Skip to content

GRA · Fintech Nexus

空间金融资本与危害、风险暴露和转型的交汇之处

金融身份

Every financial relationship starts with identifying the customer: proving who they are, checking them against sanctions and risk lists, and authenticating them each time they return. Digital onboarding has widened access and cut cost, but it has also industrialised identity fraud, from synthetic identities to deepfake-assisted account opening. Checks that are too rigid exclude people who lack documents. This page sets out how identity risk develops in finance, how it is measured, and who sets the rules.

01登记

Financial identity on the record.

02问题

财务恒等式:风险如何形成。

客户开户、欺诈与排斥。下方档案阐述风险如何发展、衡量什么、数据来自何处、如何建模以及由谁决定。

机制

Synthetic identities age into credit

欺诈者将真实信息与虚构信息组合成可通过开户审核的身份,经过数月积累良好记录,然后提取信贷并消失。没有真实受害者投诉,因此损失显现较晚,且常被记为违约。

机制

深度伪造攻击远程身份验证

人脸比对和活体检测面临注入式视频、生成的文件和克隆的语音。攻击检测的每一次改进都会遇到新的工具,因此,除非持续测试,身份核验的保证水平会逐渐下降。

机制

Mule accounts carry proceeds

诈骗和欺诈所得资金经由被招募或被冒用身份者开设的账户流转。开户审核薄弱、监测迟缓,使骡账网络得以在冻结生效之前跨机构转移资金。

机制

刚性检查会排除

Document requirements that low-income, migrant, displaced or rural people cannot meet push them out of formal finance. Tiered, risk-based due diligence and trusted digital ID can widen access without lowering integrity.

03风险暴露、损失与资本

金融身份对风险暴露、损失和资本意味着什么。

风险官、监管者或财政部会要求的风险暴露、损失、情景和资本视图,以及它们据以报告的框架。结构遵循监管和精算实务。数值为说明性:绝非价格、评级或建议。

传导渠道

How a shock in financial identity reaches credit, market, underwriting, liquidity and operational risk, and feeds back.

Channel structure after the Basel Committee’s work on climate-related risk drivers and transmission channels; links shown are typical paths, not estimated exposures.
NGFS 气候情景 · 央行与监管机构绿色金融网络IFRS S2 气候相关披露 · ISSBTCFD 建议 · 金融稳定理事会 TCFD有效管理气候相关金融风险的原则 · 巴塞尔银行监管委员会
04指标

衡量什么,以及衡量得如何。

每个指标都注明其单位、频率、背后的来源类型以及数值将附带的证据标签。此处不显示数值:数值由计划依据其自身权限提供。

指标单位节奏来源类型标签
Identity fraud at onboarding每千份申请中的尝试次数和确认案件数每月Firm fraud records观测
账户接管事件cases per million active accounts每季度Firm and industry fraud reporting已报告
呈现攻击检测错误率攻击与真实样本分类错误率每次评估Independent biometric testing观测
Remote onboarding completion and abandonmentpercent of started applications每月企业记录观测
因缺乏证件而未开立账户的成年人没有账户的成年人百分比每轮调查Global Findex估算
拥有基础身份证明的人口占人口百分比定期ID4D 数据集和国家登记册估算
与身份欺诈相关的可疑交易报告每年计数年度金融情报机构报告已发布统计
05来源

证据来自何处。

具名公共数据集与机构记录,各自注明许可类别。依协议获得的来源仅按其条款使用,绝不再次发布。

来源它提供什么许可类别
世界银行全球金融包容性数据库(World Bank Global Findex Database)Account ownership and barriers to access, including lack of documentation, by country and groupFree and open, attribution (CC BY)
世界银行 ID4D 全球数据集Coverage of foundational identification systems by country免费开放,须署名
NIST 人脸识别和呈现攻击评估生物识别算法的独立准确性和攻击检测结果公有领域
FBI Internet Crime Complaint Center reportsReported identity theft, impersonation and fraud losses公有领域
Financial intelligence unit annual reports可疑交易报告的趋势和类型Public records, terms per unit
FATF guidance on digital identity and mutual evaluation reportsAssessments of customer due diligence regimes and guidance on reliance on digital ID免费,条款依 FATF
06决定

由谁决定,以及证据服务于什么。

Fintech Nexus 负责准备证据。以下决定由所列机构依其自身规则作出,平台绝不代为决定。

谁来决定问题本页面服务的证据
金融情报机构和反洗钱监管机构哪些客户尽职调查方法符合规则,以及在哪些情形下允许采取简化措施Assurance-level mapping, typology notes, inclusion evidence
National identity authorities金融机构能否以及如何依赖国家数字身份办理客户开户保证映射、覆盖缺口、欺诈与错误记录
Bank and fintech boards应资助哪些身份证明、活体检测和监测控制措施,以及可接受多大的客户摩擦开户环节的欺诈、攻击检测结果、放弃率
标准机构(FATF、ISO、NIST)哪些身份保证和认证级别适用于金融场景独立测试结果、事件类型划分、跨司法管辖区比较

该主题产生的产出

  • Onboarding risk guides for remote and digital channels
  • 关于合成身份、骡子账户和 深度伪造
  • 关于分层尽职调查与包容性的说明
  • 国家数字身份保证等级与金融开户的对照
07核心问题

专家提出的问题。

  • 数字化开户和验证方法如何在安全、便利与包容之间取得平衡?
  • 合成身份和账户接管是如何产生的,又如何被检测?
  • Who is excluded by identity requirements, and why?
  • 数字身份 系统 如何保护隐私?
  • How does risk-based customer due diligence work in practice?
08机制

Verification, fraud and exclusion.

Know-your-customer requirements oblige financial institutions to identify and verify customers and assess the risk they pose. Digital onboarding uses document scanning, biometric matching, database checks and device signals. Strength of verification varies, and attackers exploit the weakest methods. Synthetic identity fraud combines real and fabricated information to create identities that pass checks; account takeover uses stolen credentials or social engineering; and generative AI enables convincing fake documents and voice or video impersonation.

Exclusion is the other side of the problem. Many people lack formal identity documents or proof of address, and rigid requirements can deny them access to accounts, credit and payments. Risk-based approaches allow simplified due diligence for lower-risk products, and national digital identity systems can widen access, but they raise privacy and surveillance concerns and can exclude those not enrolled.

Identity is also a systemic dependency: when a widely used identity provider fails or is compromised, onboarding and authentication can stop across many institutions.

Recovery of identity after compromise is an under-studied problem. Victims of identity theft may spend long periods restoring access and credit standing, and systems designed for onboarding often lack equally strong processes for re-establishing a person’s identity.

09方法与数据

方法及其局限。

Fraud

Fraud pattern analysis

Studying attack methods and detection; data are fragmented across firms.

Inclusion

访问研究成果

Measuring who is excluded and why; under-documented people are hard to reach in surveys.

隐私

Privacy engineering

评估选择性披露等保护隐私的设计。

局限:欺诈数据较为敏感且定义不一致,身份规则对普惠性的影响也难以与其他障碍区分开来。

10方法

如何研究,以及方法无法说明的内容。

  1. 数字身份与公共数字系统保证等级映射 · 远程身份验证攻击测试 · 排斥分析在此应用
  2. 人工智能与机器学习信用与欺诈模型评估 · 漂移与挑战者模型监测在此应用
  3. 网络安全与密码学威胁导向渗透测试 · 供应链与依赖分析在此应用
  4. 支付与金融市场技术支付链条依赖关系映射 · 流动性压力模拟在此应用
11问题

问题与解答。

What is risk-based customer due diligence?
FATF 建议下的一种方法,其中核查措施与客户和产品的风险相称。风险较低的产品可采用简化措施,这有利于普惠。
What is a synthetic identity?
由真实与虚构信息混合构成的身份,例如真实标识符搭配虚假姓名。没有任何真实的人拥有完整身份,因此无人察觉并报告其被滥用。
数字身份能否同时减少欺诈和排斥?
It can when the system is inclusive, secure and governed with privacy safeguards. A poorly designed system can exclude people and concentrate fraud risk in a single credential.
Fintech Nexus 是否核验身份?
不。它旨在促进行业对身份风险和身份保证形成共同认识。身份核验由受监管机构依照其监管者制定的规则进行。
GRA 是否核实身份?
No. GRA does not provide identity, onboarding or compliance services.
Can identity rules be both strict and inclusive?
Risk-based approaches aim to do this, applying stronger checks where risk is higher and simpler ones for basic products.
生物特征验证安全吗?
它可能很强大,但会引发隐私问题,也可能被欺骗;能否使用取决于设计和保障措施。
报告贡献

从本页贡献证据。

Reports 是联合体的季刊。您对该主题的了解可以成为下一期中的证据、案例或更正,署名归您,并在属于您本人作品时以 CC BY 4.0 发布。

Global Risks AllianceFintech NexusExponential TechnologyFinance and Insurance Readiness
Fintech Nexus

向联盟提出风险暴露或资本问题。

信息和咨询现已开放。每项咨询都会获得一个编号,并告知下一步;申请、付款、算力分配和注册只有在相应团队、条款和审查能力就绪时才会开放。