IMF Conference

The 12^th Statistical Forum of the International Monetary Fund (IMF) will take place in hybrid format (in person and virtually) in Washington, D.C. from November 20 to 21, 2024. The Forum is a platform for policymakers, researchers, the private sector, regulators, and compilers of economic and financial data to come together to discuss cutting edge issues in macroeconomic and financial statistics and to build support for statistical improvements.

The theme of this year’s Statistical Forum is Measuring the Implications of Artificial Intelligence (AI) on the Economy. The increased use of AI presents both opportunities and challenges, with significant economic and societal implications. However, the impact of AI on the economy and society remains an evolving field of study. To understand these implications, governments, businesses, individuals require robust and comparable statistics.

The 12^th Statistical Forum will explore (i) the transformative potential of AI and where its impact will most likely be felt over the short to medium term, (ii) the impact of AI on jobs and productivity, (iii) the distributional implications of AI, (iv) how AI is being used by firms (including statistical agencies) and regulated by governments, and (v) some early attempts to produce official measures of the “AI industry”, “AI investment”, and the “use of AI”. The Forum will provide participants with an opportunity to share experiences and build on topics of mutual interest through presentations and panel discussions.

The IMF’s Statistical Forum Program Committee seeks proposals for research papers reporting on work or experiences on measuring the implications of AI. Contributions are solicited on the following topics of interest, as outlined in the different proposed sessions^[1]:

Session I: Where AI is headed, and which sectors will witness the greatest impact.

• As AI develops, its impact on various sectors is becoming increasingly profound and pervasive. The prospect of "turning over the keys" to increasingly autonomous systems raises many questions. This session will explore the rise of AI, its transformative potential, and where its impact will be the greatest.

Session II: AI’s implications on jobs and productivity.

• New AI tools can potentially change the way we work, raising productivity and boosting growth and incomes. This session will discuss how economies will respond as versatile robots and machine-learning systems potentially displace blue- and white-collar workers. This session will consider important questions about how AI will affect the labor market and productivity and the type of statistics that are needed to quantify these changes.

Session III: The distributional implications of AI.

• The development of wide-scale large language models requires substantial investment. A handful of firms control the world’s most complex large language models. The firms that control AI will most likely also ‘control’ the related income flows, resulting in highly concentrated income flows with important distributional implications for the economy and society. This session will explore the type of statistics required to understand the income flows associated with AI, to whom they are flowing, and where they are flowing.

Session IV: How AI is being used by firms and regulated by governments.

• Each day, firms are discovering new use cases for AI. The number of applications is widespread, from adjusting insurance claims, to diagnosing medical conditions and producing and analyzing data. At the same time, there is a recognition that this technology must be used responsibly. This session will explore the various use cases of AI, the latest thinking about the responsible use of AI, and the type of statistics that are needed to understand the evolving uses and their regulation.

Session V: Some early attempts to produce official measures of the “AI industry”, “AI investment”, and the “use of AI”.

• This session will highlight the work by the official statistical community to measure the size, scope, and structure of the “AI industry,” “AI investment”, and “AI use” and how the community adapts the internationally endorsed economic statistics reporting standards (such as the Balance of Payments Manual and the System of National Accounts) to capture this activity better.

[1] The sessions’ organization is currently tentative. Upon finalization of the agenda, the sessions will be better organized, and can be presently differently.