Snowflake：2024实践中的生成式人工智能：探索利用企业数据的用例报告（英文版）.pdf

ByFernHalper,Ph.D.GenerativeAIinPractice:ExploringUseCasestoHarnessEnterpriseDataCHECKLIST2024tdwi.org1  TDWIRESEARCHGenerativeAIinPractice:ExploringUseCasestoHarnessEnterpriseDataByFernHalper,Ph.D.FiveconsiderationsforgenerativeAI:UnderstandwhengenerativeAIistherightchoiceConsidercontentcreationusecasesfirstEvaluatechatbotsUsegenerativeAItoderiveinsightsDon’tforgetaboutback-enddeploymentissues12345Itisanexcitingtimeforartificialintelligence.GenerativeAI—asubsetofartificialintelligencethatinvolvessystemsdesignedtogenerateoutputssuchasimages,music,text,orotherformsofmediabasedonitstrainingdata—istopofmindformanyorganizations.GenerativeAIpromisestorevolutionizebusinessesbypotentiallyenablingunprecedentedlevelsofcreativity,efficiency,andpersonalizationincontentcreation,productdevelopment,andoperationalprocesses.InNovember2023,OpenAIannouncedthatitsgenerativeAIsystem,ChatGPT,had100millionweeklyactiveusers.1Itisnotsurprising,then,thatinTDWIsurveysthemajorityofrespondentsstatethattheyareeitherexperimentingwithorexploringgenerativeAI.1“OpenAIunveilspersonalizedAIappsasitseekstoexpanditsChatGPTconsumerbusiness,”https://www.reuters.com/technology/openai-enables-customized-gpt-bots-offers-cheaper-more-powerful-models-2023-11-06/tdwi.org2  TDWIRESEARCHTDWICHECKLISTREPORT:GENERATIVEAIINPRACTICE:EXPLORINGUSECASESTOHARNESSENTERPRISEDATAAtTDWI,weseeorganizationsexploringgenerativeAItounlocknewlevelsofproductivityusinglargelanguagemodels(LLMs)alongsiderelatedtools.InarecentTDWIsurvey,forinstance,thetopusecasesforgenerativeAIincludedcreatingchatbotsforcustomersupport(39%),generatingmarketingcontent(29%),onboardingnewemployees(26%),andactingasafrontendforanalyzingcompanydata(22%).2Alloftheseusecasesmayinvolveusingcompanydata,includingstructuredandunstructureddata.Forinstance,contentcreationorchatbotsmayutilizeunstructured,text-heavyassetssuchascorporateproductinformation,troubleshootingmanuals,websitecontent,andotherdocumentationthatholdsvaluablecompanyinformation;analyticsassistantsmayusestructuredandotherkindsoftabulardatatypicallystoredindatawarehouses,lakes,andotheranalyticsplatforms.Fororganizations,it’simportanttoconsiderusecasesthattargetallkindsofdatabecausewithallenterpriseshavingaccesstothesamemodels,companydataiswhatprovidestheircompetitiveadvantage.Ofcourse,generativeAIisn’tnecessarilythebestoptionforallAIusecases.Inthisever-evolvingenvironment,itisimportanttohavewell-definedbusinessobjectivesandthenevaluatethetechnologythatwillhelptheorganizationsuccessfullyachievethem.Aspartofimplementation,it’simportanttohaveaclearpathtoproduction,onefocusedonmaintainingdatasecurityandalowoperationalinfrastructureburden.ThisTDWIChecklistReportexploresseveralpopulargenerativeAIusecasesaswellashowtoapproachgettingstartedwithgenerativeAIandputtingitintoproductionwithcompanydata.UnderstandwhengenerativeAIistherightchoiceThereisoftenconfusionaroundtermssuchasAI,machinelearning,andgenerativeAI.AIisanumbrellatermthatdescribesabroadfieldofcomputersciencefocusedoncreatingsys