By Radosław Bekanowski, Product Manager at Comarch
editor's pickWednesday 24, July 2019
The dynamic development of AI prompts banks to seek applications for it in finance.
Artificial Intelligence (AI) Machine Learning (ML) or RPA Robotic Process Automation (RPA) were until recently associated mainly with literature and movies and given the Sci-Fi label.
Similarly, Augmented Reality (AR) and Virtual Reality (VR, on their part, have been introduced to the public and promoted mainly by game developers as well as the entertainment industry.
Virtual assistants—communicating with users both via text and voice—now is features in our homes and cars thanks to the largest technology companies (Apple, Google as well Microsoft or Amazon), however the financial sector is also following suit adopting these technologies.
As a result, solutions supporting decision making, fraud prevention as well as process automation or new communication channels with customers (chatbots, voice assistants, AR, VR) are among the most widely discussed aspects of using AI for business processes in banks.
The most popular areas of using new technologies in loans
AI and Machine Learning can help resolve key industry challenges in the loan origination process such as cost and cycle times. The use of mobile technologies, AR, VR or virtual assistants enables the emergence of new sales channels, which significantly improves customer experience when applying for loans. The banking sector is noticing a growing potential in the development and implementation of new technologies, especially in the context of the growing competition from fintechs.
The potential applications of modern technologies to credit processes are:
Automating and streamlining processes
Machine Learning (ML) can be used in labour-intensive tasks that require a significant amount of resources and operational teams dedicated to verifying income, assets and insurance coverage.
Automation of manual and routine tasks performed by humans can not only lead to expedition of the origination process but enables management teams to increase the volumes of their business by delivering a more positive and personalised consumer experience.
Systems based on AI and ML run more processes that do not require special analysis by humans and can be fully automated. Barclays, a UK-based automated a number of processes such as debt analysis and verification of fraudulent practises, thereby reducing collateral for bad loans by approximately $225 million per annum.
In addition, process support systems based on AI and ML can play an advisory role in helping the user (bank employee or customer) find the best solution tailored to their needs and characteristics. Thus, on the basis of specific customer data, the system can suggest for instance the best way to finance a specific customer need.
Fraud detection and risk management
Accurate and thorough analysis of creditworthiness—based on data held by the bank and data analysis in external data sources (credit bureaus, business intelligence agencies) as well as in social media. Such analysis may be a foundation or complement for scoring or rating models used by banks.
Apart from that, algorithms based on AI and ML allow to minimise the risk related to money laundering and fraud by detecting suspicious behaviour of both customers and employees.
Virtual customer service / new sales channels
Chatbots or voice assistants are not only simple applications answering pre-defined questions, most frequently asked by users but also systems which, on the basis of subsequent conversations and interactions with users, learn to provide services at a higher level in subsequent interactions. Thanks to them, it is possible to fill out the entire credit application form based on a conversation similar to a natural one.
The use of VR and AR allows for non-standard presentation of products offered by bank partners in the marketplace, which facilitates the customer relationship with a financial institution. Products offered in this way by the bank can be credited, which, again, allows the bank to monetise cooperation with partners.
How does Comarch employ new technologies in loan origination processes?
Comarch Loan Origination enables more efficient control of credit risk and allows for cutting time needed to grant a loan. The system automates the work of client advisors managing the credit-granting process. It allows banks to optimise credit management end-to-end—loan simulation, application verification as well as analysis of customer financial situation, decision making and fund disbursement.
Comarch AML is an AI-powered fraud detection software dedicated to financial institutions obligated to monitor, investigate and report suspicious transactions to financial regulatory authorities. The system can spot and recognise relationships and similarities between data and, further down the road, learn to detect new anomalies or classify and predict specific events.
Devra is a virtual, voice-based cloud assistant whose job is to answer various financial questions asked via any device such as phone, Alexa,car or Messenger. Devra is kind of a digital equivalent of a financial advisor and it uses natural language processing technologies to recognise and synthesise speech.
Comarch AR Loans is a mobile application that allows users to find their dream house nearby with the help of Augmented Reality and easily apply for a mortgage. The application combines the concepts of marketplace and non-banking services offered by financial institutions to retail and SME customers.