AI is becoming invaluable to the payments industry

Innovation in AI (artificial intelligence) technologies has simplified the use of ML (machine learning). It is becoming a valuable tool for improving the decision making process as the outcome is more accurate and faster.

Platforms such as IBM Watson, Azure Machine Learning or Google Prediction APIs have enhanced the adoption of AI technology in several areas as they are easy to use even for users who don’t have a deeper understanding of complex AI algorithms. Today, programming languages such as Python and R have in-built libraries with ML algorithms that have made BOT development a lot easier.

Following are some good use cases for AI in the payments industry.

Fraud management

Most financial institutions are using AI to prevent fraud before it takes place. This directly helps reduce chargebacks and false positives. With fraudsters continuously finding new ways to do fraudulent transactions, a traditional rule-based system may fail to identify such transactions. This is further complicated by the advent of number of new payment channels, which increases the fraud risk further in payments.

Since building additional rules and new patterns every time is cumbersome and costly, AI is considered to be the best alternative technology to seamlessly address issues as follows:

• Ability to interpret data and gain insights through supervised machine learning and also gain new knowledge through unsupervised learning.
• Increased automation and improved accuracy, leading to fewer manual checks.

Chatbots

Chatbots powered by AI are rapidly becoming an integral part of customer relationship management, cross-selling and marketing plans. It is essential for companies to provide quick access to their account information, payment transaction data, and new products to gain market share in the highly competitive payment ecosystem.

Key advantages of chatbots that are enabling organizations to optimize their resources are:

• Cost efficient compared to traditional call centres.
• Consistent interaction model with customers irrespective of volume of requests and
• Ability to learn from new interactions.

Incoming mail processing

Many companies still handle large volumes of incoming mails coming from customers, vendors, staff and so on. Handling all such mails manually is time consuming and prone to errors. AI helps in automating incoming mail processing using supervised learning techniques. This helps in understanding key data elements of the mail, categorise them, raise tickets in systems. It can further be coupled with RPA (robotic process automation) tools to fully automate and process incoming mails efficiently.

Decision Management

Payments industry being highly competitive, customer delight and retention is one of the top priorities for financial institutions. AI helps financial institutions to gain a better understanding of their customers using large volume of transactions processed by them on a daily basis. It is useful in understanding the data, identifying patterns and helping in taking decisions in real time to improve customer experience.

Issue resolution

Many organizations face challenges in managing defects or issues that come through help desks or test cycles. To improve time-to-market, it is essential to fast track issue resolution. AI, by making use of historical issue data, can help in categorizing issues so that it can be resolved faster by resolution teams.

All these references clearly state that AI is gradually becoming an invaluable asset in operations. It will be interesting to note how the payments ecosystem will further leverage this tool in the coming days.

Venkat Sriramagiri is COO of Worldline India. The views in this article are his own.

Venkat Sriramagiri