INTRODUCTION:

Artificial intelligence (AI) enables computer programs to learn from data and use data without human input or assistance. AI systems observe their environment, analyse data, and draw conclusions, taking necessary actions. They continually improve their performance by learning from their previous judgments and enhancing their abilities. (Kaya and Schildbach, 2019). The global market for AI was valued at $65.48 billion in 2020, with a projected growth to $1,581.70 billion by 2030, according to Bloomberg. The transformative impact of AI on various industries, such as banking, financial markets, e-commerce, education, gaming, and entertainment, has been widely acknowledged. A study by NASSCOM reports that investments in AI applications in India are expected to grow at a compound annual growth rate of 30.8% and reach $881 million by 2023. (Shah Aasif, n.d.)

AI is increasingly prevalent in today's economy, including the banking sector. Banks are utilizing innovative AI applications to save time and reduce costs, employing algorithms to achieve reliable outcomes, boost sales performance, enhance customer service, and increase revenue. By leveraging deep learning and machine learning, AI can reduce errors caused by psychological and emotional factors. Extracting pertinent information from multiple inputs and generating conclusions is one of AI's key capabilities. (Kaur et al., 2020). AI usage has surged in the 21st century, with banks utilizing machine learning algorithms for detecting credit card fraud for over a decade. In 2014, Man Group began using AI for managing client investments, while Bank of America's chatbot Erica showed significant improvements in customer relations by 2016. Additionally, several institutions developed recommendation systems in 2018. (Artificial Intelligence in the Financial Sector, 2020)

API (Application Programming Interface) has revolutionized the banking industry by facilitating seamless collaboration and integration of traditional banks with digital banks, third-party entities, including both banking such as Neo Banks and non-banking institutions such as PhonePe, paytm, etc. Even the Open Banking employs Open APIs to authorize third-party developers with access to customers' financial data, enabling them to create applications and services that enhance the flexibility and feasibility of monetary transactions. This has enabled FinTech’s to offer targeted financial solutions to niche customer segments, leading to critical advancements in the banking ecosystem. With the rise of UPI payments during the pandemic, the need for innovative banking services increased, leading to the growth of banking APIs in India and thus AI has a ripple effect in the entire banking sector. (API -The Power That Drives New-Age Banking Ecosystems|M2P Fintech Blog, 2022)

As depicted in Figure 1, By 2030, $300 billion is expected to be the economic value of AI in banking, according to IHS Markit. The Asia Pacific region ranks second in terms of the amount of AI used in banking, with a business value of $11.5 billion in 2018 and a projected growth to $50.6 billion in 2024. The Asia Pacific region is expected to generate $98.6 billion in revenue from AI in banking by the year 2030. During the next ten years, demand for AI in the banking industry is projected to be driven by nations like China, Japan, South Korea, Hong Kong, and Singapore.

Figure 1: The business value for the world market for AI in banking by region

Source: IHS Markit (Global Business Value of Artificial Intelligence in Banking to Reach $300 Billion by 2030, IHS Markit Says, 2019)

REVIEW OF LITERATURE:

Ghandour (2021) conducted a study that identified several opportunities for the banking sector with the implementation of AI, including personalized services, smart wallets, decision-making and problem-solving, improved customer satisfaction and loyalty, automated processes, enhanced transactional security and cybersecurity, and promotion of digital financial inclusion. Banks worldwide are leveraging Artificial Intelligence (AI) to personalize experiences for their customers, as per a study by (Triveni et al., 2020). They emphasised that AI is being used in diverse banking sectors like data analytics, blockchain, and machine learning, promising significant efficiency gains for banks and their clients. Königstorfer and Thalmann, (2020) has suggested that commercial banks can use AI to reduce lending losses, improve payment processing security, automate compliance tasks, and enhance customer targeting. However, implementing AI in these areas poses significant challenges, as noted in their study. Malali and Gopalakrishnan (2020) observed the banking and financial industry is being significantly disrupted by AI, providing opportunities for banks to transform their operations and develop new and innovative products and services. AI has become crucial for banks to stay ahead of competitors and remain competitive in the market.

Rahman et al., (2021) showed that attitude towards AI, perceived usefulness, perceived risk, perceived trust, and subjective norms have a significant impact on the intention to adopt AI in banking services. However, perceived ease of use and awareness did not have a significant effect according to the quantitative results. Butenko, (2018) found that banks can develop a multilevel structure for implementing artificial intelligence systems and use digital transformation tools to bridge the gap between real and virtual worlds. This will allow the use of AI in banks to enhance customer service and improve business efficiency. Vedapradha and Ravi (2018) Top of FormBottom of Formstates that investment banks are also implementing AI-based technology to automate their processes and enhance customer experience. This technology primarily aids in fighting financial crimes such as money laundering, resulting in improved compliance, credit-underwriting, and smart contracts etc.

Gomez and De Pablos-Heredero, (2020) expresses that Al technology allows banks to establish new connections with their customers, understand their needs and experiences, and enhance their competitiveness by improving their offerings. Furthermore, it enables banks to promptly address customer inquiries and improve their value chain by modernizing traditional banking practices. . Boustani, (2021) The use of AI enhances the standard of banking transactions to a higher level. However, AI cannot substitute human interaction in the case of clients dealing with bank personnel. Evidence indicates that the widespread adoption of AI could coincide with an increase in employment opportunities within the banking industry. Lui and Lamb (2018), AI is becoming more popular in the finance industry as it provides tailored services, but there are issues such as bias and discrimination that need to be regulated. The Financial Conduct Authority (FCA) is at the forefront of technology regulation, working alongside FinTech companies and utilizing regulatory sandboxes to address these challenges.

Applications of AI in Banking:

The study focuses on transformative impact of AI on banking, including the development of following innovative applications that improve customer service, save time, and reduce costs.

· Digitalization instead of branch lines: It's because of AI, banking processes are now further improved by automating tasks such as document processing, fraud detection, and customer service. This can lead to faster and more efficient service for customers while also reducing costs for banks. (Vijai, 2019)

· Robo Advisors: AI-driven Robo-advisors are employed in financial services to assess a client's financial status and background. They offer appropriate investment advice, considering the evaluation and the client's objectives, and may suggest specific products or equities in a particular category. (Dumasia, 2021)

· Fraud Detection: Using unsupervised algorithms can help banks detect fraud faster and with less human effort. However, creating an accurate ML model for fraud detection is difficult and requires careful consideration to avoid false positives. (Owczarek, 2022)

· Facial Recognition for Frictionless Payment: Facial recognition payments are already used in China at self-service stores and restaurants like KFC. Customers' faces are scanned and matched to unique DNA to track preferences and offer recommendations. Payment is quick and easy, eliminating the need for a phone or credit card. (Owczarek, 2022)

· Accurate Decision-Making: In the data-driven era, banks, investment firms, and insurance providers can benefit from lower-cost management practices by relying on technology rather than experts. With the help of systems and machines that analyse vast amounts of data, managers can make more informed decisions and take suitable actions. (Malali and Gopalakrishnan, 2020)

· Cognitive process automation: Claims management in banking can be expensive and error-prone, but cognitive process automation can automate and improve the process. This leads to a guaranteed ROI, lower expenses, and faster, more accurate service processing. Through machine learning, cognitive process automation continuously improves and automates a range of processes. (Triveni et al., 2020)

· Automated Teller Machine (ATM) Helpline: ATM helplines are available for emergencies, and now ATMs themselves are equipped with AI technology. This includes automated facial recognition, machine vision cameras, predictive maintenance, and forecasting cash demand. By using machine learning, ATM security and user experience are improved. (Kaur et al., 2020)

· Anti-money laundering and fraud detection: The effectiveness of anti-money laundering and fraud detection in the banking industry can be improved by using AI models, machine learning algorithms, neural networks, and anomaly detection systems. (Ghandour, 2021)

· Chatbots for Customer Service in Multiple Languages: Retail banks integrate chatbots into their platform to assist customers with financial services, such as advance approval procedures, credit card cancellation, and balance monitoring. Chatbots use NLP software to understand customer queries and connect to relevant areas of the flexible application. AI models are trained based on current financial conditions to improve chatbot performance. (Ramana et al., 2022)

· Smart Wallets: Digital wallets enable mobile payment transactions, decreasing reliance on physical currency and promoting financial inclusion. To deliver insights on investments, performance, forecasting, and risk management, banking information systems are integrating personal robots, big data analytics, neural networks, machine learning, and predictive analytics. (Ghandour, 2021)

· Know Your Customer (KYC) and Anti-Money Laundering (AML): Financial institutions have been fined $26 billion in the past decade for not complying with sanctions, KYC, and AML regulations. To reduce compliance costs, banks are using AI to automate KYC procedures and detect AML activity. (Katyal, 2021)

· Credit Assessment: A Personal Assistant powered by Neural Network using classification models can effectively manage credit assessment and decision-making. Market conditions, lifestyle, risk appetite, financial goals, and past behaviour are analysed to develop successful financial strategies adhering to credit standards. (Veerla, 2021)

· Blockchain Technology and Banking: Blockchain is a digital ledger stored on a public database, with AI used for analysis and decision-making. It's not just for cryptocurrencies, but also solves data security and fraud prevention issues. It has many applications like transparency, cross-border remittances, and loan syndication. (Kaur et al., 2020)

As depicted in Figure 2, Front, Middle, Back-office AI applications present significant cost-saving opportunities of around $199 billion, $217 billion and $31 billion respectively within banks. AI is increasingly being utilized in the front-end to facilitate customer identification and authentication, replicate human interactions through chatbots and voice assistants i.e., conversational banking, foster stronger customer relationships, and provide personalized insights and recommendations. Furthermore, banks are implementing AI within their middle-office functions to effectively identify and prevent payments fraud, enhance processes related to anti-fraud and risk, anti-money laundering (AML) and know-your-customer (KYC) regulatory checks. By leveraging AI technology, banks can streamline operations, reduce manual efforts, and improve compliance with regulatory requirements. Back-office AI applications, such as credit underwriting and smart contract infrastructure, are also being adopted by banks to enhance operational efficiency and optimize processes.

Challenges of Implementing AI in Banking:

· Trained Manpower: There is a shortage of trained personnel with the necessary data science skills, resulting in a limited number of qualified data scientists available to work on AI. The existing workforce in banks is not familiar with the latest tools and applications. (AI in Banking A Primer, 2020)

· Lack of Quality Data: Adequate quality data is essential to ensure that the algorithm can perform optimally in real-world scenarios. In addition, if data is not in a machine-readable format, it may cause unexpected behaviour in the AI model. (Singh, 2023)

· Privacy and security: AI security is tricky as hackers can manipulate data to deceive AI systems. Users must balance personalization with privacy by choosing how much data to share. More data yields a personalized experience, while less data yields a general one. (Bhatnagar et al., 2022)

· Localization: Localization is important in banking, especially for developing models that work in different markets. When using AI in financial services, it's important to consider language, culture, and demographics to ensure a good customer experience. (Golden, 2021)

· Lack of Transparency: The lack of transparency in AI implementation by banks can lead to trust issues among customers due to the complexity of the technology. To build trust, it is crucial for banks to improve transparency in their AI systems. (Mudgal, n.d.)

· Policy Non-Compliance: Organizations using AI need to follow existing policies to avoid breaking rules. Regulators are paying more attention to AI in finance and have made rules to supervise it. These rules can affect how AI is used and managed, so it's important to keep up with them. (Mudgal, n.d.)

· Unwillingness: Some banks are hesitant to adopt or create new practices related to AI. This could also be true for certain areas in tier two and three cities, where traditional methods are deeply ingrained and standardized. (Vadapalli, 2022)

· Restrictive AI Implementation and Operational Resources: Large-scale AI systems could be prohibitively costly to set up and operate, which would be especially difficult for small institutions with constrained resources. To maintain efficient and effective AI technology operations, skilled data science skills would be necessary in addition to the initial costs. (Ghandour, 2021)

· Modelling Techniques: The challenge lies in developing modelling techniques that can accurately estimate model parameters using only a limited amount of data samples. The need for high precision in such parameter estimations further adds to the complexity of this challenge. (Triveni et al., 2020)

· Different approaches of enforcement: It is challenging for businesses to develop efficient worldwide standards and to estimate the risk associated with deploying AI technologies globally due to different enforcement strategies. (Katyal, 2021)

· Explainability of AI models: A major problem with AI is that it is quite complex. AI is based on algorithms, and people who are not familiar with these algorithms might find it difficult to grasp the functioning of AI-driven decision making. (Bhatnagar et al., 2022)

· Bias and discrimination: AI has the potential to magnify biases and discrimination present in society. Therefore, it is imperative for banks to ensure that their AI systems are fair and promote equal opportunities by eliminating any biases. (Best and Rao, n.d.)

· Integration with legacy systems: Banks often have outdated and complex legacy systems that may not be compatible with modern AI technologies. In order to seamlessly integrate their AI systems with existing ones, banks will have to invest in modernization and integration projects. (Building Resilience in Banking by Overcoming Challenges of Legacy Systems, 2023)

CONCLUSION:

The banking industry is experiencing a revolution as a result of AI, which is providing new opportunities for improving efficiency, cost savings, and customer satisfaction. AI is being utilised in various ways, such as personalising services, improving transactional security, and automating compliance activities. Despite its benefits, AI implementation poses significant challenges, such as employment displacement and the potential for bias and discrimination. However, banks can establish multilevel structures and use digital transformation tools to link the physical and digital worlds and effectively integrate AI systems. Additionally, the widespread adoption of AI could create new employment opportunities in the banking sector. It is crucial to regulate the use of AI to ensure that its benefits are maximised while minimising its negative effects. Overall, AI has become an essential instrument for banks to remain ahead of their competitors, transform their business practices, and create innovative and state-of-the-art products and services.

REFERENCES:

1. AI in Banking A Primer. (2020). https://www.idrbt.ac.in/wp-content/uploads/2022/07/AI_2020.pdf

2. API -The Power that Drives New-Age Banking Ecosystems|M2P Fintech Blog. (2022, April 28). https://m2pfintech.com/blog/api-the-power-that-drives-new-age-banking-ecosystems/

3. Artificial Intelligence in the Financial Sector. (2020). https://www.nttdata.com/global/en/-/media/nttdataglobal/1_files/services/data-and-intelligence/artificial_intelligence_in_the_financial_sector.pdf?rev=704e0e643b6b4f3bbd1ef22b04cdcd3e

4. Best, M., and Rao, A. (n.d.). Understanding algorithmic bias and how to build trust in AI: PwC. Retrieved March 21, 2023, from https://www.pwc.com/us/en/tech-effect/ai-analytics/algorithmic-bias-and-trust-in-ai.html

5. Bhatnagar, J., Kanikarla, R., Mishra, R., Singh, H., Arya, A. K., Panmand, M., and Sharma, P. (2022). AI adoption in Indian financial services and related challenges.

6. Boustani, N. M. (2021). Artificial intelligence impact on banks clients and employees in an Asian developing country. Journal of Asia Business Studies, 16(2), 267–278. https://doi.org/10.1108/JABS-09-2020-0376

7. Building Resilience in Banking by Overcoming Challenges of Legacy Systems. (2023, March 14). https://www.intelliswift.com/blogs/building-resilience-in-banking-by-overcoming-challenges-of-legacy-systems

8. Butenko, E. (2018). Artificial intelligence in banks today: Experience and perspectives. Digest Finance, 25(2), 230–242. https://doi.org/10.24891/DF.25.2.230

9. Dumasia, J. (2021, November 17). 5 applications of Artificial Intelligence in banking. https://ibsintelligence.com/ibsi-news/5-applications-of-artificial-intelligence-in-banking/

10. Ghandour, A. (2021). Opportunities and Challenges of Artificial Intelligence in Banking: Systematic Literature Review. TEM Journal, 10(4), 1581–1587. https://doi.org/10.18421/TEM104-12

11. Global Business Value of Artificial Intelligence in Banking to Reach $300 Billion by 2030, IHS Markit Says. (2019, April 10). https://news.ihsmarkit.com/prviewer/release_only/slug/technology-global-business-value-artificial-intelligence-banking-reach-300-billion-203

12. Golden, C. (2021, January 28). The Four Key Challenges of AI in Financial Services. https://appen.com/blog/key-challenges-of-ai-in-financial-services/

13. Gomez, C., and De Pablos-Heredero, C. (2020). Artificial Intelligence as an Enabling Tool for the Development of Dynamic Capabilities in the Banking Industry. International Journal of Enterprise Information Systems (IJEIS), 16(3), 20–33. https://EconPapers.repec.org/RePEc:igg:jeis00:v:16:y:2020:i:3:p:20-33

14. Katyal, I. (2021). Role of Artificial Intelligence and Analytics in Banking. The Journal of Indian Institute of Banking and Finance.

15. Kaur, Dr. N., Sahdev, S. L., Sharma, Dr. M., and Siddiqui, L. (2020). Banking 4.0: ‘The Influence of Artificial Intelligence on the Banking Industry and How AI Is Changing the Face of Modern Day Banks’ by Dr. Navleen Kaur, Supriya Lamba Sahdev, Dr. Monika Sharma, Laraibe Siddiqui: SSRN. International Journal of Management, 11(6), 577–585. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3661469

16. Kaya, O., and Schildbach, J. (2019). Artificial intelligence in banking. www.dbresearch.com

17. Königstorfer, F., and Thalmann, S. (2020). Applications of Artificial Intelligence in commercial banks – A research agenda for behavioral finance. Journal of Behavioral and Experimental Finance, 27. https://doi.org/10.1016/J.JBEF.2020.100352

18. Lui, A., and Lamb, G. W. (2018). Artificial intelligence and augmented intelligence collaboration: Regaining trust and confidence in the financial sector. Information and Communications Technology Law, 27(3), 267–283. https://doi.org/10.1080/13600834.2018.1488659

19. Magana, G. (2020). AI in Banking. https://store.businessinsider.com/products/ai-in-banking-and-payments?IR=Tanditm_source=businessinsideranditm_medium=content_marketinganditm_campaign=report_teaseranditm_content=buy_buttonanditm_term=buy_button_link-ai-in-banking-reportandvertical=banking#!

20. Malali, A. B., and Gopalakrishnan, S. (2020). Application of Artificial Intelligence and Its Powered Technologies in the Indian Banking and Financial Industry: An Overview. IOSR Journal Of Humanities And Social Science (IOSR-JHSS, 25(4), 55–60. https://doi.org/10.9790/0837-2504065560

21. Mudgal, Y. (n.d.). Artificial Intelligence Risk and Governance - AI and Analytics for Business. Retrieved March 21, 2023, from https://aiab.wharton.upenn.edu/research/artificial-intelligence-risk-governance/

22. Owczarek, D. (2022, March 29). AI in Banking. Applications and Benefits of Artificial Intelligence in Financial Services. https://nexocode.com/blog/posts/ai-in-banking.applications-and-benefits-of-artificial-intelligence-in-financial-services/

23. Rahman, M., Ming, T. H., Baigh, T. A., and Sarker, M. (2021). Adoption of artificial intelligence in banking services: an empirical analysis. International Journal of Emerging Markets. https://doi.org/10.1108/IJOEM-06-2020-0724/FULL/XML

24. Ramana, S. V., Katta, A. K., Rao, P. S., Rao, S. S., and Raja, R. (2022). Impact of Artificial Intelligence on Fraud Detection in Retail Banking Products. International Journal of Intelligent Systems and Applications in Engineering, 10(4), 124–129. https://ijisae.org/index.php/IJISAE/article/view/2206

25. Shah Aasif. (n.d.). India must be prepared to meet challenges that AI poses on economic, security front | The Indian Express. Retrieved March 20, 2023, from https://indianexpress.com/article/opinion/columns/india-ai-challenges-economic-security-front-8477517/

26. Singh, S. (2023, February 13). AI in Banking – How Artificial Intelligence is Used in Banks. https://appinventiv.com/blog/ai-in-banking/

27. Triveni, K., Bharathi. G., R., and Nawaz, M. (2020). Application of Artificial Intelligence in Indian Banking-Opportunities and Challenges. International Journal of Trend in Scientific Research and Development, Volume-5(Issue-1), 429–432. www.ijtsrd.com

28. Vadapalli, P. (2022, November 22). Artificial Intelligence in Banking 2023: Examples and Challenges | upGrad blog. https://www.upgrad.com/blog/artificial-intelligence-in-banking/

29. Vedapradha, R., and Ravi, H. (2018). Application of Artificial Intelligence in Investment Banks. Review of Economic and Business Studies, 11(2), 131–136. https://doi.org/10.1515/REBS-2018-0078

30. Veerla, V. (2021). To Study the Impact of Artificial Intelligence as Predictive Model in Banking Sector: Novel Approach. IJIRT, 7(8).

31. Vijai, C. (2019). Artificial Intelligence in Indian Banking Sector: Challenges and Opportunities. International Journal of Advanced Research, 7(4), 1581–1587. https://doi.org/10.21474/IJAR01/8987

Received on 20.05.2023 Modified on 14.07.2023

Asian Journal of Management. 2023;14(3):166-170.

DOI: 10.52711/2321-5763.2023.00028