformation:GradualandUnequal

Theassetmanagementecosystemisundergoing

atechnologicalshift,drivenbyinnovationslikeAI,

cloudinfrastructure,blockchainandbigdata.These

technologiesarenolongeradoptedinisolationbutareincreasinglyviewedascomplementaryforcesconvergingtoreshapethisindustry.Despitethisdynamic,

transformationremainsunequal.SoftwaretechnologiessuchasRoboticProcessAutomation(RPA)havedeliveredtacticalefficiencies,particularlyinrepetitive,rules-basedtasks.However,deeperautomationremainslimited,

especiallyforprocessingunstructureddata.These

tacticalimprovementsremainlimited,especiallygiventhecontinuedrelianceonoldandcostlylegacysystems.

TheintegrationofAI,despiteamarkedinterestwith83%ofrespondentswishingtointegrateit,remainsembryonicinitsmostcriticalapplications.Most

applicationsinvolvehigh-stakesdecision-making,orsensitivedata,whichsuggestslimitationsincurrentAIcapabilitiesfortheassetmanagementandservicesindustry.Thisfurtherillustratesthelimitationsofreal-timeAIdecision-making,particularlyinhigh-stakesorsensitivecontexts,with78%ofrespondentsindicatingnotusingAIforsuchpurposesnorinpre-tradeorposttradeactivities.

AIintegrationinpre-tradeandpost-trade

activitiesisstillconstrainedbyinadequate

technologicalinfrastructure,limiteddata

accessibilityandquality,skillsshortages,andbudgetconstraints.

PrimarybarrierstoadvancingAImaturityinrespondentorganization,%ofrespondents

65%

DataAccessibility

58%

LackofTechnical

Expertise

42%

BudgetConstraints

Source:BearingPointsurveyHowisAItransformingassetservicingvaluechain?–October2024-March2025.

11

12

DataQuality,Governance,andSecurityasstructuralbarriers

DataqualitycontinuestobethemostsigniicantobstacletoadvancedautomationandrelevantAIdeployment

with48%ofrespondentsstatingthattheirdataisonly‘Partiallyaccessibleand/orstandardized’.However,the

emergenceofGenerativeAIofersnewopportunities

topartiallyofsettheselimitations,asitcanhandle

unstructured,incomplete,orheterogeneousdatamoreflexiblythantraditionalAIsystems.Thisopensup

possibilitiesforextractinginformationfromunstructureddocuments(prospectus,fundsfactsheets),forreconcilingheterogeneousdataforregulatoryreporting(multiple

sourceswithinconsistentformatforFATCAcontrols)orsummarizequalitativeperformancecommentaries.

Datagovernanceisalsoinsufficientlymature.Whilea

majorityhaveformalizedpolicies,onlyaminorityhave

‘fullygoverned’supervision.Thisgaprepresentsagrowingriskforregulatedentities,particularlyinviewofthe

increasingcomplexityoftheregulatoryenvironment,whichdemandsbettercontroloverdata.

Ontheinfrastructureside,cloudplatformsarerecognizedascriticalenablers,withagrowingshareofrespondentsidentifyingthemaspivotalforindustrygrowth,especiallyforAssetManagementcompanies.52%oforganizationsreportfullcloudintegration,buthybridenvironments

remainwidespread,creatinginteroperabilitychallengesthathinderfluidscalability.

Stateofcloudadoption,%ofrespondents

ModerateCloudIntegration21%

LimitedCloudUse21%

FullCloudIntegration50%

NoCloudAdoption8%

Nocloudadoption:theorganizationreliesonon-

premisesystems,nocriticalinfrastructure,applicationordataishostedinacloudenvironment.

Limitedclouduse:cloudisusedforsecondaryor

non-criticalfunctions(archiving,backups,messaging,

collaborativetools),butsensitivedataandcorebusinessprocessesremainhostedlocally.

Moderatecloudintegration:asigniicantpartof

businessprocessesanddataaremigratedtothecloud,butcertaincriticalfunctionsremainlocalforregulatoryorsecurityreasons.

Fullcloudintegration:criticalapplications,sensitive

dataandcorebusinessprocessestendtobemigratedtothecloud,inamulti-cloudorhybridenvironmentwithAPI-irstlogic.

OperationalFrictionPoints

Thesetechnologicalandstructurallimitationstranslatedirectlyintocriticaloperationalfrictionpoints.For

custodianbanks,reconciliationerrors,reportingdelays

andlowerSTPrates(especiallyincorporateaction

activity)aredirectconsequencesofdatafragmentation,oftenmentionedaspainpointsinpost-tradeactivities.

13

Thismakesitdifficulttoproducereliableinformationquickly.Thelowscalabilityoflegacyorhybridsystemsmakesitdifficulttoabsorbpeaksinvolumeforassetmanagersortheincreasingdiversiicationofassets.

Additionally,exceptionmanagement(handlingcases

outsideautomatedflow),whichisunavoidablegiventhelackofglobalstandardization,remainsaheavyandcostlyoperationalworkload.

IncreasingRegulatoryPressure

Atthesametime,regulatorypressureisincreasing,makingthisecosystemevenmorecomplex.Recentregulationintroduces:

1.Newoperationalconstraintswith:

?DORAregulatinginancialirmsrelationshipwithITprovidersandintensifyingoversightonthird-partyriskmanagement.

?TheEUAIActregulatingtheuseofAIaccordingtothelevelofriskmakingdeploymentmore

complexforsensitiveapplications.Acrossthe

Atlantic,USregulations(e.g.,CloudAct)mayalsocauseinancialinstitutionstobereluctanttousethesolutionsofUS-basedcompaniesduetolackoftransparencyintermsofdataprocessingandstorageprovidedbyUSprovider.

2.Andtheneedfornewsourceofrevenues:

?WiththeupcomingintroductionofRetail

InvestmentStrategy(RIS)havingdirect

implicationsoncommercialmodelsand

distributionstrategies(banoninducements,costandfeetransparencyobligations,justifyvalue

ofinvestmentsandservicefortheclients).RIS

challengestherevenuestructureandrequireassetmanagerstorethinktheirapproachintermsof:

–Costoptimizationacrossdistributionchannelsandcompliancefunctions,

–Diversiicationofincomebymaximizingcaptureofmanagementfees,

–Businessmodeladaptationtosustain

proitabilityinadvisoryanddiscretionarymanagementservices.

TheDigitalImperative

Thecurrentstateoftheassetmanagementandservicesindustryrevealsatensionbetweenstrategicambition

andoperationalreality.Ontheonehand,thereisalmostunanimousrecognitionoftheimportanceofnew

technologyandastronginterestinAI.Ontheotherhand,operationalrealityissometimeslimitedbydataqualityissuesandaccessibility,lackoftechnicalexpertise,or

budgetlimitations.

Successwillbelongtoirmsthatgobeyondadoption,

byadaptingtotheconvergenceofAI,cloud,anddatainfrastructureasinterconnectedleversoftransformation.

Top-performingorganizationsarealreadyintegrating

cloudmodernizationandAIstrategyintotheirroadmap.

Collaborationisalsoemergingasastrategiclever,

particularlyforlargeandmid-sizedplayers.Manyare

pursuingpartnerships,acquisitions,orintegrationswith

intechecosystemstostrengthentheirtechnological

capabilities,whileremainingagile.Someexamplesof

thepastfewyearsincludeDeutscheB?rse’sinvestmentinClarityAItoreinforceESGanalytics,Clearstream’s

partnershipwithinvestRFP.comtostreamlinefund

selectionprocesses,orthecollaborationbetween6MonksandZeidlertoensureregulatorycomplianceforcross-

borderfunddistribution.

14

AIasaCompanion:

CurrentApplication

inAssetManagement

AIismovingfrombeingjustabuzzwordtoarealtoolthatassetmanagersareexploring.OurtalkswithseniorleadersacrossEuropeshowthatadoptionismostly

abouttryingthingsoutinspeciicareasratherthan

bigcompany-widechanges.OrganizationsaremainlystartingpilotprojectsfocusingonprocesseswhereAIofersclearbeneits.Theyarefocusingtheirefortson3maingoals:

1.makingoperationsrunbetter,

2.improvingCompliance&Riskcontrol,and

3.createnewrevenueopportunities.

TheassetservicingandassetmanagementsectorstillappearstolackmaturityinitsAIjourney,withmany

playersremainingatthestageofusingAIasan“AI

companion”ratherthanmovingtowardbecomingtrulyaugmentedorganizations.

MakingOperationsMoreEfficientandAutomated

Between2021and2023,leadingsassetservicers

recordedanaverage+17.6%growthinassetsunder

custody,withsomeexceeding+25%by20241.While

transactionvolumesandoperationalcomplexityhavegrownsteadily,automationratesinsecuritiesprocessinghavenotincreasedatthesamepace.

Accordingtomarketstudies,Straight-ThroughProcessing(STP)ratesremainunequalacrossassetclasses.For

standardequityandbondtransactions,STPrates

typicallyrangebetween70%and85%inEurope2.Infundorderprocessing,automationratesforUCITSfundsarehigh,oftenreaching85%to90%inmajor

1Basedonareviewofpublicannualreports(2021–2023)fromarepresentativesampleofglobalassetservicers

2SWIFTSecuritiesInsightsReport(2022)

15

marketssuchasLuxembourgandIreland3.However,

foralternativeinvestmentfundsandcross-borderfunddistribution,automationremainssigniicantlylower,

oftenfallingbetween30%and50%,duetopersistentrelianceonmanualprocesseslikefaxandemail.The

situationisevenmorechallengingforcorporateactions,whereSTPratesoftenfallbetween30%and60%duetothecomplexityofeventtypesandtheneedforclientinstructions.

Thisgapbetweentherapidgrowthinactivityand

therelativelyslowimprovementinautomationlevels

createsoperationalrisksandcostpressuresforasset

servicers.IthighlightsthegrowingnecessitytoaccelerateautomationinitiativesandintegrateAI-powered

solutions—particularlyinareasliketradematching

andcorporateactionmanagement.Thelatestinvolvesmorethan50peoplefromtheissueragentsidetothesecuritiesservicingside.AIsolutionscanhelpinstitutionsabsorbrisingvolumes,reducingmanualworkloads,

improvingprocessingaccuracy,andbuildingoperationalresilienceinademandingpost-tradeenvironment.

Partofourrespondentspointedoutseveralkey

areaswhereAIisbeingusedortested.(e.g.,portfoliomanagementsupport,reportingautomation,

reconciliation).

WhenaskedabouttheirAIdeploymentstatus,the

overallpicturesuggestsworkisactivelyunderwaybut

oftennotfullyimplemented.Toillustrate,approximately40%ofparticipantsreportedbeingatleastin‘Partial

implementation’stageforAIinpre-tradeactivities,while10%reportedthesamestageforpost-tradeactivities.

Drillingdownintospeciicareaswhereorganizations

havedeployedAIsolutions,participantsmostfrequentlypointedtocoreoperationalfunctions,as:

?PortfolioManagementSupport:AIdoesnotreplaceinvestmentdecision-making,butassistsportfoliomanagersbyimprovingdataanalysis,indingpatterns,orassistingwithbuilding

syntheticindexandwatchingportfolios.

?ReportingAutomation:AIisusedtosimplifycollection,aggregation,andcreationofreports

3EFAMAFundProcessingStandardizationReport2023

(forclients,regulatorsorinternaluse).

?Reconciliation:AI,viamachinelearning,is

beinglookedatfordifficultreconciliationtasks,automatingthematchingprocess,andspeedingupindingandixingerrors.

Portfolio

management

Others33%

52%

Reconciliation19%

Reporting

33%

Beyondthesecoreprocessareas,AIcontributesto

operationalefficiencybymakingexistingautomationtechnologies“smarter”.Keyexamplesobservedinclude:

?IntelligentDocumentProcessing(IDP):AI

methods,particularlymachinelearningwith

OpticalCharacterRecognition(OCR),areused

topullout,understand,andcheckdatafrom

documentsthataren’tclearlystructured(e.g.,

prospectuses,contracts,invoices)moreaccuratelyandwithabetterunderstandingofthecontextthanoldOCRmethods.

?BoostingRoboticProcessAutomation(RPA):AIaddsintelligencetoRPAbots,allowingthemtodealwithhardertasks,handleproblemsbetter,andmakerule-baseddecisionsinsituations

wherepeopleusedtobeneeded.

?InternalAssistants&KnowledgeTools:

GenAItools,suchasinternaltools,bespoketoolsdevelopedwithAIexternalexperts(e.g.,OpenAi,Google),orcommercialplatforms(e.g.,MicrosoftCopilot)actasassistantsforemployees,makingitfastertoindinformationinlargedatasources(e.g.,internalprocedures,regulations)and

summarizinghighamountsofinformation.

16

?ContentCreation&Communication:GenAIisbeingtestedtohelpwritecommunications(e.g.,emails,internalnotes,partsofpressreleases,

marketingcontent),summarizedocuments,anddraftirstrepliesforcustomerservice,giving

teamsmoretimeforothertasks.

StrengtheningControl,Compliance,andRiskManagement

AIisalsoseenasatoolstrengtheningriskmanagementanddealingwithcomplexregulations.

Intermsofcurrentapplicationswithincontrolfunctions,ComplianceMonitoringemergedasaspeciicuse

casementionedbyasmallnumberofparticipants.

Thesetoolsaimtoautomaticallychecktransactionsorportfolios(AuM)againstprospectusesandregulatoryrestrictions,flaggingpotentialbreachesforreview.

Markettrends,highlightAI’ssigniicantpotentialto

strengthencontrol,compliance,andriskmanagement.Keyexamplesfrequentlydiscussedintheindustry

include:

?FraudDetection:AIprogramsareverytrustfulatanalyzinghugeamountsofdataandindingsmallpatternsorunusualthingsthatsuggest

fraud.TheimprovementcomparedtotraditionalmethodisthereductionoffalsepositiveusingAIAgentsthatautomatethealertreview

process,makingdecisionsonupto90%ofalertsgeneratedbyscreeningtools.Thiscouldoferamoreforward-thinkingapproachthanscreeningsystem,expandinganalyststeamoroutsourcingduringpeakperiods.

?AnomalyDetection:Beyondfraud,AIcan

watchoperationaldatatoindunusualchangesoroutliersthatmightsuggestprocessing

mistakes,orsystemissues,allowingfastercheckingandixing.

?RegulatoryWatch&ImpactAssessment:

GenAIlookspromisingforquicklyanalyzingnewregulations,summarizingkeyrules,andhelpingorganizationsigureoutthepotentialimpactontheiroperations,helpingcomplianceworkbe

moreefficient.

Amaindistinctionhighlightedbythesurveyresponses

concernsAI’scurrentroleinrealtimedecision-making.

WhenaskedifAIisappliedforreal-timedecision-making,mostparticipantssaid‘No’(70%).AIisstillmainlyusedasadecision-supporttool,providinganalysis,pointing

outrisks,suggestingactions,orautomatingpreparationsteps.However,theinaljudgmentanddecision-makingauthorityclearlyremainwithhumanexperts.Thisshowsacarefulapproach,puttinghumansupervisionirst.

17

AIforimprovingclientexperienceandcreatingnewrevenue

IntheAssetManagementandServicingsectors,wherethequalityofclientrelationshipsdirectlyinfluences

retentionandbusinessgrowth,AIca