AML交易监控与新技术
由原文自动翻译 · 阅读原文 (English)
一直在研究AML交易监控,特别是关于公司如何适应各种新平台上交易的速度和数量。似乎传统的基于规则的系统难以跟上,导致大量误报,或者更糟的是,错过了细微的模式。你们中是否很多人看到在这方面对AI/ML解决方案进行了大量投资,还是更多地是关于改进现有工具和数据输入?将这些新技术整合到实际风险降低中,而不是仅仅增加复杂性,实践经验如何?
由原文自动翻译 · 阅读原文 (English)
一直在研究AML交易监控,特别是关于公司如何适应各种新平台上交易的速度和数量。似乎传统的基于规则的系统难以跟上,导致大量误报,或者更糟的是,错过了细微的模式。你们中是否很多人看到在这方面对AI/ML解决方案进行了大量投资,还是更多地是关于改进现有工具和数据输入?将这些新技术整合到实际风险降低中,而不是仅仅增加复杂性,实践经验如何?
That's a really interesting point. I've been wondering how these older systems are coping with the sheer volume and speed of modern transactions, especially with crypto. Are these AI/ML solutions already proving more effective in reducing false positives, or is it still early days for widespread adoption?
That's a really good point about the traditional systems. I've definitely seen some of the larger institutions making noise about AI/ML for anomaly detection, but the implementation seems to be slow-going due to regulatory hurdles and data privacy concerns. Are smaller firms having an easier time adopting these newer technologies?
From what I've seen, many firms are certainly discussing AI/ML for AML, but actual significant investment and successful large-scale implementation seem to be moving at a much slower pace than the hype suggests. The regulatory hurdles and data quality issues are substantial.
Traderforum · 简体中文