What Real‑World Data Reveals About ChatGPT’s Global Adoption and Impact

Introduction

OpenAI, Duke University and Harvard University released a 63‑page research paper “How People Use ChatGPT”, the first systematic analysis of internal conversation logs since ChatGPT’s launch in November 2022.

User growth and activity

From November 2022 to July 2025 ChatGPT accumulated >700 million weekly active users (≈10 % of the global adult population). Weekly active users are measured for the three consumer plans (Free, Plus, Pro) in six‑month snapshots.

Message volume

Daily messages rose from 4.5 billion in June 2024 to 26 billion in June 2025, averaging 29 000 messages per second.

Demographic shifts

Gender balance shifted from ~80 % male at launch to a slight female majority by mid‑2025. Users under 26 account for ~50 % of messages, but usage among middle‑aged and senior users is increasing. Growth rates are highest in low‑ and middle‑income countries, indicating a global “down‑shifting” effect.

Conversation intents

Messages were classified into three intents: Inquiry, Execution, Expression. Overall distribution: 49 % Inquiry, 40 % Execution, 11 % Expression. Inquiry grew faster and showed higher interaction quality over time.

Top content categories

Practical guidance 28.8 %

Information search 24.4 %

Writing assistance 23.9 %

Work vs. non‑work usage

Non‑work conversations constitute >70 % of all messages, rising from 53 % in June 2024 to 73 % in June 2025. In work‑related dialogs, writing dominates (40 % of work messages), followed by practical guidance (24.1 %). Programming requests are 4.2 % and AI companionship 2.4 %.

Work‑related activity clusters

Two clusters cover ~60 % of work messages: (1) information acquisition, recording, interpretation; (2) decision‑making, consulting, problem‑solving, creative thinking. These map to O*NET occupational categories.

Role in the knowledge economy

ChatGPT functions as a “second brain” or research assistant, supporting information retrieval, drafting, and decision processes rather than replacing labor.

Key conclusions

Non‑work usage is growing faster than work usage.

Practical guidance, information search and writing dominate overall interactions.

Writing is the primary workplace application, mainly for editing existing text.

Programming and companionship needs are marginal.

Gender balance is approaching parity.

Younger users remain the majority, but the age gap is shrinking.

Low‑income regions exhibit the fastest adoption growth.

Highly educated professionals rely heavily on ChatGPT for decision support.

Economic impact

An estimate for 2024 suggests U.S. users would forgo an average consumer surplus of $98 per month without generative AI, amounting to at least $970 billion in annual consumer surplus for the United States.

Reference

Full report: https://cdn.openai.com/pdf/a253471f-8260-40c6-a2cc-aa93fe9f142e/economic-research-chatgpt-usage-paper.pdf

Code example

来源：大数据文摘
本文
约2000字
，建议阅读
5
分钟
ChatGPT的真正价值，已远远超出工作场景的生产率提升，而是正在深刻改变人类的生活方式与思维习惯。