The recent release of the “Government AI Readiness Index 2025” by Oxford Insights has placed renewed focus on evolving levels of national preparedness in artificial intelligence (AI). In this context, the World Bank’s “Digital Progress and Trends Report 2025: Strengthening AI Foundations” offers a critical analytical lens for understanding the global trends and structural challenges shaping these outcomes.

Aygun Ahmadova, Acting Head of the Digital Trade Hub of the Center for Analysis of Economic Reforms and Communication, stated that the report presents a systematic analysis of global AI development trends, the challenges countries face in this field, and the emerging opportunities, and that it puts forward practical and policy-oriented approaches, particularly for developing countries, in terms of the effective adoption and scaling of artificial intelligence.

The core thematic line of the report is the importance of the foundational “4C” pillars of AI readiness. This approach enables developing countries to view artificial intelligence not merely as a technological innovation, but as a tool for economic, social, and institutional transformation, while clearly identifying the baseline conditions that must be established for countries to derive tangible benefits from AI.

The foundational “4C” pillars of AI readiness bring together four interdependent elements:

Connectivity - connectivity, including energy and digital infrastructure;

Compute - computing power based on AI chips, data centers, and cloud technologies;

Context - context consisting of data, training resources, models, and applications;

Competency - competencies encompassing digital skills and human capital.

Together, these elements form the foundation of inclusive and effective AI ecosystems and enable countries to adopt, adapt, and innovate with artificial intelligence in a responsible manner.

The “4C” approach highlights the necessity for developing countries to consider artificial intelligence not only as a technological application, but as a strategic and institutional transformation tool. Achieving real economic and social benefits from AI requires a balanced and coordinated approach between energy and digital infrastructure, computing power, high-quality and locally relevant data resources, and the development of human capital. Once these foundations are in place, artificial intelligence can become not a passive consumption of foreign technologies, but a driving force for innovation aligned with local priorities and for sustainable development.

The report emphasizes the priority of global coordination and investments to strengthen the “4C”s. Within this framework, it particularly underlines that when investments across the “4C”s are not planned as a unified system but instead implemented in isolation, artificial intelligence fails to deliver the expected results.

According to Aygun Ahmadova, improving connectivity or increasing computing power alone is not sufficient for AI development. The absence of locally relevant data and weak human capital can prevent these investments from translating into tangible outcomes. For this reason, the report recommends strengthening the “4C”s in parallel and in stages, and establishing a balance between the approaches of “adopt,” “adapt,” and “innovate,” depending on a country’s level of readiness.

The report also notes that, for developing countries, success in artificial intelligence lies not only in competing at the global level, but in targeted applications in sectors where local needs are high, such as public services, agriculture, education, healthcare, and energy. From this perspective, investments in the “4C”s are assessed not merely as technological modernization, but as a fundamental condition for strengthening economic resilience, social inclusion, and institutional capacity.

It is noted that priorities cannot be the same for countries at different stages of AI readiness, and therefore investment decisions should be aligned with existing national capacities. To better understand a broader and more systematic approach, the report presents a framework for “prioritizing investments based on levels of AI readiness.” This framework helps countries build AI policy on phased, balanced, and realistic foundations by indicating which steps are most appropriate at each stage in terms of connectivity, computing power, data context, and skills.

For this reason, the “4C” approach should be adopted as one of the core pillars of AI policy in Azerbaijan. The priority for the country is not the scale of technology, but the depth of its use and its level of penetration into society. When energy and digital infrastructure, computing capabilities, local data ecosystems, and skills develop together, artificial intelligence can become a real policy instrument that accelerates economic growth, improves the quality of public services, and supports sustainable development.

According to the approach presented in the report, in the AI era, a leading position will primarily be attained by actors that have access to extensive data resources, high computing power, and advanced cloud infrastructure.

Thus, as a result of investments directed toward generative artificial intelligence, AI chips, data centers, and computing power (compute) are becoming key strategic areas. The growing demand for computing power strengthens the positions of both infrastructure providers and cloud service providers that distribute large language models.

At the same time, the report notes that developing countries face a shortage of companies capable of supplying local text, image, audio, and other digital data used for training and developing AI systems. This situation is reflected in the extremely low share of these countries in venture capital investments and startup activity in this field. It is stated that in 2023, 56 percent of global venture capital investments related to AI system training and development went to the United States alone, 17 percent to the European Union and the United Kingdom, while the rest of the world was represented by only 6 percent. The distribution of startups reflects a similar picture: the United States leads with a 40 percent share, China follows with 6 percent, and India lags significantly behind with a 2 percent share. These indicators demonstrate the weak development of local data resources, relevant businesses, and innovation ecosystems in developing countries.

The report emphasizes that this situation creates serious inequality within the AI value chain and causes many countries to participate mainly at the stage of using technology, rather than at the stages of creating and adapting it. Limited availability of language- and locally contextualized data, weak digital footprints, and the fact that data is largely controlled by global platforms further deepen this problem.

The statistical indicators presented in the report clearly highlight the key issues that should be kept at the center of Azerbaijan’s AI policy. For the country, the priority should not only be the application of ready-made AI solutions, but also the creation of digital data in the Azerbaijani language and local context, the structuring of public and private sector data, and the promotion of local companies and startups engaged in data collection and processing. This approach can enable Azerbaijan to secure a long-term and sustainable position in the AI ecosystem.

In this context, the report specifically emphasizes that the adaptation of AI models to local conditions is of decisive importance for developing countries. AI models developed at the global level often fail to take into account local economic priorities, institutional characteristics, linguistic and cultural contexts, and resource constraints. Therefore, effective use of AI in areas such as healthcare, education, agriculture, and public services is possible only through adapting existing models to local needs or developing national solutions based on open-source models.

The report notes that open-source AI models create significant opportunities for developing countries. Open-source models enable adaptation to local languages, sectors, and governance needs without high licensing costs, while also reducing risks of technological dependency. This approach is of particular strategic importance for Azerbaijan. AI policy in the country should focus on building local adaptation capabilities based on open-source models, creating Azerbaijani-language and sectoral data repositories, and strengthening cooperation between the state, startups, and the scientific community. This is one of the key conditions for Azerbaijan to transform from a mere consumer of AI into a regional actor that adapts and generates innovation.

The report states that among more than 900 significant AI models released from the 1950s to the present day, 62 percent of the leading contributions come from the United States, 25 percent from other high-income countries, 13 percent from China, only 0.2 percent (just 2 models) from India, and 0.1 percent (just 1 model) from Argentina.

While the United States accounts for 26 percent of AI-related scientific publications in high-income countries and 35 percent of global citations, China accounts for 70 percent of publications in the upper-middle-income country group, and India accounts for 68 percent in the lower-middle-income group.

This disparity shows that although scientific activity is conducted across a broader geography, the development of high-impact models especially technologies such as large language models that require high computing power, extensive data resources, and significant financial investment, is concentrated in a limited number of countries with strong industrial sectors. As a result, dominance in the AI ecosystem is determined not only by scientific knowledge, but by industrial infrastructure and resources capable of transforming that knowledge into large-scale products and global applications.

At the same time, the report provides a balanced analysis of the potential impacts of AI on the global economy and labor markets, particularly the changes it may create in human capital, the value of skills, and employment structures. Alongside expanded access to education and healthcare, risks related to labor market polarization and the availability of quality jobs are also highlighted. Although AI creates new professions and jobs, these opportunities are not evenly distributed across the labor market and contribute to polarization. It is noted that high-income and highly skilled AI professions are concentrated mainly in a small number of companies, cities, and countries, while new roles related to generative AI have not emerged at the expected scale. At the same time, the rapid growth of low-skilled, low-income, short-term “gig”-type jobs (non-contractual, project- and task-based work via platforms) is deepening structural inequality in the labor market.

In this context, the key task is to ensure that the productivity gains created by AI translate into real and sustainable employment opportunities for a broader population, rather than benefiting only a limited group. Governments should realign education and labor market policies to strengthen workforce preparation in areas such as digital and analytical skills, data literacy, working with AI systems, their application and technical support. This should cover not only highly specialized AI experts, but also mid-level technical specialists, digital service providers, and application-oriented professions.

At the same time, citizens should focus on acquiring new knowledge and competencies-such as skills in using AI tools, digital services, technical and applied professions, as well as creative and problem-solving-oriented fields-to adapt to the changing labor market. This approach can ensure that the opportunities created by AI are not limited to narrow, high-income professional groups, but instead contribute to broader employment and social inclusion.

Aygun Ahmadova added that artificial intelligence has already become an integral part of development policy, and its impact on economic growth, social welfare, and environmental sustainability directly depends on how countries manage the technology. The thematic case studies presented show that while AI creates significant opportunities in areas such as digital services, education, agriculture, and energy, without an inclusive approach and targeted government support these opportunities do not translate into broad public benefits. The report calls on governments to develop proactive policies rather than reacting passively to AI-driven changes. Strengthening energy and digital infrastructure, investing in human capital, improving data governance, and enhancing institutional readiness are presented as key priorities. The overarching objective is to ensure that artificial intelligence becomes not a new source of inequality, but an accelerator of inclusive and sustainable development.