The adoption and integration of artificial intelligence (AI) in Ghana’s and Kazakhstan’s financial sectors signifies a transformative change, driven by technological advancement and pursuit of greater efficiency, improved risk management and enhanced customer experience. The study provides a comparative analysis of AI adoption in developing countries, focusing on key areas such as banking, investment management, legal compliance and financial inclusion. AI adoption is gradually gaining attention in Ghana, where fintech start-ups and traditional banks are using AI for mobile banking, fraud detection, and credit scoring. However, challenges such as poor infrastructure, data security concerns and lack of a skilled workforce impede the widespread implementation of AI and its full realization. In contrast, Kazakhstan has made significant progress in adopting AI, driven by government initiatives, robust digital infrastructure, and growing fintech ecosystem. Financial institutions in Kazakhstan use AI for algorithmic trading, regulatory compliance and customer service automation, positioning the country as a regional leader in fintech innovation. Despite differences in the countries’ approaches to adopting AI, both economies face similar challenges, such as algorithmic bias, regulatory uncertainty and capacity-building needs. The present paper explains why tailored growth strategies are needed to address these issues. It highlights the importance of investment, public-private partnerships and legal frameworks in upskilling professionals and creating technological infrastructure. The two countries should develop roadmaps for AI-tailored growth policies in their financial sectors to ensure their effective adoption and implementation for financial development.

AI-powered innovations, AI-driven solutions, AI-based tools, Economic transformation, Economic potential, Financial inclusion, Financial sector, Fintech.

The rapid development of artificial intelligence (AI) has led to a fundamental transformation of the global financial landscape. Many countries are experiencing significant changes in their financial sectors as a result of this development (Rose Innes & Andrieu, 2022). This transformation is particularly notable in economies such as those of Ghana and Kazakhstan. According to the Adrian (2024), the adoption of AI technologies in these countries presents unique opportunities and specific challenges to their financial ecosystems.

Ghana, one of the major economies in West Africa, has made substantial progress in its financial technology sector. According to estimates from the Bank of Ghana (2023), the value of mobile money transactions in 2023 reached approximately $80 billion. Similarly, Kazakhstan, a leading economy in Central Asia, has witnessed significant development in the adoption of digital banking. The Organization Economic Cooperation and Development (2023) estimates that over 73% of the Kazakhstani population actively use digital financial services. The intersection of these advancements with AI technologies presents a compelling case for comparative studies.

Both economies’ financial industries are at a pivotal stage where the adoption of AI could help overcome the limitations of conventional banking infrastructure. Recent research has shown that AI technologies in developing markets can reduce operational costs by over 22% and raise financial inclusion by up to 35% (International Monetary Fund, 2024a). However, contextual variations between Ghana and Kazakhstan, ranging from regulatory policies to technological infrastructure, have led to differing adoption patterns and outcomes (Asian Development Bank, 2021).

This comparative analysis of two distinct economies aims to examine their similar developmental aspirations and different approaches to financial innovation. For example, Ghana is focused on mobile money activities (a type of mobile banking that does not require an internet connection) and microfinance innovations (Coffie & Hongjiang, 2023), while Kazakhstan has demonstrated a strong commitment to blockchain technology and the development of central bank digital currencies (International Monetary Fund, 2024b). These parallel yet oppositely directed approaches provide useful insights into the adaptability of AI in various emerging market contexts.

Although the financial sectors of Ghana and Kazakhstan differ in their economic, cultural and geographical characteristics, as illustrated in Table 1, the two countries have similar trajectories in their pursuit of economic modernisation and technological progress. Both countries are developing economies with promising financial markets, and the adoption of AI-driven technologies in their financial sectors reflects a wider global trend toward digital transformation. Artificial intelligence has emerged as a keystone of innovation in finance, helping institutions improve efficiency, enhance customer experience and manage risk effectively. However, the pace and extent to which AI-based innovations are adopted differs significantly between countries, influenced by factors such as infrastructure, regulatory frameworks and economic priorities. This paper presents a comparative analysis of the adoption of artificial intelligence in the financial sectors of Kazakhstan and Ghana. It explores the drivers, challenges, and implications of this technological and economic transformation.

Ghana is a country located in West Africa with a rapidly growing economy. It has made significant progress in using innovation and technology to transform its economy, particularly in the financial sector. It is known for its mobile money services, which have revolutionized financial inclusion by providing millions of unbanked residents with access to financial and banking services (Aker & Mbiti, 2010). The interoperability of mobile money services offered by telecommunications operators and banks has promoted digital innovation in the country’s financial sector, particularly with regard to cashless transactions. Recently, financial institutions in Ghana have begun integrating AI-based solutions, such as chatbots, fraud detection systems and algorithms for credit scoring to streamline transactions and improve customer services and experience. According to the Bank of Ghana (2021), the central bank has developed guidelines and regulatory sandboxes to promote innovation while safeguarding customer protection. Despite this technological progress, challenges remain in relation to technological infrastructure, the shortage of skilled AI personnel, and data privacy.

Kazakhstan is a country in Central Asia with abundant natural resources and a growing economy. It has focused on digital transformation as part of its wider economic transformation strategy. The financial sector in Kazakhstan has seen a rapid increase in the adoption of artificial intelligence (AI), driven by government initiatives such as the Digital Kazakhstan programme and the development of the Astana International Financial Centre (AIFC) fintech innovation hub (AIFC, 2020). Similar to Ghana, banks and financial institutions in Kazakhstan have adopted AI-driven tools for risk management, investment analysis and customer service, establishing the country as a regional leader in financial technology. However, Kazakhstan faces challenges related to data privacy, ethics, regulatory harmonization, infrastructure and the need to upskill the workforce to fully realize the potential of AI adoption.

Adopting AI in the financial sectors of these countries has far-reaching implications for socioeconomic development, financial inclusion and regulatory supervision. AI-powered technologies have the capacity to improve access to financial services, lower transactional costs and mitigate associated risks. However, the rapid integration of AI-driven innovations raises concerns regarding data security, algorithmic bias and potential job displacement. Furthermore, as the regulatory sandboxes in both countries are evolving, it is necessary to strike a balance between promoting innovation and protecting consumer rights.

Kazakhstan maintained a higher and more stable credit-to-GDP ratio, peaking at 37.73% and stabilising at 25-26% post-2020. In contrast, Ghana’s ratio declined sharply from 17.93% in 2015 to 9.96% in 2023, indicating potential problems in the credit market. Ghana’s population is growing rapidly, which could put a strain on its resources. Between 2015 and 2023, Ghana’s population grew from 28.7 million to 33.8 million. Kazakhstan’s population is growing slowly. It grew from 18.1 million in 2015 to 20.3 million in 2023. Ghana experienced more volatile growth, peaking at 5.78% in 2017 before slowing to 1.0% in 2023. In contrast, Kazakhstan demonstrated resilience and recorded 3.57% GDP per capita growth in 2023, showing recovery from pandemic shocks. Thus, Kazakhstan outperforms Ghana in terms of financial stability, post-pandemic recovery and economic resilience, while Ghana faces challenges regarding access to credit and erratic growth. Kazakhstan’s successes may be attributed to its pragmatic and comprehensive approach to adopting and using various strategies to boost technology and AI adoption in different industries, including the financial sector.

This comparative analysis explores the following major questions: What are the main factors influencing the adoption of AI in the financial sectors of these economies? How do regulatory conditions in both economies influence the pace and scope of AI integration? What are the main benefits and challenges, including the regulatory impact, in each country? What socio-economic implications does AI-driven innovation present for Ghana and Kazakhstan?

This paper aims to enhance understanding of the opportunities and challenges associated with adopting AI in developing economies and to provide policy recommendations for governments, financial organizations and stakeholders. It expands the understanding of how AI-driven technologies could be leveraged in diverse emerging market contexts, thereby highlighting the importance of AI technologies in the financial sectors of developing economies. Given that financial systems of different countries are becoming increasingly interconnected, the experiences of Ghana and Kazakhstan can provide fresh insights for other developing nations (Bell & Hidary, 2024).

Artificial intelligence (AI) has transformed the financial landscape, changing the way financial transactions are carried out, markets are operated, and risks are managed. The theoretical framework of AI in financial services represents a paradigm shift in the way that banking and financial organizations operate, particularly in developing economies (World Bank, 2025). According to a McKinsey & Company report (2023) on global banking, AI-driven technologies are projected to generate $200–300 billion annually in emerging market banking sectors. The concept underlying the adoption of AI-driven innovations in finance encompasses machine learning, natural language processing and predictive analytics. These innovations are fundamentally transforming conventional banking models (Pattnaik et al., 2023).

AI-driven innovations have significantly improved customer experience, operational efficiency, and decision-making processes in the banking sector. Major elements of the AI-powered tool concept include fraud detection systems, credit scoring, chatbots, and personalized financial and investment advice. Davenport and Ronanki (2018) reveal that AI-based chatbots such as Erica (Bank of America) and Eno (Capital One) have enhanced customer engagement by providing 24/7 support and personalized recommendations. Fraud remains one of the biggest problems in the financial sector, but it is now widely mitigated using machine learning algorithms that can detect fraudulent activities in real time. Ngai et al. (2011) demonstrate how Mastercard uses AI to analyse operational patterns and detect anomalies. AI-powered tools, particularly those based on neural networks, have enhanced the precision of credit scoring by employing non-traditional data sources, such as social media activity and use of mobile phones (Lessmann et al., 2015), when determining the credit score of customers.

AI-powered innovations are also used to manage investments; they transform investment management through algorithmic trading, portfolio optimisation and predictive analytics. AI-based trading algorithms analyse vast amounts of market data in order to execute trades at the most opportune moments. Biais et al. (2015) argue that these tools are used by high-frequency trading companies, such as Renaissance Technologies, to gain a competitive edge. Some banking platforms have adopted various Robo-advisors in their activities. For example, D’Acunto et al. (2019) state that companies such as Betterment and Wealthfront use AI-powered robo-advisers to provide automated, low-cost investment recommendations based on customer preferences and risk tolerance. Meanwhile, Chen et al. (2019) have demonstrated that AI-driven models can be used to predict market trends and asset prices based on historical data, news sentiment analysis, and macroeconomic factors.

In risk management, AI-driven technologies are used to assess related risks and develop mitigation approaches. AI has enhanced the methods of assessing and mitigating risks by providing more precise modelling and real-time monitoring. According to Barboza et al. (2017), AI-powered techniques such as deep learning have improved the accuracy of credit risk models by identifying complex, non-linear relationships in data. Also, transactional risks are managed with the help of AI-based models, which are used to monitor operational processes to detect potential risks, e.g. system failures or breaches of compliance. AI tools can be used to simulate market scenarios and predict potential losses in adverse conditions. This can help with stress testing and capital distribution Kou et al. (2014).

One major policy in the financial industry is ensuring regulatory compliance when providing financial services, both domestically and internationally. AI-powered innovations can be used to modernize compliance processes by automating tasks like know-your-customer (KYC) verification, anti-money laundering (AML), and others. Sironi (2021) explains how AI-powered systems analyze operational patterns in order to detect suspicious transactions and reduce false positives. Similarly, Arner et al. (2016) demonstrate how AI models verify client identities using facial recognition and document analysis. This has led to reductions in both onboarding time and costs. Deloitte (2020) also shows that natural language processing (NLP) tools can extract vital information from unstructured data in order to generate regulatory reports.

AI-powered innovations have transformed financial markets by improving liquidity, price discovery, and market productivity. AI algorithmic tools provide liquidity by consistently quoting bid and ask prices, thereby reducing bid-ask spreads (Hendershott & Riordan, 2013). According to Loughran and McDonald (2016), AI tools are effective in analysing new articles, social media activity and earnings calls to measure market sentiment and predict price levels. Gomber et al. (2017) also support this view. AI-based systems facilitate smart order routing, optimising trade execution by routing orders to the most favourable locations based on real-time market conditions.

Despite its numerous advantages, the adoption of AI in the financial sector faces difficulties related to data privacy, regulatory uncertainty, and algorithmic bias. For example, Zarsky (2016) points out that using personal information in AI tools raises privacy and data security concerns. Mehrabi et al. (2021) also warn that AI-powered models can perpetuate biases arising from the data training processes, which may result in unfair credit scoring or hiring decisions. The Financial Stability Board (FSB, 2024) indicates in its 2024 report that the absence of clear and consistent regulatory guidelines for AI in finance poses threats to compliance and accountability.

Theoretical and conceptual framework of AI adoption in the financial sector

The adoption of AI-powered solutions in the financial sector is supported by numerous theoretical works that describe the processes, drivers and outcomes of technological innovation. Examples include diffusion innovation theory (Rogers, 2003), the technology-organization-environment (TOE) theory (Tornatzky & Fleischer, 1990) and institutional theory (Meyer & Rowan, 1977).

Fig. 1 presents Roger’s (2003) Diffusion of Innovations theory, which provides a foundation for understanding how new technological models and applications, such as AI-powered solutions, are adopted and diffused in communities.

Figure 1. 

The Diffusion Innovation Theory. Source: Aygül et al. (2015).

According to this theory, the adoption of innovations is determined by the relative benefits obtained, compatibility, trialability, complexity and observability. In the case of AI-powered applications in the financial sector, these factors help to explain why some institutions and countries adopted AI earlier than others. For example, the potential of AI-driven solutions to enhance efficiency and reduce costs could encourage their adoption, while the perceived complexity or incompatibility with existing systems could hinder it. Thus, the early adoption of AI in Ghana’s financial sector is evident in customer-facing applications such as chatbots: Leo and Kukua, developed by United Bank of Africa (UBA) and Fidelity Bank respectively, implemented to improve customer service, followed a bottom-up pattern driven by private sector initiatives. In Kazakhstan’s financial sector, the adoption of AI exhibits a top-down diffusion model involving active government participation through the Generative Nation concept.

The Technology-Organisation-Environment (TOE) framework, developed by Tornatzky and Fleischer (1990), identifies three contexts that facilitate the adoption of technological innovations: technological, environmental, and organizational, as illustrated in diagram 2. In the financial sector, the technological context relates to the availability of AI tools and infrastructure. The organizational context includes elements such as firm size, resources, and leadership. The environmental context encompasses market competition, regulatory frameworks, and customer expectations. The TOE theory is useful for analyzing the adoption of AI in Ghana and Kazakhstan, as it outlines the interplay between technological potential, organizational readiness and external environmental conditions.

Ghana faces a variety of challenges, ranging from technological constraints, such as limited digital infrastructure, to varying levels of readiness, such as limited technical capacity among adopters, and limited specific AI policies, which create uncertainty for financial institutions (Quaye et al., 2024). Meanwhile, Kazakhstan is investing heavily in technological infrastructure such as supercomputing and 5G networks, actively developing governance frameworks, and coordinating organizational transformation with state initiatives such as the International AI Centre in Astana and financial institutions (Geneva Internet Platform digWatch, 2025).

The institutional theory by Meyer and Rowan (1977) highlights the importance of regulatory policies, norms, and cultural expectations in forming organizational behavior (Scott, 2014). In the context of adopting AI, institutional theory explains how regulatory frameworks, societal attitudes and industry standards influence its adoption and integration into the financial sector. For example, the development of regulatory sandboxes in Ghana and the establishment of the Astana International Financial Centre (AIFC) in Kazakhstan demonstrate the influence of institutional factors on the adoption of AI. While Kazakhstan is proactively establishing comprehensive governance frameworks, such as the AI Law and standards for AI content labelling in financial services, Ghana currently lacks strong regulatory policies specifically for AI in finance.

The conceptual framework for this paper integrates the DOI theory, TOE framework, and institutional theory to analyze the adoption of AI in the financial sectors of Ghana and Kazakhstan, showing how technological, organizational, and environmental factors are intertwined and collectively determine the pace, extent and outcomes of the adoption of AI in the financial sector. The framework also includes the socio-economic implications of AI adoption, particularly with regard to financial inclusion, employment, and regulatory oversight.

Although Ghana and Kazakhstan have made significant strides in adopting and integrating AI-powered solutions to achieve their respective economic agendas, the pace and extent of this adoption varies. Among the obstacles that prevent the two countries from realising the AI full potential in their financial sectors are inadequate infrastructure, absence of a comprehensive regulatory framework on data privacy, and insufficiently skilled workforce

Ghana and Kazakhstan are located in different regions, both economically and geographically (West Africa and Central Asia, respectively), and have varying levels of socioeconomic development, regulatory policies, and infrastructure. By comparing these two economies, the paper aims to contribute to a wider understanding of how regional contexts could influence the adoption of AI in the financial sector. Ghana and Kazakhstan were selected for this comparative analysis due to their similar contexts and challenges, and regional representation. The study applies this framework to explore AI application in the countries’ regulatory environment and its broader implications for economic growth and financial inclusion in both countries.

This paper employs a comparative approach in order to achieve in-depth understanding of the impact of AI-powered technologies on the financial sectors of Ghana and Kazakhstan. The analysis is carried out in accordance with the theoretical and conceptual framework of AI implementation in the financial sector; it involves private sector initiatives, major technologies, infrastructure, regulatory frameworks, and market dynamics (Table 2). This methodology is adopted to determine the degree and extent of the efforts and progress made by each of the countries.

Figure 2. 

The framework of Technology-Organisation-Environment (TOE). Source: (Yong, 2023).

From the table above it is clear that both Ghana and Kazakhstan have made great progress in using AI and digitalization for their economic development. Despite the differences in the initiatives, infrastructure, technologies and regulations employed to achieve their economic development agendas, they are reaping similar benefits.

Ghana’s financial sector has made a digital breakthrough: according to the Bank of Ghana (2024) mobile money penetration is at 84% of the overall adult population. Ghana’s fintech ecosystem has had an annual growth of about 15% since 2020, with AI-based innovative solutions playing a significant role in this development. The introduction of AI-powered chatbot by the Ghana Commercial Bank (GCB) has minimized the time of customer service response by over 60% (Mhlanga, 2024).

As concerns regulatory frameworks and innovations, Asante and Akowuah (2023) point out that AI-powered financial products were tested with the help of Bank of Ghana’s Regulatory Sandbox. Innovative solutions were then introduced in credit scoring and fraud detection. It has been estimated that about 45% of licensed fintech companies in Ghana use AI-driven technologies in their activities.

The financial sector of Kazakhstan has shown notable advancement in digital transformation, and Astana International Financial Centre (AIFC) has become the hub for fintech innovation (National Bank of Kazakhstan, 2023). Between 2020 and 2023 about $500 million has been invested in the country’s banking sector to promote AI-powered technologies. The results of a survey conducted by the National Bank of Kazakhstan (2024a) show that 31% of financial market participants - including second-tier banks, microfinance organizations, insurance companies, and securities market participants - use AI in their operations. Among them, second-tier banks are the most active in adopting AI, with a usage rate of 60%.

Kazakh banks have been early adopters of AI technologies in Central Asia. According to Business and Finance Consulting (2024), these technologies helped Kaspi bank in its credit scoring by processing more than 1 million applications monthly. Furthermore, AI implementation in anti-money laundering systems enhanced detection capacity by 35% (Hall, 2024).

Digitalization and the adoption of AI have provided considerable economic benefits by increasing financial inclusion (African Banker, 2023) and reducing operational and processing costs. For instance, Deloitte (2023) integrated AI in its operations and thus reduced its operational expenses through automation of routine tasks and improvement in decision-making process. In Ghana, the introduction of AI-driven chatbots and automated customer services systems led to an over 20% reduction in the labor costs (Krijnsen et al., 2023). In a similar way, Kazakhstan’s banks used AI-based systems to reduce their processing costs by about 15% (National Bank of Kazakhstan, 2024a).

AI technologies have improved financial inclusion in developing countries by offering access to digital financial services to the population groups that were previously underserved by conventional banking systems (World Bank, 2020). The number of users of digital banking services in Kazakhstan, particularly in remote communities, rose by over 25% according to the Hall (2024). MTN Mobile Money (2023) reported that the integration of AI on mobile money platforms led to increased access to financial services in rural areas by 30%. AI-based technologies have also encouraged competition among financial institutions, which is beneficial to the financial sector as it makes it more vibrant and up to date. According to McKinsey & Company (2023), the development of AI improved market competitiveness, helping financial organizations to develop and offer personalized products and services to their clients. Grzybowski et al. (2023) in their study of Ghana found out that the country’s fintech startups increased their market share by 10% thanks to using AI-powered technologies to develop innovative solutions. Zhumadilova et al. (2023) showed that the introduction of AI-enabled analytics enhanced customer segmentation thereby promoting competitive advantage.

Although the integration of AI caused job displacement in some industries, in others it promoted employment opportunities in tech-driven roles (World Economic Forum, 2024). In Ghana, data analytics and AI-related employment rose by 40% (Asante & Akowuah, 2023). Kazakhstan’s financial sector also witnessed a shift towards tech-oriented jobs (Deloitte Kazakhstan, 2022).

AI has improved access to financial services by easing processes and reducing barriers (World Bank, 2024). AI-driven credit scoring models have helped more people to access loan facilities according to GCB Bank PLC (2023). Digital platforms have simplified the process of opening accounts and increased accessibility in Kazakhstan (National Bank of Kazakhstan, 2024a) and AI-powered virtual assistants have improved customer satisfaction (Zhumadilova et al., 2023). In Ghana, AI-based innovations have transformed customer service by providing 24-hours support and personalized interactions, with 60% of customer inquiries handled by AI chatbot (Deloitte, 2023; Krijnsen et al., 2023). McKinsey & Company (2023) claim that the integration of AI in the financial sector has increased product innovation, which caused the development of personalized financial products. AI is used by banks and fintech businesses in Ghana to create customised saving plans (Coffie & Hongjiang., 2023), and in Kazakhstan to provide investment advisory services (Hall, 2024). Fraud detection is another major advantage brought about by AI, which significantly improves risk management by facilitating compliance monitoring (Krijnsen et al., 2023). Bank of Ghana (2024) stated that the adoption of AI-driven systems helped it reduce fraud by 25%; according to the National Bank of Kazakhstan (2024a) AI allowed financial organizations to strengthen their risk management capabilities.

Despite the progress made, digital transformation remains a challenge, with gaps in digital and AI literacy as well as in access to the internet (ITU, 2024). Rural Ghana is lagging in the adoption of AI because of limited infrastructure (World Bank, 2024); similar gaps exist in digital service between rural and urban areas in Kazakhstan (Hall, 2024). The lack of cultural acceptance and understanding in both countries can lead to low trust in AI technologies and thus impede the adoption of AI (MTN Mobile Money, 2023; Hall, 2024)

The comparative analysis of the adoption of artificial intelligence (AI) in the financial sectors of Ghana and Kazakhstan shows the two countries’ similar challenges and prospects, which also reflect broader dynamics of technological innovation in developing countries. This paper highlights the need for studying the socioeconomic impact of AI in developing economies and particularly in Ghana and Kazakhstan, which are different in terms of geography and culture but face similar challenges in AI adoption.

The paper has explored the drivers, processes, and implication of the adoption and integration of AI into the economies of these nations based on the diffusion of innovation theory, Technology-Organisation-Environment framework and institutional theory. It shows that AI has serious potential to improve financial inclusion, refine operational efficiency, and drive economic development; at the same time, it outlines the obstacles that ought to be addressed to achieve the full potential of AI.

The study reveals the drivers of AI-powered solutions in these countries. In Ghana the adoption of AI was driven by the willingness to improve financial inclusion and operational efficiency. It was based on successful development and implementation of mobile money services and their interoperability with banks. The regulatory sandbox initiatives by the central bank of Ghana fostered innovation, encouraging financial institutions to experiment with AI-driven instruments, such as chatbots, fraud detection systems, and algorithms for credit scoring. In Kazakhstan, the government implemented the flagship Digital Kazakhstan program and created the Astana International Financial Centre (AIFC), making the country a regional leader in fintech innovation. Under a broader economic diversification agenda of the country, banks in Kazakhstan have developed and implemented sophisticated AI tools, such as predictive analytics and algorithmic trading.

The factors that limit both countries’ capacity of integrating AI and realising its full potential, include inadequate technological infrastructure, data security concerns, and shortage of skilled AI workforce. The lack of comprehensive regulatory framework for data protection and cybersecurity remains an impediment to AI adoption and integration in Ghana. Kazakhstan’s major difficulties are related to regulatory enforcement and harmonisation, and workforce upskilling.

The legal environment in both Ghana and Kazakhstan are indicative of their respective priorities and challenges. Legal sandboxes intended to facilitate innovation and ensure consumer protection are introduced in Ghana to promote financial inclusion. In Kazakhstan, a focus on economic diversification and the country’s leading position in the region has led to the development of more robust regulatory policies.

The adoption of AI in the financial industries of Ghana and Kazakhstan has both short- and long-term implications for economic growth, financial inclusion, and regulatory oversight. In Ghana, AI-powered tools are expected to further improve access to financial services for the unbanked and underserved people. The adoption of AI-driven innovation has positioned Kazakhstan as a regional hub for fintech innovation attracting investment and talent from across the region.

This paper contributes to the theoretical understanding of AI adoption in developing economies by integrating the most useful theories such as DOI theory, TOE framework and institutional theory. The comparative study of Ghana and Kazakhstan reveals the interplay between technological, organisational and environmental elements influencing AI adoption, and offers a nuanced perspective on the drivers of and obstacles to technological innovation in the financial sector.

The results provide useful insights for policymakers, financial institutions and other stakeholders. The study highlights the importance of creating conditions for adopting and integrating AI. This requires comprehensive regulatory frameworks for cybersecurity and data protection, promotion of digital skills and establishment of legal sandboxes to facilitate innovation growth. Financial institutions need to create technological infrastructure, upskill the labour force, and integrate AI-driven innovations that align with the national priorities and customer needs.

It is therefore recommended that Ghana concentrates on building foundational AI infrastructure and governance framework to boost trust; Kazakhstan should use its national AI strategy to introduce more effective AI applications in its financial sector; and both countries should try to balance innovation with governance in order to maximise AI benefits for financial inclusion and stability and to spur economic growth and performance.

The creation of an ecosystem for effective AI adoption requires collaboration of governments, financial organisations and technology providers. It is also important to work with societal attitudes and build trust in AI-driven financial services. Finally, the socioeconomic impact of adopting AI in terms of employment and financial inclusion should be monitored and addressed when necessary.

This research opens many directions for future studies. First, further research into socioeconomic impact of AI adoption, especially in terms of employment and income inequality, could offer better understanding of the trade-offs involved. Next, comparative analysis of other developing countries’ experience may help identify best practices and lessons to be learned. Finally, conducting longitudinal studies to track the evolution of adopting AI in Ghana and Kazakhstan could offer insights into the long-term outcomes and sustainability of such initiatives.