Artificial Intelligence in Finance: Putting the Human in the Loop
Artificial Intelligence in Finance: Putting the Human in the Loop
复制标题
金融领域的人工智能：让人类参与进来
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Brian W. Tang
中科院分区：
文献类型：
--
作者：
D. Zetzsche;D. Arner;Ross P. Buckley;Brian W. Tang
金融已成为经济中最全球化和数字化的部门之一。它也是最受监管的行业之一，特别是自2008年全球金融危机以来。全球化、数字化和货币正在推动金融领域的人工智能以越来越快的速度向前发展。
本文制定了一个监管路线图，以理解和解决人工智能在金融中日益重要的作用，重点是人类责任：“将人类置于循环中”的想法，特别是为了解决“黑匣子”问题。
第一部分描绘了人工智能在金融领域的各种用例，强调了为什么人工智能在金融领域发展如此迅速，并将继续发展下去。第二部分重点介绍了由于人工智能在金融领域的发展而可能出现的一系列潜在问题。第三部分考虑了人工智能在金融服务领域的监管挑战以及解决这些挑战的工具，第四部分强调了人类参与的必要性。
我们发现，人工智能在金融领域的使用带来了三个监管挑战：（1）人工智能增加了用户、开发人员、监管机构和消费者之间关于算法能力和效果的信息不对称;（2）人工智能增强了数据依赖性，因为不同的数据源可能会改变运营、效果和影响;（3）人工智能增强了相互依赖性，因为系统可以与意想不到的结果相互作用，增强或减弱有效性，影响力和可解释性。这些问题通常被总结为“黑箱”问题：没有人理解某些人工智能是如何运作的，或者为什么它做了它所做的事情，这使得问责制变得不可能。
即使监管机构拥有无限的资源和专业知识-他们显然没有-通过传统手段监管人工智能的影响是具有挑战性的。
为了应对这一挑战，我们主张通过外部监管加强受监管金融市场参与者的内部治理。因此，第四部分建议，最有效的前进道路是将人纳入循环的监管办法，通过外部监管加强内部治理。
在金融领域，危机后对个人和管理责任体系的关注提供了一个独特而重要的外部框架，通过监管责任将人纳入循环，在某些情况下通过人工智能审查小组来加强内部责任。我们认为，这种方法--人工智能定制的经理人责任框架，在某些情况下由独立的人工智能审查委员会加强，作为对传统三道防线的增强--可能是解决人工智能相关问题的最有效手段，不仅是在金融领域--特别是“黑匣子”问题--而且可能是在任何受监管的行业。
Finance has become one of the most globalized and digitized sectors of the economy. It is also one of the most regulated of sectors, especially since the 2008 Global Financial Crisis. Globalization, digitization and money are propelling AI in finance forward at an ever increasing pace.
This paper develops a regulatory roadmap for understanding and addressing the increasing role of AI in finance, focusing on human responsibility: the idea of “putting the human in the loop” in order in particular to address “black box” issues.
Part I maps the various use-cases of AI in finance, highlighting why AI has developed so rapidly in finance and is set to continue to do so. Part II then highlights the range of the potential issues which may arise as a result of the growth of AI in finance. Part III considers the regulatory challenges of AI in the context of financial services and the tools available to address them, and Part IV highlights the necessity of human involvement.
We find that the use of AI in finance comes with three regulatory challenges: (1) AI increases information asymmetries regarding the capabilities and effects of algorithms between users, developers, regulators and consumers; (2) AI enhances data dependencies as different day’s data sources may may alter operations, effects and impact; and (3) AI enhances interdependency, in that systems can interact with unexpected consequences, enhancing or diminishing effectiveness, impact and explainability. These issues are often summarized as the “black box” problem: no one understands how some AI operates or why it has done what it has done, rendering accountability impossible.
Even if regulatory authorities possessed unlimited resources and expertise – which they clearly do not – regulating the impact of AI by traditional means is challenging.
To address this challenge, we argue for strengthening the internal governance of regulated financial market participants through external regulation. Part IV thus suggests that the most effective path forward involves regulatory approaches which bring the human into the loop, enhancing internal governance through external regulation.
In the context of finance, the post-Crisis focus on personal and managerial responsibility systems provide a unique and important external framework to enhance internal responsibility in the context of AI, by putting a human in the loop through regulatory responsibility, augmented in some cases with AI review panels. This approach – AI-tailored manager responsibility frameworks, augmented in some cases by independent AI review committees, as enhancements to the traditional three lines of defence – is in our view likely to be the most effective means for addressing AI-related issues not only in finance – particularly “black box” problems – but potentially in any regulated industry.