Data Critique

Overview of the Dataset

Our data set includes the country, rank, and the 2024 index ranking of the government AI readiness. This data has become a trusted resource for policymakers and is referenced by leading organizations. The 2024 index explores this readiness by examining 40 indicators across three core pillars: Government, Technology Sector, and Data & Infrastructure. The Index asks the question: How ready are governments to implement AI in the delivery of public services? This aims to be a tool that supports evidence-based decisions that helps policymakers use AI to its best potential for the benefit of its citizens. To calculate the total score, they took the arithmetic mean of each dimension and the arithmetic mean of each pillar. The final score is the arithmetic mean of the three pillars. All indicators, dimensions, and pillars were weighted equally.

Methodology

Government Pillar

The methodology section provides the original sources for components organized by each pillar. For the Government Pillar, a combination of desk research and established indexes was used:

  • Vision Dimension: OECD AI Policy Observatory and UN IDIR AI Policy Portal
  • Government and Ethics Dimensions: Desk research, Worldwide Governance Indicators, Global Cybersecurity Index
  • Digital Capacity: Worldwide Governance Indicators, UN e-Government Survey, GovTech Maturity Index, World Economic Forum Executive Opinion Survey, Global Index on Responsible AI
  • Adaptability Dimension: Worldwide Governance Indicators, World Economic Forum Executive Opinion Survey, Global Data Barometer

Technology Pillar

The Technology Pillar relied primarily on worldwide rankings without additional desk research:

  • Maturity Index: CB Insights, UNCTAD, Global Innovation Index
  • Innovation Capacity: World Bank World Development Indicators, DealRoom, UNESCO, World Economic Forum Executive Opinion Survey, Scimago
  • Human Capital Capacity: UNESCO, GitHub Innovation Graph Data, QS Engineering & Technology Ranking, Network Readiness Index

Data & Infrastructure Pillar

The Data and Infrastructure Pillar combined desk research with worldwide indexes:

  • Infrastructure Index: UN E-Gov Telecommunications Infrastructure Index, Top500, EIU Inclusive Internet Index, GSMA Mobile Connectivity Index, World Economic Forum Executive Opinion Survey
  • Data Availability Dimension: Global Data Barometer, ITU, SPI GitHub Report, desk research with GovTech Maturity Index I-34 support
  • Data Representativeness Dimension: EIU Inclusive Internet Index, GSMA Mobile Connectivity Index

About Oxford Insights

The Government AI Readiness Index 2024 was made by Oxford Insights. This organization is a research institution located in the United Kingdom that studies how governments adapt and use technology, especially in the field of artificial intelligence. Oxford Insights is widely known for its expertise in the analysis of digital governance. The institution has worked with international organizations. The United Nations and the G20 referenced their report in global policy discussion. The report doesn’t mention any funders or sponsors. It only demonstrates the institution name as Oxford Insights and gives the credit to it. Starting from 2019, Oxford Insights provide an AI readiness report each year. The report gives information to support governments worldwide to help them understand how ready they are to use AI responsively and effectively. 

Data Limitations and Imputation

Some values were purposely excluded from the spreadsheet. Countries with data for less than half of the indicators were excluded from the final index. As a result, these countries do not appear in the final rankings: Democratic People’s Republic of Korea, Dominica, Micronesia (Federated States of), Monaco, Nauru, Palau, Tuvalu. For most indicators with some missing data, the value for each country was estimated using the average of its peer group, which was defined by both geographical region and World Bank income group. However, for 11 countries, this imputation was not possible because they were either the only country in their peer group or all countries in that group lacked data for that indicator. These countries are: Afghanistan, Algeria, Canada, Iran (Islamic Republic of), Iraq, Libya, Maldives, Seychelles, Syrian Arab Republic, United States of America, and Yemen. For these countries, no imputation was made where data for the entire peer group were missing.

Ideological Effects of the Dataset

The ideological effects of the way in which our sources have been divided into data are that our dataset turns a country’s readiness for artificial intelligence into scores. This approach makes it seem like a quantifiable standard, but in fact, it is not entirely so. Because it mainly focuses on the parts that can be statistically analyzed, such as infrastructure and government policies. However, these data also overlook some aspects that are difficult to measure, such as the public’s acceptance of AI. This is because it can better reflect the priorities of developed countries, while those countries with incomplete data are prone to being marginalized. So our dataset seems quite objective, but it might also give people the impression that a country’s progress means higher numbers, which also implies a better ranking.