International Scientific Symposium Horizons of Economic Research in a Global Context (HERGC)

In the past several years, the financial markets have been facing transition from traditional finance to algorithmic finance. The introduction of the newly developed data-driven processes has increased interest among the researchers, as well as the business, due to the impact of these new processes on the daily management and business operations. Therefore, this paper examines these new changes and their
implications and analyzes how advanced algorithmic technologies create new opportunities to influence
market dynamics and the behavior of market participants.
Although artificial intelligence is in its early stages, currently it is present in the processes of financial
decision-making, investment management and portfolio optimization, credit scoring and risk assessment, algorithmic and high-frequency trading, fraud detection and financial security, as well as reducing human bias and errors. The human factor is still very important, but AI assists with greater precision, contributes to more accurate risk assessments and decreases the risk of fraudulent transactions. The further advancement and development of artificial intelligence is limitless. Nowadays, investors are facing daily changes in the trading systems, and the faster they adapt to the new technologies, the better
the outcomes will be. Considering the importance of these new possibilities, in this paper it is deeply
elaborated on the benefits of AI implementation in finance. It is provided comparison between the traditional and algorithmic finance and the future challenges that investors may face because of the implementation of artificial intelligence. In this context, there are several types of AI used in finance, such as Machine Learning (ML), Deep
Learning, Natural Language Processing (NLP), Reinforcement Learning, Expert Systems (Rule-Based AI), Robotic Process Automation (RPA), Hybrid AI Systems. On one hand, Machine Learning in finance is used to analyze large datasets, and its main application is in algorithmic trading, risk management, credit scoring, fraud detection, customer analytics etc. On the other hand, Deep Learning is an advanced option of ML and it is based on artificial neural network algorithms (Huang, et al, 2020). It is mainly
used for analysis of big data, price prediction from complex data and algorithmic trading with sequence modeling. Additionally, Natural Language Processing (NLP) has become very important in financial research, and it is widely used in public opinion monitoring, financial report interpretation and risk assessment. From extracting key information, this technology is used in understanding the emotions of the investors, which may affect the stock market dynamic or monetary value (Xiao, et al, 2024).
Reinforcement Learning is used in improving decisions in complex financial environments with fewer model assumptions. The evolution of this technology is the fact that it combines classical RL theory with deep learning, and it works in real and simulated environments. Its main limitations are storage and simple function approximations (Hambly, et al, 2023). Expert Systems (Rule-Based AI) are the simplest form of AI, and they mainly use rules coded into the system rather than static facts (Grosan and Abraham, 2011). Expert Systems are more traditional and are mainly used in credit scoring and loan approval,
fraud detection, portfolio management and investment advice. Moreover, Robotic Process Automation (RPA) has also big importance in finance. This technology assists in improving efficiency, accuracy and data handling and its application are mainly in market risk models, credit risk models, integrated risk assessment models and operational and regulatory compliance models. Main benefits of RPS in finance are for data gathering, processing, analysis, and reporting (Kothandapani, 2023). Last but not least,
Hybrid AI Systems have several key components, such as neural networks, fuzzy logic, genetic algorithms. These components help in pattern recognition and predictive analysis, optimes decision making and problem solving and handles uncertainty (Corral de La Mata., et al, 2024). Undoubtedly, these technologies are contributing to better efficiency and accuracy in the decision-making process.
The automation of various tasks helps many companies to adapt faster to the complex and dynamic financial environment. Also, implementation of new generation AI approaches will assist in overcoming many financial problems and potentially transform the entire financial services system. For instance, Fintech becomes very important in the financial markets, because it is very beneficial and allows better credit access for individuals and small businesses. Banking sector, especially traditional banks, are
becoming more dependent on fintech services. However, new technologies come together with new types of risks. It is also worth mentioning that although artificial intelligence has many positive effects and contributes to better efficiency and effectiveness in the financial markets, it still has some drawbacks. As mentioned in the paper of Boppiniti (2021), main issues of the new technologies are
related to data privacy, ethical concerns, transparency and accountability in decision-making processes.
Considering these challenges, many regulatory bodies started working on addressing these issues. Evolving the regulatory framework is mainly needed for two reasons, to protect the data privacy of the consumers and to allow uninterrupted fintech development (Cao and Zhai, 2022).