当前位置:首页 > 报告详情

代理式 KYC:金融机构的自主客户筛选 [重复].pdf

上传人: 明**** 编号:1013175 2025-12-21 40页 772.82KB

1、 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.I N D 3 7 3Agentic KYC:Autonomous customer screening for financial institutionsChetan SharmaPrincipal Solutions Architect,FSIAWSYaniv GuetaPrincipal Solutions Archit

2、ect,FSIAWS 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.AI maturity scale from rules to reasoningLow agencyMore human oversightHigh agencyLess human oversightRule-based RPAFollow a set of rulesGenerative AI assistantsAchieve a specific,predefined goalGoal-driven AI agentsOperat

3、e toward a higher-level objective,not just a taskFully autonomous agentic systemsIndependently set and execute goalsFollowAssistCollaboratePioneerINCREASING AUTONOMY AND BUSINESS IMPACT 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.AgendaBusiness Challenge&OpportunityDemo:Agenti

4、c KYCArchitecture Deep DiveImplementation ConsiderationsQ&A 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.KYC process-Customer PerspectiveJohn visits bank website Fills application form manually Uploads documents:ID,payslips,statements Submits and waits.Receives Under Review ema

5、il And wait.for 2-5 days for approvalHe wants to open a new bank accountMeet John 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.KYC process Bank Perspective Receives Johns case in queue Meet Shirley KYC AnalystManually reviews each document Extracts data points to spreadsheet Ch

6、ecks sanctions lists manually Calculates risk score using templates Writes assessment report Mail to manager andwait for approval 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.High Error Rates and Compliance RisksKey Business ChallengesRegulatory Complexity ManagementExtended On

word格式文档无特别注明外均可编辑修改,预览文件经过压缩,下载原文更清晰!
三个皮匠报告文库所有资源均是客户上传分享,仅供网友学习交流,未经上传用户书面授权,请勿作商用。
根据报告的内容,全文主要介绍了Agentic KYC(客户身份验证)系统,该系统通过自动化流程提高金融机构的KYC效率。以下是关键点: 1. **AI成熟度**:从基于规则的RPA到具有推理能力的AI助手,再到完全自主的智能系统,AI在KYC领域的应用逐渐成熟。 2. **KYC流程挑战**:传统KYC流程存在错误率高、合规风险、管理复杂、开户时间长、数据隐私和安全问题、成本高、资源负担重以及欺诈风险等问题。 3. **Agentic KYC架构**:系统通过KYC Orchestrator协调多个专门代理(如文档代理、制裁筛查代理、风险评估代理等)自动处理客户申请。 4. **自动化处理**:系统自动提取文档数据、检查制裁名单、分析媒体信息、计算风险评分,并生成评估报告。 5. **技术栈**:使用Amazon Bedrock框架,包括Strands Agents、CrewAI、LangGraph等,实现多代理协作和自动化流程。 6. **实施考虑**:涉及多代理模式、单代理协调器、网状群组、分层代理图等,并强调可解释性在监管环境中的重要性。 7. **AWS Responsible AI维度**:包括可控性、隐私与安全、安全性、公平性、真实性、可解释性、透明度和治理。
"AI加速KYC,效率翻倍?" 自动化KYC如何降低风险?" AI赋能KYC流程!"
客服
商务合作
小程序
服务号
折叠