ChatGPT Deep Research

ChatGPT Deep Research is OpenAI's autonomous research agent that browses the web for 5-30 minutes to produce cited, comprehensive reports on complex topics. Updated to GPT-5.2 in February 2026 with better source steering, MCP server connectivity, and improved report UI. It operates differently from standard ChatGPT -- instead of answering immediately, it plans a research strategy, browses dozens of sources autonomously, and synthesises findings into a structured report with citations. Pricing is tied to ChatGPT plans: Free (5 lightweight queries/month), Plus ($20/month -- 25 queries/month), Pro ($200/month -- 250 queries/month including priority access). Business ($25/user/month) and Enterprise also include Deep Research. The 10-query monthly limit is the first constraint Plus users hit. Pro is only cost-effective for researchers who run Deep Research daily and need the GPT-5.4 Pro model for enhanced accuracy. The agent scored 26.6% on Humanity's Last Exam (vs. 9.1% for o1) and is state-of-the-art on the GAIA benchmark. February 2026 added MCP server support so users can connect to specific data sources and restrict web searches to trusted sites -- a meaningful enterprise unlock. Best fit: knowledge workers in finance, science, policy, legal, and engineering who need analyst-grade research outputs and can absorb the 10-25 query/month cap. Not appropriate for teams that need structured literature review of academic databases (Elicit/Consensus do this better) or for daily high-volume research where Pro's $200/mo becomes the only viable tier. Not affiliated.