AI may already be adding hundreds of billions to the economy—without showing up in the data

Artificial intelligence (AI) is rapidly emerging as a transformative force in the global economy, yet its economic impact remains shrouded in uncertainty due to a significant measurement gap. As AI technologies proliferate across various sectors, the challenge of accurately quantifying their contributions poses pressing questions for policymakers, businesses, and researchers alike. A recent policy brief by the Peterson Institute for International Economics sheds light on this issue, arguing that AI's economic activity could be vastly underestimated in current metrics.

The Measurement Gap in AI Economics

The conversation around AI's economic impact often contrasts starkly with traditional economic indicators such as productivity figures and GDP growth rates. While tech giants tout AI as a game-changing technology reshaping industries and enhancing productivity, the official data fails to reflect such sweeping transformations.

Anton Korinek and Patrick McKelvey, the authors of the Peterson Institute brief, argue that the very framework used to analyze economic performance does not accommodate the specific nature of AI advancements. They identify two primary challenges:

1. Fragmentation Across Industries: AI applications are scattered across various sectors, including cloud services, software, and data processing. This fragmentation makes it difficult to isolate AI's economic impact, as current statistics do not provide a comprehensive view of AI's contributions to the overall economy.

2. Rapid Improvement of AI Capabilities: The current economic metrics lack the means to account for the accelerated pace at which AI technologies are evolving. According to Korinek and McKelvey, AI is projected to generate approximately $250 billion in economic activity by 2025, comparable to the entire U.S. airline industry. They estimate that the industry's capacity to produce AI output is growing at an astonishing 2,600% annually. Additionally, the cost of achieving the same level of AI performance has plummeted by about 94% per year, significantly increasing the value derived from AI investments.

Building New Statistical Frameworks

To address this measurement gap, Korinek and McKelvey propose the establishment of a dedicated statistical framework for AI, akin to how governments track energy or international trade. This framework would aggregate AI activities across industries while adjusting for the rapid technological improvements characteristic of AI advancements. They assert that without accurate measurement, policymakers may struggle to make informed decisions regarding taxes, labor markets, and public spending, potentially resulting in misguided policies that overlook the true state of the AI economy.

Their analysis indicates that if official statistics adequately captured AI's contributions, the projected U.S. economic growth for 2025 could be about 4 percentage points higher than currently reported. However, the authors caution that this estimate represents an upper bound, emphasizing that it reflects the maximum plausible impact rather than a central estimate.

Diverging Opinions on AI's Economic Role

While Korinek and McKelvey present compelling arguments regarding the measurement gap, not all experts are convinced by their conclusions. Diane Coyle, a Bennett Professor of Public Policy at the University of Cambridge, acknowledges the existence of the measurement gap but questions the scale of AI's impact as suggested by the brief.

Coyle raises several points of contention:

• AI as an Intermediate Input: She argues that AI primarily serves as a tool used to enhance other products and services rather than functioning as a standalone product. Since GDP measures final goods and services consumed by end-users, the economic impact of AI may only be significant if it demonstrably improves the final product's quality or efficiency.

• Limited Evidence of Productivity Gains: Coyle emphasizes the lack of systematic evidence indicating that AI is consistently boosting productivity at the firm level. If one department integrates AI while others do not, the overall gains may stagnate, leading to bottlenecks that hinder broader organizational improvements.

Coyle's insights highlight a critical aspect of the ongoing conversation surrounding AI's role in the economy: the need for clarity in defining and measuring AI's contributions. As she succinctly states, the challenge lies not only in how we measure economic impact but also in reaching a consensus on what precisely should be measured.

Implications for Policy and Business Strategy

The implications of the measurement gap extend beyond academic debates; they resonate deeply within the realms of policy formulation and business strategy. As AI continues to evolve and disrupt various industries, understanding its economic impact is essential for crafting effective policies that support innovation while safeguarding public interests.

For policymakers, the absence of accurate data on AI's contributions could lead to misguided decisions about labor markets, taxation, and public spending. Without a comprehensive understanding of AI's economic landscape, governments may struggle to foster an environment conducive to growth and innovation.

For businesses, the challenge lies in navigating the complexities of AI integration and its implications for productivity. Companies that recognize the transformative potential of AI may find themselves at a competitive advantage, provided they can measure and leverage its contributions effectively. However, those that overlook AI's significance may risk falling behind in an increasingly tech-driven marketplace.

Real-World Examples of AI in Action

To illustrate the practical implications of AI's economic influence, consider a few real-world examples:

• Healthcare: AI-powered diagnostic tools and treatment optimization algorithms are revolutionizing patient care, leading to improved health outcomes and cost savings. For instance, AI systems can analyze medical images more quickly and accurately than human radiologists, enabling earlier detection of diseases such as cancer.

• Finance: In the financial sector, AI is being used to enhance risk assessment, automate trading, and streamline customer service through chatbots. These advancements lead to increased efficiency and reduced operational costs, contributing to the sector's overall productivity.

• Manufacturing: AI-driven automation is transforming manufacturing processes, allowing for more flexible production lines and reduced downtime. Smart factories equipped with AI technologies can optimize supply chain management and minimize waste, driving significant cost savings.

These examples highlight the diverse applications of AI across sectors, underscoring the need for a robust framework to measure its economic contributions effectively.

Looking Ahead: The Future of AI Measurement

As we move forward in this rapidly evolving landscape, the call for enhanced measurement of AI's economic impact becomes increasingly urgent. The need for comprehensive data is not merely an academic exercise; it is essential for shaping sound policies, guiding business strategies, and ensuring that the benefits of AI are realized across society.

Establishing a dedicated statistical framework for AI will require collaboration among governments, industry stakeholders, and academic researchers. By working together, these entities can develop methodologies that accurately capture the complexities of AI's contributions to the economy while addressing the challenges associated with rapid technological advancement.

Conclusion

The debate surrounding AI's economic impact is far from settled. While some experts argue for the recognition of AI's transformative potential, others caution against overstating its current contributions. As the discussion continues, one thing remains clear: bridging the measurement gap is crucial for understanding AI's role in shaping the future of work, productivity, and economic growth.

As policymakers and businesses grapple with the implications of AI, an accurate understanding of its economic contributions will be vital in navigating the complexities of the modern economy. By prioritizing the establishment of a dedicated statistical framework for AI, we can ensure that we are equipped to harness the full potential of this groundbreaking technology while addressing the challenges it presents.