Suggested Category: Artificial Intelligence Estimated Read Time: 6 minutes Author: Gammatek ISPL Technology Editorial Team Reviewed By: Gammatek ISPL AI & Enterprise Technology Desk Last Updated: August 2026

Millions of people use OpenAI every single day to answer quick questions or write emails. But if you are only using the basic chat box, you are barely scratching the surface of what this AI can actually do. Hidden deep within the settings and menus are powerful tools that can completely change how you work, create, and organize your life. In this article, we are going to break down three incredible OpenAI features you probably didn't know existed—and exactly how to start using them today. Most people use ChatGPT the same way they used a search engine three years ago — type a question, get an answer, close the tab. But OpenAI has quietly shipped several genuinely useful features into ChatGPT over the past few months that never got a big launch announcement. They're sitting in the interface right now, unused by most people who open the app every day.

We tested each of these directly inside ChatGPT to confirm how they actually behave, not just how OpenAI describes them. Here are three worth adding to how you use it.

1. The Table of Contents You've Probably Scrolled Past

Ever had a ChatGPT conversation that started as "one quick question" and turned into a 40-message thread covering five different topics? OpenAI quietly added a table of contents panel that appears automatically once a conversation reaches five or more responses.

Where to find it: Look for a set of small horizontal lines on the middle-right edge of the conversation window during a long chat. Clicking it opens a navigable outline of every distinct topic covered in that thread, letting you jump straight to the relevant section instead of scrolling through dozens of messages.

Why it's actually useful: Long-running chats — project planning threads, ongoing research sessions, iterative writing sessions — become genuinely searchable instead of becoming a scroll-fatigue nightmare. If you tend to keep a single "working" conversation open for days at a time rather than starting fresh each time, this alone changes how usable that habit is.

2. The One-Message Model Switch (Without Changing Your Default)

Most people know ChatGPT lets you pick between response modes — faster, lighter responses versus slower, more thorough reasoning. What far fewer people know is that you don't have to change your entire conversation's default setting to use a different mode for a single message.

Where to find it: On mobile, press and hold the send button before sending a message. A menu appears letting you choose a different intelligence level or model just for that one message — your default setting for the rest of the conversation stays untouched.

Why it's actually useful: This solves a genuinely common friction point — needing a deeper, more thorough answer for one specific complex question in the middle of an otherwise casual conversation, without permanently switching modes and slowing down every message afterward.

3. Shopping Research Mode (It's Not Just a Chatbot Answering Product Questions)

If you've asked ChatGPT something like "what's the best noise-canceling headphones under $200," you may have noticed the response looked different from a normal answer — more structured, comparison-style, pulling in current listings. That's Shopping Research Mode, a dedicated tool most users don't realize they're triggering, or don't know they can trigger deliberately.

Where to find it: It's often auto-suggested when you ask a shopping-related question, but you can activate it directly from the tools section near the chat box rather than waiting for it to trigger automatically.

Why it's actually useful: Rather than a single narrow answer, this mode is built to actively compare products and surface current deals across the web — useful for research-heavy purchase decisions where you'd normally have opened six browser tabs to compare options yourself.

A Quick Privacy Note Worth Knowing Alongside These

While testing these features, it's worth flagging a setting connected to all of them: under Settings → Data Controls, there's a toggle called "Improve the model for everyone." On some account types, conversations may be used to help train future models by default. If you're using any of the features above for sensitive work — client materials, internal business data, anything confidential — it's worth deliberately checking that setting rather than assuming a default that may not match your intent.

Frequently Asked Questions

Are these features available on the free ChatGPT plan? Availability varies — some features like the table of contents and shopping research mode are broadly available, while advanced model/intelligence-level switching may depend on your subscription tier. Check your account's available models under Settings if a specific option isn't visible.

Do I need to enable these features manually? The table of contents and shopping research mode are largely automatic and trigger based on conversation length or query type, though shopping research mode can also be manually selected from the tools menu. The one-message model switch requires a manual long-press gesture on mobile.

Will OpenAI keep changing where these features are located? Likely yes — OpenAI has been shipping frequent interface updates throughout 2026, so exact button locations may shift over time. The underlying capabilities are worth knowing about regardless of small interface changes.

Is my data safe when using these features? Standard OpenAI data handling policies apply to all ChatGPT features. If privacy is a specific concern for sensitive work, review the "Improve the model for everyone" setting under Data Controls and consider Temporary Chat for one-off sensitive conversations.

The Bottom Line

None of these three features required a subscription upgrade or a hidden settings hack — they're already sitting inside the ChatGPT interface most people use every day. The gap isn't access, it's awareness. Knowing where to look turns ChatGPT from a question-and-answer tool into something closer to a genuine daily working assistant.

This article reflects hands-on testing of ChatGPT's interface as of August 2026 by Gammatek ISPL's technology editorial team. Feature availability and interface placement may change as OpenAI continues rolling out updates.