ChatGPT characteristics are four settings in Personalization, Warm, Enthusiastic, Headers & Lists and Emojis, that you nudge up or down to change how replies sound without changing what ChatGPT can do. They sit on top of your personality, the Base style and tone setting, where you choose Default or one of six presets: Professional, Friendly, Candid, Quirky, Efficient or Cynical, as of October 2026.
So there are two layers. The personality sets the overall voice, and the characteristics fine-tune it. Both live in the same Personalization screen, and both work alongside your custom instructions and anything ChatGPT has saved to memory. That last part is where most of the confusion starts.
Turning warmth down doesn't make answers shorter
Here's the bit people don't expect. When you set Warm or Enthusiastic to less, OpenAI says ChatGPT keeps the same response length and completeness. It only changes the tone. Less warm sounds emotionally neutral, and less enthusiastic sounds calmer and more objective, but you still get the full answer.
So if you want shorter replies, the warmth slider isn't the one to touch. You'd get the same amount of text in a flatter voice. For brevity, the Efficient personality is the better lever, because it gives a direct answer first and then concise reasoning or steps.
The four characteristics, in both directions
Each characteristic has a more (+) and a less (-) setting. Once you set one, ChatGPT tries to follow that style in the current chat and in new chats going forward.
| Characteristic | More (+) | Less (-) |
|---|---|---|
| Warm | Sounds more sincere and kind, with a friendly tone | Same length and completeness, emotionally neutral tone |
| Enthusiastic | Shows more excitement, curiosity and active interest | Same length and completeness, calmer and more objective |
| Headers & Lists | More markdown, such as headers, lists and tables when it helps | Fewer formatting elements, more paragraphs |
| Emojis | A bit more emoji colour in the tone | Sparing in chatty replies, none in informational answers |
OpenAI uses one prompt, tips for preparing for a job interview, to show each setting. With more headers and lists you get a checklist split into company research, your stories, practice, questions and logistics. With less, you get the same advice as a short run of paragraphs. If you copy ChatGPT's answers into emails or documents, Headers & Lists set to less saves a lot of tidying up.
The six personalities, plus Default
Default is the standard ChatGPT style, clear and neutral. The other six each have their own tone. OpenAI also retired a seventh preset, Nerdy, and moved anyone using it to the default personality.
| Personality | How it sounds | Best for, according to OpenAI |
|---|---|---|
| Default | Clear and neutral | The standard ChatGPT style |
| Professional | Polished and precise, formal language | Workplace communication and documentation |
| Friendly | Warm and chatty, calm clarity and light wit | A conversational sounding board, planning and reflection |
| Candid | Direct and encouraging, honest feedback with next steps | Gut checks on plans, drafts or decisions |
| Quirky | Playful and imaginative, humour and unexpected ideas | Brainstorming and creative writing |
| Efficient | Concise and plain, direct answers | Technical tasks, checklists and troubleshooting |
| Cynical | Sarcastic and dry, blunt help with wit | Entertaining replies that are still actionable |
OpenAI's own comparison asks every personality how to spend a rainy day indoors. Efficient gives a short list of activities, Friendly leans into comfort ideas, Professional offers a structured menu, and Cynical jokes about avoiding all those projects you keep putting off. In coding prompts, Efficient skips the intro altogether.
One detail matters if you write in other languages. Professional uses forms suitable for the workplace in languages that have formal and informal conjugations.
How personality, memory and instructions stack
Changing your personality doesn't change what ChatGPT can or can't do, or the safety rules it follows. It only guides how it communicates. It also applies across all your chats, including existing conversations, which is different from characteristics.
Personality doesn't control every kind of content either. Ask for an email draft, a code snippet, a social post or a resume, and ChatGPT matches the tone to your instructions and the context, not to the personality. OpenAI's example is Friendly being asked for Python code: you get clear, working code without the chatty style.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Then there's memory. Your personality works alongside saved memories and custom instructions, and if a saved memory conflicts with a personality's style, such as a preference for serious and professional replies, it may override or reduce that personality's visible traits. Instructions you give during a conversation can do the same. If you want to see what ChatGPT has stored about you, my guide to ChatGPT memory shows where to look.
Where to find the settings
- On the web, select your profile icon in the bottom-right corner. On iOS or Android, select your profile icon in the sidebar to open Settings.
- Open Personalization.
- Use the dropdown next to Base style and tone to pick a personality.
- Under Characteristics, nudge Warm, Enthusiastic, Headers & Lists and Emojis up or down.
- If a style doesn't stick, check your saved memories and custom instructions for anything that contradicts it.
Why ChatGPT keeps ignoring your style
- A saved memory says something different, such as asking for formal replies, and it's quietly winning.
- Your custom instructions describe a tone that clashes with the personality you picked. For ideas on wording them, see these custom instructions examples.
- You asked for a specific format, such as an email or code, so ChatGPT followed the task instead of the personality.
- Something you said earlier in the conversation adjusted the behaviour for that chat.
Which setup I'd use for work
I'd pick Professional or Efficient as the personality if most of your chats end up in emails, reports or client documents, then set Headers & Lists and Emojis to less so the output needs less cleaning up. I'd try Candid if you want ChatGPT to push back on your drafts instead of praising them. I'd skip the warmth slider if what you're after is shorter answers, because it won't give you that. And if nothing seems to change, it's nearly always a memory or an instruction getting in the way, not a broken setting.
Curious about the neighbouring question? Try ChatGPT Credits.
This guide is part of my ChatGPT Prompts and Limits: 30 Guides to Fun Prompts, Photo Edits and Daily Caps. Check the length with my free caption length checker.
Frequently asked questions
What are ChatGPT characteristics?
They're four settings in Personalization, Warm, Enthusiastic, Headers & Lists and Emojis, that you nudge up or down to change how ChatGPT's replies sound without changing what it can do.
How do I change ChatGPT's personality?
Select your profile icon, open Personalization and use the dropdown next to Base style and tone. The change applies across all chats, including existing ones.
Do personalities change what ChatGPT can do?
No. A personality only guides how ChatGPT communicates. It doesn't change its capabilities or the safety rules it follows.
Why is ChatGPT ignoring my style setting?
A saved memory or custom instruction that conflicts with the style can override it, and so can instructions you give during the conversation.
Which personality is best for work?
OpenAI describes Professional as best for workplace communication and documentation, and Efficient as best for direct instructions, checklists and troubleshooting.
