OpenAI has announced that deep research in ChatGPT is now enabled by GPT-5.2, bringing enhanced controls over research workflows along with richer production, visualization and review capabilities.
First introduced last year, Deep Research is an agent feature within ChatGPT designed to handle complex multi-step web research tasks. The latest update improves how users navigate the search process and how results are generated and presented – without leaving the chat interface.
Advanced search controls and personalized results
With GPT-5.2 integration, deep research gives users better control over both how information is collected and how results are delivered. Users can now:
- Connect supported apps within ChatGPT and search for specific websites
- Limit the search to selected fields or defined time ranges
- Apply filters and comparison parameters
- Track your search progress in real time
Users can also interrupt an active research session to add follow-up questions or introduce new resources – making the workflow more interactive and iterative.
On the production side, in-depth research supports:
- Structured reports and tables
- Visual data charts and summaries
- Full-screen report mode for focused review
These tools allow users to customize both the depth and the presentation, tailoring the results for analysis, sharing or decision making.
Structured Reporting and Transparency
In-depth research now produces integrated reports that combine analysis with documentation. Depending on the assignment, reports may include:
- Key findings and summary insights
- Visual comparisons and trend charts
- Methodology explanations detailing how the results were compiled
- Added context comment sections
- Source citations for verification
All reports remain embedded within the chat, allowing users to review, refine and export results directly to ChatGPT.
Availability
The Deep Search update with GPT-5.2 is rolling out starting today, with enhanced controls and reporting features becoming available as the deployment progresses.
