ChatGPT has a new “deep research” agent that OpenAI says uses reasoning to synthesize large amounts of online information and complete multi-step research tasks. “It accomplishes in tens of minutes what would take a human many hours,” OpenAI suggests, claiming it will “synthesize hundreds of online sources to create a comprehensive report at the level of a research analyst.” Powered by a version of the upcoming OpenAI o3 model optimized for web browsing and data analysis, the company says the deep research agent will typically take 5 to 30 minutes to complete its work. The agent is described as an ideal research tool for areas such as finance, science and engineering.

OpenAI explains in a blog post that its new deep research model contrasts with GPT-4o in that the latter “is ideal for real-time, multimodal conversations” while the new model specializes in “multi-faceted, domain-specific inquiries where depth and detail are critical,” relying on “deep research’s ability to conduct extensive exploration and cite each claim.”

The result, says OpenAI — which also details the deep research concept in a YouTube video — “is the difference between a quick summary and a well-documented, verified answer that can be usable as a work product.”

The new agent “can operate autonomously to ‘plan and execute a multi-step trajectory to find the data it needs, backtracking and reacting to real-time information where necessary,’” writes The Verge, explaining that “instead of simply generating text, it shows a summary of its process in a sidebar, with citations and a summary showing the process used for reference.”

Users can query the agent using text, images or files like PDFs or spreadsheets, and OpenAI says it will soon accommodate embedded images and charts.

The Verge embeds a demo showcasing “a request for info on changes in the retail industry over the last three years, with a response that includes bullet points and tables.”

The new capability “was designed for ‘people who do intensive knowledge work in areas like finance, science, policy, and engineering and need thorough, precise, and reliable research,’” reports TechCrunch, which adds it could also be useful “for anyone making ‘purchases that typically require careful research, like cars, appliances, and furniture.’”

The deep research agent is subject to the problems common to AI, with OpenAI acknowledging its limitations “saying it can ‘sometimes hallucinate’ and make up facts, struggle with telling the difference between authoritative info and rumors, and register how certain it should rate a response,” The Verge warns.

The deep research agent closely follows OpenAI’s January launch of Operator, “a tool that can use a web browser to complete tasks for you, and is similar to the Project Mariner research prototype Google showed off in December,” The Verge reports, adding that the Google tool “is not available to the public yet.”

The deep research agent is available now to Pro users (who get up to 100 queries as part of their $200 per month subscription) with access for Plus and Team subscribers queued-up next.