Financial Tehnologies

Each payment could be defined as a transfer of funds that fulfills an obligation of the payer vis-à-vis the payee

New dynamics and innovative solutions are being observed in the evolving payments landscape, including:

• disruptive online payment services in the retail segment of the market (such as Alipay, Venmo/PayPal, Skrill, Dwolla, 2Checkout, and TransferWise);
• new regulations like Europe’s revised Payment Services Directive (PSD2) and sandbox approaches that provide a relaxed regulatory environment to test new services;
• new payment systems with improved user experience and faster availability of funds, including the Australian New Payments Platform and the United Kingdom’s Faster Payments Service; and
• new cyber-resilient payment systems for real-time settlement of payments between financial institutions, such as the TARGET Instant Payments Settlement (TIPS).

Wholesale payments are transfers of often high-value funds between financial institutions. Guaranteeing an orderly and timely settlement of wholesale payments is crucial for the wellbeing of financial markets. Wholesale payment systems, therefore, must ensure high maturity in information security, operational reliability, and cyber-resilience.

Retail payments are defined as payments between nonfinancial institutions. Their average value tends to be several orders of magnitude smaller than wholesale payments. By contrast, their share in the total number of payments is significantly higher. Operators of these systems are primarily focused on improving the user experience, the timeliness of processing, integration with other platforms, and ensuring adequate scaling.

Domestic payments are transactions between payers and payees in the same jurisdiction, whereas the geographic scope of cross-border payments covers multiple countries. As payment systems in the Euro zone are increasingly integrated with or replaced by single systems, payments within this zone could be considered domestic payments.

Safety requirements and efficiency standards formulated by international standard organizations like the Committee on Payments and Market Infrastructures (CPMI) are directly reflected in the technological blueprints of the next generation of wholesale payment systems.

Preventive controls are complemented by behavioral analysis techniques such as network intrusion detection and security information and event-management solutions, as well as intelligence-based penetration testing like CBEST.

Supervising authorities put forward highly ambitious incident recovery objectives, both in terms of recovery point and recovery time. Current cyber-resilience requirements stipulate that critical operations should resume within two hours of an incident.

Licensed nonbank payment services can only access wholesale systems through their main competitors — banks.

Global momentum has been observed for adopting the single open methodology prescribed in ISO 20022, a multipart international standard for electronic data interchange between financial institutions ( www.iso20022.org ).

Emerging user demands for wholesale payments primarily deal with users’ desire to optimize processing efficiency and interact with the system. Users increasingly demand immediate payment processing at any given time, but significant technical barriers exist.

Wholesale payment systems do not operate in a vacuum—they are highly interconnected with the systems of their users and other third parties. … Cyber-events such as the 2016 compromise of the Bank of Bangladesh’s information systems demonstrated that a compromise might crystallize at any organization or step in the payment processing chain and then impact the entire ecosystem. Risk assessments increasingly cover risks related to connected third-party infrastructure.

Retail payment innovation emphasizes faster processing … aim for always-on systems that provide near-real-time (typically less than 10 seconds) availability of funds to the beneficiary or a rejection of the payment.

Physical items are increasingly being turned into smart objects that are able to interact with other objects in their environment (the Internet of Things, or IoT). Automatic payment for services by these smart objects is an often-recited interaction pattern, such as cars automatically paying an insurance premium in a pay-as-you-go model or in pay-per-use leasing models.

Payment initiation services like SOFORT, iDeal, and Trustly—which establish software bridges between e-commerce websites and online banking platforms—are examples of participants that benefit from the revised regulatory framework.

Mobile wallets contain tokenized versions of credit cards and allow for payments through the use of near-field-communication (NFC) chips and QR codes.

Mobile network operators offer alternative payment systems based on carrier billing systems like Boku or deposits with the carrier, as is the case for M-Pesa in Africa. Mobile carrier–backed payment systems like M-Pesa have significantly improved financial inclusion in developing countries. The need for mobile networks with extensive coverage is crucial in this context and an important technical challenge for the operators of these systems.

Nonbank service providers that recently entered the retail payment market are following a similar strategy. For example, Alipay and WePay have distilled typical usage scenarios for their services and have started to develop an extensive partner network. Currently, the usage scenarios are strongly focused on online-to-offline models, which combine online customer acquisition with physical delivery of services and products. Examples of these scenarios are paying for utility bills, online shopping, topping off mobile phone credit, and buying train tickets.

Advanced statistical analysis of payment data will enable more effective risk and compliance assessments.

Payment-related data can be used to continuously improve the user experience, which can deliver a competitive advantage in an environment with increasingly demanding end users.

Payment service providers can try to monetize the data. Conducting advanced market analyses on payment data can provide valuable insights into the market segments and assist platform partners in driving their reward programs and identifying cross-selling opportunities.

A prime driver for the current uncertainty regarding the timing and cost of international transactions is the omnipresence of correspondent banking. Correspondent banking is the practice of using one or more intermediaries between banks in different countries. This is typically needed if the payer’s bank has limited access to the foreign financial market in which the payee is located. Currency conversion is often required.

Ant Financial has redefined digital financial services, specifically mobile payment and microloan services

Although the initial technological challenge was expected to be connecting all banks, it turned out that the real challenge was controlling risk given a rapid increase in transaction volume.

A leap forward occurred when the system adopted advanced machine learning methods—including boosting, deep learning, and graph-based machine learning—for accurate credit modeling. The new system is characterized by three digits: 3, 1, and 0; a merchant takes less than 3 minutes to complete a loan application, obtains the decision in 1 second, with zero human intervention. The integration of systems engineering and algorithmic advances into the microloan operation makes the cost of each loan less than 2 RMB ($0.31 USD). Combining a convenient user experience with low operational costs, Ant Financial now serves tens of millions of merchants in China with accessible and affordable loans.

Financial service providers face three challenges when digitizing service for the future economy

1. connection: how to link users, merchants, and service partners in a low-cost, fast, and intelligent way;
2. risk: how to control for aspects of financial risk; and
3. trust: how to grant equal opportunity for all to be trusted, and trustworthy, in the digital space.

Ant Financial focuses on five technologies: blockchain, AI, security, IoT, and computing (BASIC). Blockchain helps to build a trusted global inter-connected system capable of storing, exchanging, and processing values; AI enables companies to build intelligent systems that better serve customers and business partners, and drive new product design; security is a pillar that makes digital systems safe and stable; IoT (Internet of Things) is a bridge that connects the physical and digital realms to transformative effect; and computing engines provide the digital space with computational power.

Ant Financial uses AI to create a financial brain for the digital world. Recent years have witnessed the huge success of machine learning and deep learning in machine perception areas such as speech recognition and image analysis, but financial services need more, including prediction and decision-making. These capabilities, combined with a comprehensive financial knowledge graph, are the foundation of the financial brain at the core of Ant Financial's risk, credit, and customer service engines. The brain enabled Ant Financial to reduce its payment loss rate to less than one in a million, automatically answer millions of customer inquiries a day, automatically assess car damages based on computer vision and a vehicle knowledge base, and improve other services.

Deep learning and natural language processing (NLP) technologies helped intelligent customer service robots achieve higher customer satisfaction rates than live service staffs. During the popular Singles' Day 2016 shopping occasion, 97% of customer service inquiries on Ant Financial's Alipay service were handled by the intelligent customer service robots.

Ping An developed a system where the customer only needs to take a few pictures of the damaged car to file a claim from the accident site. The claim is processed within seconds and the customer given a precise payment calculation. The system involves a series of key modules: picture quality assessment, verification of insurance, car segmentation, identification of damage and related parts, payment calculation, and fraud detection. A number of AI techniques, such as image processing, image segmentation, and object recognition, were developed to support the functions. The system has been running in production at Ping An for over a year, successfully processing over 30,000 claims each day.

There are over 70-million registered enterprises, including households, in China. Their information comes from three major sources: commercial registration and daily operation; public news announcements and social posts; and business relationships including the supply chain, investments, and legal actions. To organize and analyze such rich and dynamic data, Ping An developed Euler Graph, an enterprise knowledge graph. The graph covers nearly all of China's 70-million enterprises, using data from all three sources. Millions of legal proceedings are automatically interpreted and over 40-million lawsuit relationships have been extracted and incorporated into the graph. Signals on enterprises are collected from over 300 news and social sites, totaling hundreds of thousands of articles daily, and updated every 10 minutes.

Dynamics of networked market systems are better under-stood as complex adaptive systems, in which independent software components interact without centralized control or oversight.