This Report is the outcome of the work of the High-Level Panel of Experts on Artificial Intelligence (AI). The report is intended to serve as a reference for policymakers. The Panel was established by the mandate of the Finance Ministers and Central Bank Governors of the Group of 7 (G7), with the objective of exploring the implications of AI for the economy and the financial sector. The full implications of AI are unclear, and we know that there will be winners and losers. Opinions on the best approach to the technology will thus inevitably diverge. There were indeed some differences in views and emphasis among the members of the Panel in the preparation of this Report.
Similarly, G7 governments and central banks will likely prefer different approaches to AI, depending on culture, legal systems, national priorities, perception of risk and specific needs. There is no correct one-size-fits-all approach. AI has emerged as a transformative technology across a spectrum of activities with potential to lead to significant changes in economic structure and financial systems. Being a general-purpose technology with a potential for rapid innovation across a broad array of economic processes, AI calls for careful management and proactive policies. Governments have three roles to play in AI development: AI enablement (R&D, education, infrastructure, and financing); the use of AI in government itself; and the enactment of laws and regulations for the private sector, ensuring that the use of AI technologies facilitates governments’ objectives of economic growth, stability, equity and wellbeing. The speed of innovation and adoption of AI, its subsequent impact on the economy, labor markets, finance, and sustainability is unknown, but could be rapid.
As a result, policymakers should remain vigilant to AI developments and be prepared to adapt policy flexibly. AI already poses several urgent questions for policymakers. These include whether scaling laws will continue to hold and how that might impact concentration of market power; the trade-offs between proprietary and open-source systems; the availability of data for AI training; and the sustainability of AI business models due to the high associated capital and energy costs. Regulatory authorities should remain vigilant about market developments, as excessive concentration could reduce innovation and slow the spread of productivity gains. AI’s macroeconomic impact will show up in productivity and employment. Estimates of the productivity impact of AI vary widely between 0.1 and 1.2 percentage points per year, depending on AI’s influence on task automation and innovation. AI could affect employment through three main potentially interacting channels: a displacement channel, where AI-driven automation replaces jobs; a productivity channel, where AI increases overall efficiency; and a reinstatement channel, where AI creates new job opportunities, particularly in industries where human labor maintains a comparative advantage. Whether AI will complement or replace workers is an open issue. As AI capabilities advance, ever more sophisticated tasks could be automated, potentially increasing job losses in both cognitive and manual occupations and further exacerbating income and wealth inequalities.