最有看点的互联网金融门户

其他国际资讯

如何保证AI在金融科技领域应用的透明性和可靠性?

其他国际资讯

如何保证AI在金融科技领域应用的透明性和可靠性?

How Do We Ensure Transparent and Trustworthy AI in Fintech?

随着数据流量的日益增大,人工智能系统也变得越来越智能,但与此同时这些系统是否也变得更加有道德了呢?

只要人们在设置人工智能系统的时候抱有正确的价值观,那么这些系统就将在改造现有的金融科技上拥有无限可能性。

IEEE网站一篇名为《将道德融入设计中》(Ethically Aligned Design)的文章表明, 人工智能系统的道德设置甚至比金融合规监管还要复杂,因为这些系统目前只是试探性地应用于法规中。

报告指出:

"对于如何将人类的规定和道德观念传输到自动化人工智能系统中,社会还没有一个普遍承认的标准。但是由于这些系统通过自动化程序做决定,而且其控制环境的能力在日益增强,让它们去采纳、学习以及遵守其服务社区的规定和道德标准也就变得非常必要了。根据它们的不同功能以及服务的不同人群,它们也应该尽可能以一种透明和诚实的方式交流及解释自身的行为。"

这篇文章还强调了在设计人工智能系统时要考虑道德因素的重要性。文章中将科技学家定义为参与到调研、设计、制造人工智能系统或者为其传输信息的所有人和机构,其中包括大学、组织及大型公司等。文章认为,对这些人的教育是至关重要的。

AI-Austin执行理事,IEEE全球人工智能系统道德设计规划组副主席Kay 表示:

"科技学家为整个金融科技市场做出判断。而这些人的确有能力运用这些道德原则,但是我们仍然需要对人工智能领域的工作人员给与更多的教育。我们所谈论的是人类的幸福——我们需要鼓励金融科技行业来思考这个问题。"

Kay 还表示,将人工智能道德设计初步融入金融科技的最好方式就是在各个人工智能领域的公司设立一个首席道德官的职位。

关于人工智能在金融科技领域的应用,有关企业已经开始探索将这项技术应用于保护数据、身份隐私,遵守金融法规等领域,为客户更加透明的服务。如果这些功能对你来说都很耳熟,那很有可能是因为它们对于金融机构来说很有吸引力。这些科技包括智能合同,这些合同拥有自动执行代码,这完全符合人们对人工智能的定义。

Fay 说:

"在我看来人工智能系统将会管理智能合同,而由人工智能管辖的全套系统将会在未来更加成熟完整。"

人们必须细心妥善管理这个系统,避免例如DAO黑客之类的危害,这意味着在建立这个系统的时候就应该考虑到消费者的需求。对于这一点,金融科技是特别擅长的。而现在,在设计人工智能系统的最初阶段,人们也应该将消费者的需求考虑进去。

AI systems may be getting smarter with the ever-expanding flow of data, but are they also getting more ethical?

When it comes to fintech, artificial intelligence has a seemingly endless range of possibilities to transform present systems, so long as the AI system that is applied is constructed with the correct values in mind.

Explored in the IEEE Ethically Aligned Design paper on the topic, the ethics of AI design is as complex as financial compliance regulations—perhaps more so, as some of these systems are at the point now where they are being tentatively applied to those regulations.

From the report:

Society does not have universal standards or guidelines to help embed human norms or moral values into autonomous intelligent systems (AIS) today. But as these systems grow to have increasing autonomy to make decisions and manipulate their environment, it is essential they be designed to adopt, learn, and follow the norms and values of the community they serve, and to communicate and explain their actions in as transparent and trustworthy manner possible, given the scenarios in which they function and the humans who use them.

The report cites a need for ethical consideration within the actual design of the AI system, making education of “technologists”—which are, as defined by the paper, “anyone involved in the research, design, manufacture or messaging around AI/AS including universities, organizations, and corporations—critical.

“If we’re talking about technologists, who are making decisions for the whole fintech market—we do see that they are able to use these [ethical] principles—then we do need to do more education with the people who are working within the design of the AI industry,” says Kay Firth-Butterfield, executive director for AI-Austin and executive committee vice-chair for The IEEE Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems. “We talk about human wellness—we need to encourage fintech industrialists to think about that.”

According to Firth-Butterfield, one of the best ways to begin knitting ethical AI design into the fabric of fintech could be the creation of a “chief values officer” for every company working on artificial intelligence.

For use within fintech, companies are exploring the technology for usage in factors such as data privacy and identity, and adherence to financial regulation for more transparency to the consumer; if that sounds familiar that’s probably because these are the use cases that tend to make other technologies so attractive to financial institutions—technology like, oh, say, smart contracts, which as self-executing code fits the definition of AI.

“I see smart contracts as part of the system that AI will manage… as part of a system that will be very integrated in the future,” says Firth-Butterfield.

This integrated system will have to be managed quite carefully to avoid exploitations like the DAO hack, which means the system will have to be built with the needs of the consumer in mind. Luckily, this is an approach fintech particularly excels in. Now that approach just needs to be wired in to the bare-bones design of the technology.

GELP

用微信扫描可以分享至好友和朋友圈

发表评论

发表评论

评论

您的个人信息不会被公开,请放心填写! 标记为的是必填项

4 × 二 =

取消

2017第二届上海国际金融科技创新大会

小未 19小时前

SoFi收购Zenbanx对fintech的未来意味着什么

潘, 妍媛 1天前

Investopedia十大金融关键词:Trump、Fintech表现抢眼

张沛祺 2天前

韩国政府放松对金融科技、AI和VR领域监管

高旭 | ASIA-FIRST 2天前

知名Fintech专家Catherine Bessant将加盟苏黎世保险集团

张沛祺 02-18