What data does OpenAI Signals v3.0 cover?

It covers work-related messages from individual ChatGPT plans, collected quarterly from Q1 2025 through Q3 2026, across about 100 countries. Enterprise plans and Codex usage are not included.

How does OpenAI classify work tasks in the report?

OpenAI uses the U.S. O*NET occupational database to group tasks into 165 Intermediate Work Activities (IWA), which lets it compare usage patterns across countries.

What does Japan's usage look like in the report?

In Q3 2026, planning and decision-making was Japan's largest category at 35.8%, followed by deliverable creation at 22.3% and information input at 13.7%. Planning and decision-making rose from 31% in Q1 2025, while communication and interaction fell from 13.5% to 7%.

Large Language ModelsAI Business & IndustryImpress Watch2h agoOct 27, 2026, 16:00 JST

OpenAI Signals v3.0: top ChatGPT tasks fall to 51.1% of work use

Send on LINE

3 Key Points

What happened
OpenAI published OpenAI Signals v3.0, classifying work-related ChatGPT messages across about 100 countries into 165 task types; in the 74 countries with full data, the top 10 activities fell from an average 56% to 51.1%.
Why it matters
The figures suggest work-related ChatGPT use is spreading across a wider range of tasks rather than staying clustered in a few dominant ones, with OpenAI citing national industry mix and access to AI tools as possible factors.

WHO IT HITSCorporate teams that set AI usage policy—especially those in HR, operations, and IT—may see more varied day-to-day ChatGPT use across job functions, making it harder to manage with a single one-size-fits-all guideline, based on the report's finding that work use is less concentrated.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

OpenAI Signals v3.0 is a new statistical view of how ChatGPT is used for work, built on the U.S. O*NET database and its 165 Intermediate Work Activities. The headline finding is a gradual loosening of concentration: in the 74 countries with complete quarterly data, the top 10 task types accounted for an average 56% of work-related messages in Q1 2025 but 51.1% by Q3 2026. OpenAI notes that the mix of top tasks differs by country and suggests national industry structure and access to AI tools may play a role.

The Japan and U.S. breakdowns add country-level color to that trend. In Japan, planning and decision-making rose from 31% to 35.8% between Q1 2025 and Q3 2026, and development and advice grew from 8.8% to 10%, while deliverable creation and communication and interaction declined. The report's scope is limited to individual ChatGPT plans; enterprise plans and Codex are excluded, so the figures describe consumer-side work use rather than corporate deployments.

FAQ

What data does OpenAI Signals v3.0 cover?: It covers work-related messages from individual ChatGPT plans, collected quarterly from Q1 2025 through Q3 2026, across about 100 countries. Enterprise plans and Codex usage are not included.
How does OpenAI classify work tasks in the report?: OpenAI uses the U.S. O*NET occupational database to group tasks into 165 Intermediate Work Activities (IWA), which lets it compare usage patterns across countries.
What does Japan's usage look like in the report?: In Q3 2026, planning and decision-making was Japan's largest category at 35.8%, followed by deliverable creation at 22.3% and information input at 13.7%. Planning and decision-making rose from 31% in Q1 2025, while communication and interaction fell from 13.5% to 7%.

Impress WatchRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.