Empowering teams to unlock insights faster at OpenAI

This is part of our series sharing internal examples of how OpenAI is using its own technology and APIs. These tools are being used internally, only at OpenAI, and are shared here as illustrative examples of how frontier AI is supporting use cases across our teams. We’re also sharing the internal tool names for a clearer look at how frontier AI helps our teams get work done.

Buried in the noise

Millions of support tickets come in every year. Each one carries something valuable: a frustration, an idea, a request.

But until recently, those signals were difficult to understand. Dashboards hinted at trends but missed the why. Deep dives took weeks of work from a data scientist. A product leader might want to know how a new feature landed with a unique audience type. But answering required a data scientist to conduct a detailed analysis.

Curiosity was getting rationed.

“The process required deep technical expertise, and it was cutting off our curiosity,” says Molly Jackman, Head of Business Data.

A new way to ask 

We built a research assistant to unlock curiosity that scales. It combines two modes of exploration: dashboards for patterns and a conversational interface for digging deeper. You can start with a chart of trending issues, then ask follow-up questions in plain language.

We built it by blending what already worked. On one side, classifiers and charts that structured millions of tickets into product areas and themes. On the other, GPT‑5, which could summarize raw tickets and generate flexible reports in plain language. The combination gave us both speed and depth that was simple enough for anyone to use.

“What are healthcare customers saying about new integrations?”

“What’s driving support tickets this quarter?”

“What major features are hitting the mark?”

“The magic is that you don’t have to predefine your questions, you can just follow your curiosity.”

Molly Jackman, Head of Business Data

Making it reliable

Speed is meaningless without accuracy.