ChatGPT Features: Kicking the Tires on the Deep Research Tool Applied to Health Insurer Stock Price Collapse

One of my favorite perks of the ChatGPT premium plan is a little-known feature called Deep Research. I have found it to be the equivalent of asking a supercharged intern to spend the weekend combing journals, earnings calls, and news clips—except the “intern” returns in about ten minutes and never complains about footnotes. In this video I a real-world puzzle: Why have UnitedHealth, Humana, and Centene seen their share prices crater over the last year? Here’s how the experiment unfolded and why I think this feature is a a great AI tool for busy clinicians, educators, and knowledge junkies.

---

A Quick Tour of the Workflow

1. Find the Tool. Hit the Tools dropdown, scan past “Study and Learn” ( check out my other video on the study and learn feature) and “Agent Mode,” and click Deep Research.

2. Speak or Type Your Prompt. I dictated a mouthful:

“Give me a detailed deep-dive research report on why UnitedHealth, Humana, and Centene have had such a precipitous drop in share price over the last year. Explore shrinking margins, missed earnings, practical examples, and balanced projections for how they might course-correct.”

1. Answer Clarifying Questions. Deep Research volleyed back: What date range? Focus on Medicare Advantage? I filled in the blanks, asking it to look at the last 12 months and to highlight Medicare Advantage while still touching commercial and Medicaid lines.

2. Let It Cook. The model spent ~10 minutes churning. Pro tip: open a new tab and work on something else—ChatGPT won’t mind.

3. Review the Report. The output is a multi-section briefing peppered with embedded links to earnings calls, policy memos, and reputable news coverage. I skimmed, fact-checked a few citations, and was impressed by the breadth (if occasionally repetitive sourcing). I have the copy of the final output at the bottom of this article for paid subscribers.