OpenAI has recently unveiled a new agentic tool for ChatGPT called “Deep Research”. Built using the upcoming OpenAI o3 model, this new tool can “create a comprehensive report at the level of a research analyst” from simple text prompts.
The announcement was made via a blog post on OpenAI’s official website. “Today, we’re launching deep research in ChatGPT, a new agentic capability that conducts multi-step research on the Internet for complex tasks. It accomplishes in tens of minutes what would take a human many hours,” the company stated.
OpenAI has decided to release Deep Research as part of ChatGPT. To use it, one has to select the “deep research” option in the message composer and enter a valid prompt. Users can add further context with notes, images, or spreadsheets. Then the AI will analyze several sources on the internet and compile them in a single report. To do this, the AI will utilize “end-to-end reinforcement learning” on hard browsing and reasoning.
See Related: ECB Report Reveals Shifting Trade Concerns In European Manufacturing Sector
According to the company, Deep Research can take 5 to 30 minutes to complete any given report. Users can perform other tasks in the meantime. A sidebar will appear to summarize the report’s progress and sources used. Once finished, the AI will send a notification, and the report will be available in the chat as text. The company has promised to add images, data visualization, and other modalities in the coming weeks.
Currently, Deep Research is only available to ChatGPT Pro users with a limit of 100 queries per month. OpenAI plans to give access to Plus and Team subscribers in the next phase and then to Enterprise users. The company has also promised a higher rate limit for paid users in the next version of Deep Research.
