AI and central banks: Transforming financial systems

The implementation of AI in financial systems is changing the central banking systems around the world with more prospects that may transform the world economy in a positive or negative manner. Monetary authorities standing at the crossroads of monetary policy and financial stability are more and more actively adopting AI to optimize their work and counter threats.

Due to this, AI is very important for financial institutions because it can work through large amounts of data and make pattern identifications. In most of its operational fields including payments, lending, insurance, and asset management, AI is improving on the efficiency and cutting on the cost. For example, through chatbots, AI facilitates timely customer support and reduces the costs related to it for financial institutions, while at the same time, increasing efficiency in the detection of frauds and meeting the regulations.

In payment industry the problems which are pertained for a long time solved by AI includes know your customer (KYC) and anti money laundering (AML). In this way, AI not only contributes to cost and compliance reduction but also counteracts the tendency of correspondent banking’s fragmentation observed throughout the world, therefore worsening the situation.

AI also presents major improvements in the area of credit appraisal. Generally, the routine credit reference employs the conventional rating models that do not give a complete picture of the subject’s creditworthiness. Yet, for this purpose, AI can process a larger number of parameters, including non-financial ones, which results in more or less accurate conclusions. This capability can make credit accessible to people considered as ‘invisible primes,’ those who would be creditworthy despite the poor credit scores.

In sum, countless opportunities come hand in hand with the integration of AI into the financial sector; however, new threats have emerged as a result. Cybersecurity continues to be an area of focus,

especially as AI increases the probability and complexness of cyber attacks. Introducing generative AI or gen AI can increase the range of hackers’ possibilities: phishing can become more believable, malware can be more efficient. Moreover, AI systems as well as datasets are subject to attacks like data poisoning, prompt injection which threatens the systems’ efficiency.

Bias and discrimination can also not be looked at as a disregard towards the black community. In case AI models are trained with data that contain proportional representation of biases that are existent in a society, then the crafted models will have an aspect of repeating the same. This entails probabilities of bias in determination of loans and insurance, hence denial of credit to some members of the society.

Other operational risks are associated with dependence in the procurement of the models on a few providers. This concentration elevates third party dependency risks and can intensify the financial stability risks in procyclicality and market fluctuation. Since most financial institutions use AI, coordinated actions may amplify movements in the markets.

Therefore, central banks are not mere spectators to the role performed by AI but are rather using AI to achieve their goals. Regarding the specific objectives, AI assists central banks in the collection of data, statistical analysis of the market and payment systems, and the management of payment systems. Initiatives such as the BIS Innovation Hub’s Project Aurora as well as Project Agorá demonstrates how AI can further AML efforts and even optimize payment network by utilizing such instrument as tokenization and private algorithms.

Furthermore, central banks have singled out the applicability of gen AI in the area of cybersecurity. Thus, AI has the potential to build up the strength of financial structures from cyber threats by automating repetitious processes and improving threats recognition.

Therefore, AI will continuously prove to be an essential tool in central banking and overall financial systems. That is why, despite the great advantages in relation to the increase of effectiveness and financial liberalization, it also contains definite risks that must be effectively addressed. As the providers of financial and economic stability, central banks have the ability and the responsibility to direct the integration of AI toward the organization’s beneficial outcomes.