Use of AI in Banking – Part I

Artificial Intelligence (AI) is a key technology of digital change and is increasingly utilised in the financial world. Suitable application areas for AI are information-intensive, recurrent processes. Some financial services providers are already using AI applications to review transactions, provide investment advice, detect fraud, or credit-worthiness auditing. Together with high-performance hardware and software platforms, AIs methods allow large amounts of data to be recognized without explicit programming to recognize complex relationships. Among the best known AI applications are the “intelligent assistants” or “personal voice assistants” such as Amazon Alexa, Google Home, or Deutsche Telekom Magenta: applications which capture, process, and interpret natural language queries and commands, and “reply” with answers via simulated human speech. Chaining queries together to add context allows simple simulated human-robot conversations to be built up.

Using speech recognition and parsing algorithms in conjunction with semantic technologies, these devices are easy to use and support the user in a defined set of tasks while acting as a new distribution channel. Well-known applications are audio and music services, news services (weather, traffic, stock exchange, sports, navigation, etc.). Even financial services providers recognize the added value of this communication interface and are already using Alexa to announce stock quotes, stock news and general information.

The banking sector is predestined for the use of AI due to the amount of data available and the extensive documentation of processes that are characterized by numerous repetitions. Banks use AI primarily in intelligent agents (software agents or chat bots), as part of lending and underlying processing or in securities trading.

The first part of the article discusses the use of intelligent chatbots, personal voice assistants (PVA), and AI in marketing. The follow-up contribution deals with the topics of credit checks and fraud prevention, investment advice / investment and compliance.

Smart chatbots

In the field of AI-controlled assistance, banks are increasingly using intelligent chatbots with which users can also communicate and interact directly. Advanced bot platforms work with AI algorithms compared to simple chatbots that access a predefined content library through scripting applets.

The central element of the intelligent chatbot systems is the deposited knowledge base (database) and the implemented algorithms for process / process control. A chatbot works on the principle of pattern power, pattern matching. The knowledge base is based on a recognition pattern (rules) for possible questions of the users with appropriate answer texts and actions. From the questions asked by the users, the algorithm compares the keywords with the stored answers in the knowledge base. If corresponding keywords are stored in the knowledge base, a response is issued or an action is executed.

Personal Voice Assistants (PVA)

PVA are advanced chatbot platforms that translate the user’s linguistic input as part of an imitated dialogue into hands-on instructions, thereby assisting the user (see Apple Siri, Amazon Echo, or Google Home). Applications enable customers to query the account balance for individual and joint accounts, overall financial status, and information about entries and debits on the account by voice command. It supports Google Home speakers as well as compatible third-party speakers and other devices using Google Assistant.


The Predictive Behavioural Targeting technique enables banks to use the artificial neural networks (CNN) and deep learning methods of selfless user surveys, user data, and external data sources to create statistical predictions on user behaviour patterns. The formation of profiles consisting solely of search words and Big-Data-based evaluation algorithms allows probabilities to be approximated for upcoming purchase decisions. Instead of buying large datasets about the offline and online behaviour of specific user groups and segments, banks can instead optimize their campaigns with the help of future-requested keywords.


The author Prof. Dr. Dirk Neuhaus, MBA is Professor of Information Systems in Financial Services.


Prof. Dr. Dirk Neuhaus, MBA
Professor of Information Systems in Financial Services
Hochschule der Sparkassen-Finanzgruppe
Simrockstr. 4
53113 Bonn

Phone: (0228) 204 – 9936
Fax: (0228) 204 – 9939

Related posts