The authors analyze global large language model (LLM) adoption patterns of U.S. and Chinese models and explore three key drivers of international LLM adoption: pricing strategies, multilingual capabilities, and government-led diplomacy initiatives. The authors aim to provide insights to policymakers, technology leaders, and industry observers who seek to understand the evolving U.S.-China competition for artificial intelligence supremacy.

U.S.-China Competition for Artificial Intelligence Markets

Analyzing Global Use Patterns of Large Language Models

Austin Horng-En Wang, Kyle Siler-Evans

ResearchPublished Jan 14, 2026

The authors analyze global large language model (LLM) adoption patterns, with a focus on the competitive dynamics between the United States and China. Using website traffic data across 135 countries from April 2024 through May 2025, they tracked site visits to major U.S. and Chinese LLM platforms to assess market penetration, identify geographic adoption patterns, and examine the impact of the January 2025 DeepSeek R1 launch. The authors explore three key drivers of international LLM adoption: pricing strategies, multilingual capabilities, and government-led artificial intelligence (AI) diplomacy initiatives. The authors aim to provide insights to policymakers, technology leaders, and industry observers who seek to understand the evolving U.S.-China competition for AI supremacy.

Key Findings

  • Global LLM use is growing rapidly; site visits to major LLM platforms increased threefold from April 2024 to August 2025, rising from an estimated 2.4 billion to nearly 8.2 billion monthly visits.
  • U.S. models have maintained overwhelming market dominance and captured approximately 93 percent of global LLM site visits in August 2025.
  • Site visits to China-based LLMs increased by 460 percent in just two months. The rise of DeepSeek did not cannibalize traffic to other Chinese models, which continued their upward trajectories throughout 2025.
  • Chinese LLMs' global market share surged from 3 percent to 13 percent in two months, mostly carried by DeepSeek, even as the website traffic for U.S. LLMs continued to increase steadily during this period.
  • Chinese models captured more than 10 percent penetration in 30 countries and 20 percent of market share in 11 countries. Gains were most pronounced in developing countries and countries with close political and economic ties to China.
  • Chinese models are one-sixth to one-fourth the cost of U.S. rivals. However, with ubiquitous free-tier offerings, most users never directly encounter these price differences.
  • Although U.S. models have historically supported more languages, Chinese LLMs have largely closed this gap.
  • In the area of AI diplomacy, China engages more countries earlier and more frequently than does the United States. Although this may be significant for government-to-government partnerships or large corporate deals, the authors are skeptical that embassy activities meaningfully influence the choices of everyday users, which are likely to dominate the use trends documented in this report.

Topics

Document Details

Citation

Chicago Manual of Style

Wang, Austin Horng-En and Kyle Siler-Evans, U.S.-China Competition for Artificial Intelligence Markets: Analyzing Global Use Patterns of Large Language Models. Santa Monica, CA: RAND Corporation, 2026. https://www.rand.org/pubs/research_reports/RRA4355-1.html.
BibTeX RIS

Research conducted by

This research was independently initiated and conducted by the Center on AI, Security, and Technology within RAND Global and Emerging Risks using income from operations and gifts and grants from philanthropic supporters.

This publication is part of the RAND research report series. Research reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND research reports undergo rigorous peer review to ensure high standards for research quality and objectivity.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.