Ant Financial – Pioneering China Fintech with Machine Learning

Ant Financial is using machine learning to transform China's fintech industry.

Imagine yourself in an unfortunate fender bender – wouldn’t it be nice if you could just pull out your smartphone, take a photo, and have a damage assessment system powered by machine learning determine how serious the ding was and process the claim automatically with your insurer while simultaneously listing nearby repair shops and the approximate costs to fix the damage? That is one of the many ways that Ant Financial (Ant), the fintech affiliate of Alibaba, is using machine learning to transform personal finance in China [1]. Originally launched by Alibaba in 2014 to operate its ubiquitous online payment service Alipay, Ant now serves a network of over 600 million customers  [2].

Chinese consumers today use their smartphones to transact for just about anything, be it paying for taxi rides, meals, or even hospital bills. Given such prevalence in mobile payments and transactions, the ability to harness the vast amounts of data generated by its consumer base will be an invaluable competitive advantage for Ant, and that is why machine learning will play such an integral part in optimizing its existing businesses and shaping new ones. Looking forward, in order to sustain the strong momentum it has built in creating an open, shared credit system and financial services platform, applying machine learning to its product development will be at the core of the Company’s goal of upending traditional financial service offerings.

To Ant’s management, they understand that artificial intelligence is central to their business development, and that focus is manifested across the Company’s new and existing products and services. For example, it continues to invest in the “smart customer service” system that uses deep learning to carry on conversations with customers and answer their queries, and it has recently surpassed human performance in terms of customer satisfaction [3]. Earlier this year, it launched the “intelligent assistant”, a voice-recognition system that allows users on the Alipay platform to buy airline tickets and book hotels by using their voice, making travel booking more convenient and user friendly for millions of its customers. The Company’s also building advanced algorithms to analyze user data to target specific demographics. For example, Ant’s data scientists found that young girls who like tight jeans have their phone screens replaced more frequently. Therefore, it began selling screen insurance to those who fit that demographic [1].

In the medium term, management is beginning to focus on larger, more institutionalized organizations (compared to their current target audience of retail consumers and SMEs) as the digital transformation of these traditional organizations present tremendous business opportunities. For example, Ant recently announced that it will share its full suite of AI capabilities with 27 asset managers in China. Asset managers will be able to open corporate accounts on Ant’s AI-powered Caifuhao platform and leverage its vast AI infrastructure and customer data to gain better insights into consumers. Caifuhao thus creates an ecosystem that allows asset managers to better connect and engage with customers and provide valued-added services such as tailored wealth management products and investor education [4].

Given China lacks a formal credit system like that of the U.S., Ant has an incredible opportunity to leverage its infrastructure and customer data to establish a private social credit system. Considering that almost half of China’s population is users of Ant products and services, Ant can deploy its AI and machine learning capabilities to track and analyze consumer’s spending habits, payment history, and other personal information to algorithmically produce a consumer’s score. That score can then be used in a number of complementary ways to optimize its existing businesses, such as assessing a customer’s creditworthiness when granting spending credit on its Huabei platform, determining lending rates when making loans through its online bank MYbank, or even for figuring out whether to grant or deny certain perks and offers on Alipay [5]. This system is not only valuable in and of itself, but it can also be shared with traditional financial institutions to help them optimize their loan making process and risk management capabilities, thus increasing the efficiency and effectiveness of the financial industry as a whole.

As Ant looks to expand its presence internationally and especially in the U.S. given its attempted acquisition of MoneyGram, two important questions that merit further consideration is: 1) how will the Company add value to consumers in developed markets where existing AI and machine learning capabilities are so established and prevalent among leading technology and financial services companies, and 2) how can Ant acquire customers in those markets that already have robust digital payment systems (e.g. Apple Pay and Google Pay) and where most transactions are already cashless.

(Word count: 775 words)









Adding Value In Nike’s Production Line


Symantec: Using Machine Learning to Improve Malware Research

Student comments on Ant Financial – Pioneering China Fintech with Machine Learning

  1. I found the topic to be very interesting! Specifically, it opened my eyes to the huge opportunities Ant Financial has on the B2B space. By coupling Alibaba’s data on consumption with Ant Financial’s data on transactions, the company now has close to full visibility on consuming habits of a big part of the Chinese population. This vision is incredibly valuable to all consumer brands that are trying to reach their target market, and how well this data is used could be potentially the differentiation between Alibaba and Tencent on the e-commerce space.

    On your questions, I would argue that Ant Financial is probably looking to expand internationally through acquisition. By acquiring local companies’ userbase and adding new services, they would be able to create value in both developed and non-developed markets. In a non-developed digital transaction market for example, they could acquire a messaging or financial education app and transform that into a mobile wallet. On the other hand, in a developed market, they could acquire an already popular wallet and introduce new services such as loans and insurances. Either way, I believe it’s critical for Ant to use it’s capitalization to move quickly in attacking the market.

Leave a comment