Purpose

The purpose of this paper is to examine the artificial intelligence (AI) methodologies to fight against money laundering crimes in Colombia.

Design/methodology/approach

This paper examines Colombian money laundering situations with some methodologies of network science to apply AI tools.

Findings

This paper identifies the suspicious operations with AI methodologies, which are not common by number, quantity or characteristics within the economic or financial system and normal practices of companies or industries.

Research limitations/implications

Access to financial institutions’ data was the most difficult element for research because affect the implementation of a set of different algorithms and network science methodologies.

Practical implications

This paper tries to reduce the social and economic implications from money laundering (ML) that result from illegal activities and different crimes against inhabitants, governments, public resources and financial systems.

Social implications

This paper proposes a software architecture methodology to fight against ML and financial crime networks in Colombia which are common in different countries. These methodologies complement legal structure and regulatory framework.

Originality/value

The contribution of this paper is how within the flow already regulated by financial institutions to manage the ML risk, AI can be used to minimize and identify this kind of risk. For this reason, the authors propose to use the graph analysis methodology for monitoring and identifying the behavior of different ML patterns with machine learning techniques and network science methodologies. These methodologies complement legal structure and regulatory framework.

中文翻译：

---

人工智能打击洗钱网络：哥伦比亚案

目的

本文的目的是研究打击哥伦比亚洗钱犯罪的人工智能方法。

设计/方法/方法

本文使用网络科学的一些方法论研究哥伦比亚的洗钱情况，以应用AI工具。

发现

本文使用AI方法论来识别可疑操作，这些方法在数量或数量或特征上在经济或金融体系内以及公司或行业的正常实践中并不常见。

研究局限/意义

访问金融机构的数据是研究中最困难的因素，因为它会影响一套不同算法和网络科学方法的实施。

实际影响

本文试图减少因非法活动以及针对居民，政府，公共资源和金融系统的各种犯罪而造成的洗钱活动对社会和经济的影响。

社会影响

本文提出了一种软件体系结构方法论，以打击哥伦比亚的ML和金融犯罪网络，这在不同国家都很普遍。这些方法补充了法律结构和监管框架。

创意/价值

本文的贡献在于如何在金融机构已经规定的流程中管理ML风险，如何使用AI来最小化和识别此类风险。因此，作者建议使用图分析方法通过机器学习技术和网络科学方法来监视和识别不同ML模式的行为。这些方法补充了法律结构和监管框架。