OpenAI Unveils Deep Research Mode in ChatGPT

The AI startup OpenAI has launched a new agent feature in ChatGPT—Deep Research. The chatbot has acquired the ability to conduct multi-stage research and tackle complex tasks by analyzing hundreds of online sources.

Today we are launching our next agent capable of doing work for you independently—deep research.

Give it a prompt and ChatGPT will find, analyze & synthesize hundreds of online sources to create a comprehensive report in tens of minutes vs what would take a human many hours. pic.twitter.com/03PPi4cdqi

— OpenAI (@OpenAI) February 3, 2025

The mode operates on the “thinking” AI model o3, optimized for browsing web pages, data analysis, and using Python tools. It employs logic to search, interpret, and investigate vast amounts of text, images, and PDF files. The tool “solves in tens of minutes what would take a human many hours.”

Deep Research will also aid in creating AGI, the team claims.

The new feature is primarily intended for individuals engaged in intensive work in finance, science, politics, or design who require precise research. It will also assist consumers needing “hyper-personalized” recommendations, such as when purchasing cars, appliances, and furniture.

Deep Research takes between 5 to 30 minutes. Initially, it is available to users with a ChatGPT Pro subscription. Support for Plus, Team, and Enterprise will follow. There is no information on access to the service without paid plans.

OpenAI tested the Deep Research mode using Humanity’s Last Exam. The test includes over 3,000 expert-level questions across various scientific fields. The company’s AI model performed significantly better compared to competitors.

Previously, OpenAI added the o3-mini model to ChatGPT. It is available to users without a paid subscription.

In January, OpenAI partnered with the US National Laboratories to apply the company’s AI models for scientific research and nuclear weapons security.