Report Overview
"A Blueprint for Building National Compute Capacity for Artificial Intelligence" is a comprehensive report published by the OECD in February 2023. The report provides the first blueprint for policy makers to help assess and plan for the national AI compute capacity needed to enable productivity gains and capture AI's full economic potential.
Key Insight: Many countries have developed national AI strategies without fully assessing whether they have sufficient domestic AI compute infrastructure and software to realize their goals. This policy blind-spot may jeopardize domestic economic objectives and create "AI compute divides" between countries and within national AI ecosystems.
Key Data Points
Key Insights Summary
AI Compute Definition and Importance
The report defines AI compute as "one or more stacks of hardware and software used to support specialized AI workloads and applications in an efficient manner." It distinguishes AI compute from general-purpose compute and highlights its critical role in AI development and deployment.
Exponential Growth in AI Compute Demands
The computational capabilities required to train modern machine learning systems, measured in floating-point operations per second (FLOPS), has grown by hundreds of thousands of times since 2012, despite algorithmic and software improvements that reduce computing power needs.
Emerging AI Compute Divides
An imbalance in compute resources risks reinforcing socioeconomic divides, creating further differences in competitive advantage and productivity gains. Private sector initiatives increasingly benefit from state-of-the-art AI compute resources compared to public research institutes and academia.
Three-Dimensional Framework for National AI Compute Plans
The blueprint provides guidance for policy makers on how to develop a national AI compute plan along three dimensions: capacity (availability and use), effectiveness (people, policy, innovation, access), and resilience (security, sovereignty, sustainability).
Measurement Challenges
Standardized measures of national AI compute capacity remain a policy gap. The report identifies obstacles to measuring and benchmarking national AI compute capacity across countries and proposes indicators and frameworks to address these challenges.
Global Policy Initiatives
The report documents various national and regional initiatives focused on high-performance computing, cloud-based services, and supply chain security for AI compute components, highlighting different approaches taken by countries to build AI compute capacity.
Content Overview
Document Contents
Abstract
Artificial intelligence (AI) is transforming economies and promising new opportunities for productivity, growth, and resilience. Countries are responding with national AI strategies to capitalize on these transformations. However, no country today has sufficient data on, or a targeted plan for, national AI compute capacity.
This report provides the first blueprint for policy makers to help assess and plan for the national AI compute capacity needed to enable productivity gains and capture AI's full economic potential. It provides guidance for policy makers on how to develop a national AI compute plan along three dimensions: capacity, effectiveness, and resilience.
Executive Summary
The demand for AI compute has grown dramatically for machine learning systems, especially deep-learning and neural networks. According to research, the computational capabilities required to train modern machine learning systems has multiplied by hundreds of thousands of times since 2012.
As governments invest in developing cutting-edge AI, compute divides can emerge or deepen. An imbalance of such compute resources risks reinforcing socioeconomic divides, creating further differences in competitive advantage and productivity gains.
This report offers a blueprint for policy makers to develop national AI compute plans aligned with national AI strategies and domestic needs. It takes stock of existing and proposed indicators, datasets, and proxies for measuring national AI compute capacity.
Introduction
Ensuring countries have sufficient AI compute to meet their needs is critical to capturing AI's full economic potential. Many countries developed AI plans without a full assessment of whether they have sufficient domestic AI compute to realize these goals.
The development of standardized measures for AI compute remains a policy and data gap. Policy makers require accurate and reliable measures of AI compute and how much national capacity they have, to make better-informed decisions and reap the full benefits of AI.
The OECD.AI Expert Group on AI Compute and Climate advances understanding and measurement of AI compute to help policy makers understand their AI compute needs and work towards addressing them.
Evolving Trends in Compute
The report analyzes trends in supercomputer performance and AI-specific compute demands. Few economies have supercomputers ranking as top computing systems, with emerging economies sparsely represented on the Top500 list.
State-of-the-art AI systems increasingly depend on high-performance compute. Researchers estimate that the computational capabilities required to train modern ML systems has grown by hundreds of thousands of times since 2012.
A compute divide can emerge and worsen between the public and private sectors because, increasingly, public sector entities do not have the resources to train cutting edge AI models. Industry, rather than academia, is increasingly providing and using the compute capacity and specialized labour required for state-of-the-art ML research.
Measuring AI Compute
The report defines AI compute as "one or more stacks of hardware and software used to support specialized AI workloads and applications in an efficient manner." This definition highlights several properties central to a common understanding of AI compute.
Measuring AI compute capacity and needs is particularly challenging. At present, very few tools and indicators exist to measure AI compute. Literature on AI compute typically focuses on the performance measurement of compute systems, such as application performance benchmarks like MLPerf or throughput benchmarks like the Top500 list.
The preliminary results of the public survey on AI compute launched by the Expert Group highlight measurement challenges. Of respondents, 52% reported challenges accessing sufficient AI compute, with cost (44%) cited as the top barrier.
Blueprint for National AI Compute Plans
The blueprint provides a framework for policy makers to develop national AI compute plans aligned with national AI strategies. It centers around three fundamental questions:
- How much AI compute does the country have?
- How much AI compute does the country need?
- How does it compare to other countries?
To answer these questions, policy makers can consider three overarching categories as part of a national AI compute plan:
- Capacity: What is the availability (supply) and use (demand) of national AI compute capacity?
- Effectiveness: How effectively is national AI compute capacity being used?
- Resilience: How resilient is a country's compute capacity (e.g., secure, sovereign, sustainable)?
Each category includes detailed subcomponents and considerations for policy makers.
AI Compute in National Policy Initiatives
Countries and regions take varying approaches to providing the digital infrastructure and access required for the development and use of AI. Different national goals for AI lead to different investment strategies.
The report documents various national initiatives, including:
- High-performance computing initiatives in countries like Canada, Chile, France, Germany, Japan, Korea, Slovenia, Spain, the UK, and the US
- Cloud-based services such as GAIA-X in the EU
- Supply chain initiatives focused on securing semiconductor manufacturing
While several countries have broader national HPC or cloud resources initiatives, few national AI plans have specifically targeted initiatives for assessing national AI compute capacity and needs.
Gap Analysis and Preliminary Findings
The report identifies several key gaps in existing measurement tools and discusses preliminary findings:
- AI policy initiatives need to take AI compute capacity into account
- National and regional data collection and measurement standards need to expand
- Policy makers need insights into the compute demands of AI systems
- AI-specific measurements should be differentiated from general-purpose compute
- Workers need access to AI compute related skills and training
- AI compute supply chains and inputs need to be mapped and analyzed
Conclusion
AI is a general-purpose technology impacting nearly every facet of the global economy, prompting governments to formulate and publish national AI strategies. The successful implementation of national AI strategies could become one of the factors defining a country's ability to deliver innovation, productivity gains, and long-term growth.
However, many countries have developed AI plans without a full assessment of whether they have sufficient domestic AI compute capacity to realize these goals. Concerns are growing about reinforcing divides between those who have the resources to create and use complex AI models to generate competitive advantage and productivity gains, and those who do not.
Understanding of AI compute and its relationship to the diffusion of AI across OECD and partner economies can improve the implementation of national AI strategies, and guide future policymaking and investments.
Note: The above is only a summary of the report content. The complete document contains extensive data, charts, and detailed analysis. We recommend downloading the full PDF for in-depth reading.