Artificial Intelligence in Science: Returns, Reallocation, and Reorganization

TL;DR

This paper examines how AI influences scientific research, primarily reorganizing resource allocation and team structures, with modest immediate impacts on outcomes, and highlights potential future productivity gains from large language models.

Contribution

It provides a comprehensive analysis of AI's role in science, emphasizing organizational changes over immediate efficiency improvements, using detailed proposal and publication data.

Findings

AI adoption correlates with modest short-term scientific outcome improvements.

AI-enabled projects reallocate resources toward human capital and larger teams.

Activities like ideation and experimentation expand in AI-enabled projects, aligning with large language model capabilities.

Abstract

Investment in artificial intelligence (AI) has grown rapidly, yet its returns to scientific research remain poorly understood. We study how AI reshapes the production of science using a comprehensive dataset of research proposals submitted to a large international funding agency, including both funded and unfunded projects. Combining keyword extraction with large language model classification, we identify the presence, type, and functional role of AI within each proposal and link these measures to detailed budget allocations, team structure, and subsequent publication outcomes. We find that, in the short run, AI adoption is associated with modest improvements in scientific outcomes concentrated in the upper tail. Instead, its primary effects arise in the organization of research: AI-enabled projects reallocate resources toward human capital, involve larger teams, and undertake a broader set of tasks. These patterns are consistent with a reorganization of the scientific production process rather than immediate efficiency gains, in line with theories of general-purpose technologies. Task-level analyses further show that activities expanded in AI-enabled projects, particularly ideation and experimentation, are increasingly compatible with large language model capabilities, suggesting potential for future productivity gains as these technologies mature.