[Netease smart news April 26 news] From ATMs and mobile wallets to chat robots and robot consultants, technology has mercilessly changed how we store, transfer and manage money. Capital One is at the forefront of this transformation – one of the largest banking institutions in the United States, and one of the first banking institutions in the United States to seriously invest in digital technology and artificial intelligence. Several large financial services companies have publicly announced ambitious plans for AI, including consumer-oriented chat bots. However, after several months, they still did not ship. Companies that are not core technology organizations such as Google or Facebook often have difficulty launching technology products. Capital One is one of the few exceptions to this rule. When the company set up an internal software engineering department and transferred a lot of business to the cloud, Capital One's transformation into digital business began. Under such circumstances, it is almost foreseeable that agility will become a key competitive advantage for the financial industry to participate in technological competition. Recently, Fortune 500 companies injected back-end infrastructure and customer-facing channels into smart machine learning and natural language processing. In March of this year, a company launched the first gender-neutral, nlp-enabled chat robot Eno ( Later in this article are called.) In 2016, Virginia-based lenders became the first financial services company to launch customer accounts on Amazon's Alexa platform, allowing users to check their balances, pay bills, and conduct extensive voice-based interactions. The company also announced plans to become the first financial services company to launch similar services on Cortana, Microsoft's personal assistant software. Capital One's latest artificial intelligence drive project is the chat robot Reynolds, which dramatically transforms the media from speech to text. Ken Dodelin, vice president of digital product management, explained: “SMS is the most widely used feature of smartphones.†97% of smartphone users send text messages. So think that this is a benefit to us. We launched the first natural language SMS chat robot from a U.S. bank. “With Inno, people can use natural language to talk with us about their credit card accounts and checking accounts. We can talk with our customers in natural language.†Many chat robots are friendly, charming and harmless, but some are used by phishing scammers. Hijack sensitive information. As a highly regulated business, Capital One understands the risks of sharing data with artificial intelligence and establishes specific security protocols to protect customers' services using artificial intelligence. Based on the web platform, Capital One has taken a different approach to protecting personal information. For example, in order to facilitate the use of new Alexa technology, users first need to log on to a website in the capital using their secure username and password. Customers can also set a personal key for Alexa with a four-digit password. This prevents unauthorized individuals from instructing Alexa to disclose sensitive information about other users. At the same time, the company's artificial intelligence developers set a clear limit for Eno's use case: the system allows bills to be paid but largely limits the interaction between consumers and system searches. Margaret Meyer, vice president of software engineering, stated that many of the design investments were made by the customer themselves when building an artificially oriented solution. For example, the "second look" feature of the "capitalized" wallet application uses the machine's logging feature to actively alert customers when they encounter suspected unusual transactions, such as double charges and other non-standard patterns. "We usually use analytics to find out what models customers use, whether they are spending patterns or concerns about fraud." We are studying machine learning in many ways, adds Meyer. Steph Hay, head of artificial intelligence design, said that the biggest challenge for artificial intelligence design is the environment. Hay believes that the environment is the intersection between design and technology. She asked, "What can I learn from a conversation, how can I respond to your intentions and emotional state? Is this isolated in this dialogue, or can we look at the previous conversation? We can see Looking beyond the behavior of other channels? In addition, how do we begin to make better assumptions and provide smarter answers based on each person's data, not just your data?" For Hayworth, the challenge seems to be the emergence of a "sustained pendulum swing" in the use of the company's vast data set and microscopic level to understand the interaction between a person and a bank. Although Capital One is one of the most tech-savvy companies in the banking industry, only a small portion of its 43,000 employees are directly involved in the technology industry. In fact, they only have less than 50 people to build and distribute Eno. One of the key principles of the binding technical team is the agile development framework. "The core design, product, and technology teams work together." This makes conversations "Hey, I don't understand. Please help me understand. These teams are actually learning together," Meyer explained. (Source: VB Business Channel Compilation: Machine Xiaoyi Review: Zhou Wensheng) 燑br>
