Skip to main content

Why ChatGPT custom instructions are such a big deal

Are you bored with telling ChatGPT who you are and what kind of responses you want, even before you get to your actual prompt? You’re not alone. Fortunately, the Custom Instructions feature is here to save the day. Implement it properly, and you’ll only have to tell ChatGPT one time that you’re a developer and want it to analyze your code, or you’re a teacher who wants help with lesson plans. It’ll remember it the next time you open a new chat, saving you precious time.

For anyone who uses ChatGPT regularly, this is a game changer.

Recommended Videos

What are ChatGPT custom instructions?

ChatGPT custom instructions are a way to preempt ChatGPT by giving it some key information about you and what you want it to do for you. When I get ChatGPT to teach me board game rules, I no longer need to tell it that I’m providing it with board game rulebooks and that I don’t want it’s usual extraneous style of talking. I want it to get straight to the point as soon as I give it a PDF.

With Custom Instructions, it does.

With the new feature, you’re able to tell ChatGPT two key pieces of information: who you are and how you want ChatGPT to respond to your prompts. You input this information into your profile, and then any time you start a new chat, ChatGPT will look over your quick cover letter to the AI before it begins a conversation, thereby preloading it with the kind of instructions that, until now, you’d have to give it every time you opened a new chat.

This isn’t a feature that will be hugely useful for those using ChatGPT for a range of applications, or who frequently have varied discussions with the chatbot. If you use ChatGPT for work, though, or want to frame most of your prompts from a similar baseline of understanding, this tool is a godsend.

ChatGPT custom instructions example by OpenAI.
In this example, OpenAI shows how Custom Instructions can jump right into providing concise, applicable responses, instead of rambling in its usual imprecise fashion. OpenAI

What can you do with custom instructions?

ChatGPT custom instructions are mostly a time-saving mechanic, but they can also make the process of using ChatGPT far more streamlined — especially for professionals. That can make it a far more effective tool, and a great way to further enhance your productivity.

If you’re using ChatGPT for research, you can tell if you’re a student or researcher and ask it to provide detailed citations and references with whatever information it provides.

If you’re a programmer using ChatGPT to write code, make sure it knows you’re a programmer, what languages you like to work in, and that you want clear and concise comments throughout. You could also tell it you want it to follow the DRY, KISS, or YAGNI principles.

Want ChatGPT to just be more precise and concise in its responses? Let it know that you want it to keep its answers short, with no extraneous text, including apologies and reminders. You might ask it to keep all answers to a certain length, or to use bullet points and other formatting to make the response more easy to read.

If you’d like additional help in breaking down complicated topics, ask ChatGPT to respond to each prompt twice, once in a detailed fashion designed to help those familiar with a topic, and another in a more easy-to-understand way, similar to how a teacher might approach a topic for someone just learning it for the first time.

You can even make your chats with ChatGPT more fun. Let it know you want it to talk with a certain patois, or to pepper in some jokes and phrases. Tell it you want an informal conversation and don’t mind it being a little sarcastic sometimes.

Ultimately, just think about anything you often feel the need to ask for, then tell ChatGPT before you actually start hitting it with the prompts you want it to answer: That’s what you want to put in the Custom Instructions fields.

How to turn on ChatGPT custom instructions

To use ChatGPT’s Custom Instructions feature, you’ll need to be a ChatGPT Plus subscriber and reside in the U.S. At the time of writing, custom instructions are not available in the EU or United Kingdom.

Like the recently disabled web browsing, plugins, or the code interpreter, Custom Instructions is a beta feature of ChatGPT, so you’ll need to enable it before you can use it. Here’s how:

Step 1: Open a browser and navigate to the OpenAI website, or open the ChatGPT iOS app.

Step 2: Log in if necessary.

Step 3: For web users, select the three-dot menu icon in the bottom-left, then navigate to Settings Beta features and turn the Custom Instructions toggle to On. For iOS users, navigate to SettingsNew Featuresturn on Custom Instructions.

Step 4: The custom instruction fields will then appear in your Settings menu. Fill the inputs with whatever information you feel is necessary, then start a new chat with ChatGPT for it to consider your inputs form the outset.

Interested in what the competition is up to? Here are some of the best tools using the advanced GPT-4 language model.

Jon Martindale
Jon Martindale is a freelance evergreen writer and occasional section coordinator, covering how to guides, best-of lists, and…
ChatGPT’s latest model is finally here — and it’s free for everyone
OpenAI's ChatGPT blog post is open on a computer monitor, taken from a high angle.

We knew it was coming but OpenAI has made it official and released its o3-mini reasoning model to all users. The new model will be available on ChatGPT starting Friday, though your level of access will depend on your level of subscription.

OpenAI first teased the o3 model family on the finale of its 12 Days of OpenAI livestream event in December (less than two weeks after debuting its o1 reasoning model family). CEO Sam Altman shared additional details on the o3-mini model in mid-January and later announced that the model would be made available to all users as part of the ChatGPT platform. He appears to have delivered on that promise.

Read more
Chatbots are going to Washington with ChatGPT Gov
glasses and chatgpt

In an X post Monday commenting on DeepSeek's sudden success, OpenAI CEO Sam Altman promised to "pull up some releases" and it appears he has done so. OpenAI unveiled its newest product on Tuesday, a "tailored version of ChatGPT designed to provide U.S. government agencies with an additional way to access OpenAI’s frontier models," per the announcement post. ChatGPT Gov will reportedly offer even tighter data security measures than ChatGPT Enterprise, but how will it handle the hallucinations that plague the company's other models?

According to OpenAI, more than 90,000 federal, state, and local government employees across 3,500 agencies have queried ChatGPT more than 18 million times since the start of 2024. The new platform will enable government agencies to enter “non-public, sensitive information” into ChatGPT while it runs within their secure hosting environments -- specifically, the Microsoft Azure commercial cloud or Azure Government community cloud -- and cybersecurity frameworks like IL5 or CJIS. This enables each agency to "manage their own security, privacy and compliance requirements,” Felipe Millon, Government Sales lead at OpenAI told reporters on the press call Tuesday.

Read more
DeepSeek: everything you need to know about the AI that dethroned ChatGPT
robot hand in point space

A year-old startup out of China is taking the AI industry by storm after releasing a chatbot which rivals the performance of ChatGPT while using a fraction of the power, cooling, and training expense of what OpenAI, Google, and Anthropic's systems demand. Here's everything you need to know about Deepseek's V3 and R1 models and why the company could fundamentally upend America's AI ambitions.
What is DeepSeek?
DeepSeek (technically, "Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co., Ltd.") is a Chinese AI startup that was originally founded as an AI lab for its parent company, High-Flyer, in April, 2023. That May, DeepSeek was spun off into its own company (with High-Flyer remaining on as an investor) and also released its DeepSeek-V2 model. V2 offered performance on par with other leading Chinese AI firms, such as ByteDance, Tencent, and Baidu, but at a much lower operating cost.

The company followed up with the release of V3 in December 2024. V3 is a 671 billion-parameter model that reportedly took less than 2 months to train. What's more, according to a recent analysis from Jeffries, DeepSeek's “training cost of only US$5.6m (assuming $2/H800 hour rental cost). That is less than 10% of the cost of Meta’s Llama.” That's a tiny fraction of the hundreds of millions to billions of dollars that US firms like Google, Microsoft, xAI, and OpenAI have spent training their models.

Read more