There are plenty of tips and tricks for ChatGPT out there, but most people still treat the chatbot like it's a more polite version of a search engine.
What they don't know is that there's a whole hidden menu of codes you might not know about, shared around online by ChatGPT enthusiasts. I've found them to be very useful in skipping to the kind of answers I'm looking for.
Article continues below
1. ELI5 cures confusion
ELI5 stands for “Explain Like I’m 5,” which isn't an insult when facing a whole new realm of knowledge. At least as a starting point, a five-year-old's understanding is plenty. So, when caught in a web of confusion, type ELI5: followed by the topic you want to know about.
If a home insurance document says something about, “actual cash value versus replacement cost”, I can type, "ELI5: What is the difference between actual cash value and replacement cost in home insurance?", and suddenly, I am not parsing legal soup anymore.
I even used it when practicing guitar and working to understand how guitar modes are both the same and different musical notes. I asked ChatGPT, "ELI5: what are guitar modes?" and the AI explained how they're "basically different flavors of a scale. Imagine you have the same set of notes, but you start on a different one each time. Even though the notes are the same, it changes the mood. They’re like different emotional versions of the same musical ingredients.”
2. TLDR breaks walls of text
TLDR is fairly common internet slang, so ChatGPT's understanding of the idea of “too long; didn’t read” isn't surprising. But it's also a useful ChatGPT command.
Just type TLDR and then paste in whatever long text you want condensed. You get an immediate summary without any need to explain what you want in further detail.
I've used it for recipe blogs that insist on opening with a memoir, for parsing subscription cancellation policies, and even for very long group texts that I didn't get to before facing an hour of reading to catch up on.
I recently had the pleasure of getting a very long list of instructions for an upcoming event that wandered somewhat widely and was apparently formatted in the dark with pool cues.
One TLDR request later, and I had five brief bullet points and assurances that I mainly had to "arrive a little early and make sure your ticket is easy to access on your phone.”
3. Fun and fancy jargon
Jargonize is a sillier sort of code, but it has its uses — it makes what you write immediately more complex, often unnecessarily so.
I wouldn't use it in most cases as it largely ends up sounding like the most annoying versions of LinkedIn posts, but it can be entertaining to sound like a management consultant in unlikely contexts.
Just write "Jargonize:" and whatever you want to formalize it.
For instance, when telling a friend about my need to stop buying random gadgets I don't use often, I made it sound like a much more intense issue with Jargonize, which informed my friend that, “I’m looking to adopt a more intentional approach to purchases and reduce low-frequency gadget acquisitions that add clutter without delivering long-term value.”
Making ordinary life into corporate theater is amusing in small doses, even if the real thing is often irritating.
4. Humanize adds life to rote answers
Almost the inverse of Jargonize is the Humanize code. When something is serious, or it's important to avoid sounding passive-aggressive, ChatGPT is surprisingly useful.
It can call out and advise on how to go beyond cliches and stiff formality in condolence notes, and avoid accidental insults.
Paste in your text after writing "Humanize:" in the prompt, or just explain what you're looking for. Clunky messages that sound fake-friendly or just want a text sound less passive-aggressive and more genuinely clear. It's useful.
As the chatbot pointed out, "sometimes the line between 'gentle reminder' and 'mid-level emotional warfare' is very thin."
5. Feynman's answers
Richard Feynman created the Feynman Technique as a streamlined method for building real understanding. The physicist is often quoted as saying if you cannot explain something simply, you probably do not understand it properly.
The prompt is slightly longer than the others, but still just a short phrase. Write, "Teach me this using the Feynman Technique," followed by the topic.
The AI then goes through the four steps of the technique, explaining it simply, identifying gaps, simplifying further, and reviewing.
You can also ask ChatGPT to quiz you, point out what you are missing, and then help you explain it back. This is especially useful for anything you keep “sort of” understanding and then immediately forgetting, like how chargers work. That is where it becomes more than just a neat explanation machine.
It turns into a learning loop. I've used it for music theory, economics, and plenty of other topics. Using the Feynman Technique, ChatGPT can walk through it in steps and then check whether I actually get it.
These codes are just tiny instructions that help ChatGPT pick the right mode to interact with you. They aren't complicated, and that is really the whole point. Most of the time, simple prompts work better than sprawling multi-paragraph ones. The longer ones have their place, but they won't be go-tos for everyday use.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.
English (US) ·