### **ChatGPT Gets a Personality Upgrade: Now It Can Learn Your Manners (And You Can Turn Them Off)**

Have you ever found yourself starting every conversation with ChatGPT the same way? “Act as a senior software developer.” “Write in a friendly, casual tone.” “Always explain things simply, like I’m a beginner.” This repetitive setup, while effective, can feel like a chore. Well, those days are over. OpenAI has rolled out a powerful new feature called “Custom Instructions,” and it’s fundamentally changing how we interact with the AI by giving it a persistent memory of who you are and how you want it to behave.

In short, you can finally teach ChatGPT some manners—or a specific lack thereof—and have it stick.

#### What Are Custom Instructions?

Tucked away in your settings, the Custom Instructions feature presents you with two simple text boxes that hold immense power:

1. **”What would you like ChatGPT to know about you to provide better responses?”**
2. **”How would you like ChatGPT to respond?”**

The first box is for context about you. You can tell it your profession, your location, your goals, or your specific interests. For example, “I’m a third-grade teacher in Texas, so tailor science explanations for an 8-year-old’s understanding.”

The second box is where the magic of “manners” and personality comes into play. This is your director’s chair. You can dictate the AI’s tone, style, format, and overall demeanor for every conversation going forward.

#### Crafting Your Ideal AI Companion

This second instruction box allows for an incredible degree of personalization. You can finally stop the AI from being overly verbose or annoyingly apologetic. It’s all up to you.

Want a more polite and formal assistant? You could instruct it:
* “Always address me as ‘Sir’.”
* “Begin your responses with a polite greeting and end with a courteous sign-off.”
* “Avoid using slang, contractions, or overly casual language.”

Prefer a straight-to-the-point, no-nonsense partner? Try this:
* “Be direct and concise. Do not use filler phrases or unnecessary pleasantries.”
* “Get straight to the answer. No pre-amble.”
* “Use bullet points for lists and bolding for key terms.”

Or maybe you want something with more personality:
* “Respond with a witty and slightly sarcastic sense of humor.”
* “Adopt the persona of a wise, old-school detective when solving problems.”
* “Always end your responses with a thought-provoking question related to the topic.”

The possibilities are endless. You can ask it to debate you, to always play devil’s advocate, to structure its code responses in a specific format, or to never, ever say “As a large language model.”

#### The “Off Switch”: You’re Always in Control

The best part of this new feature is that you aren’t locked in. OpenAI has included a simple toggle to “Enable for new chats.” If you’re starting a conversation where you want the default, vanilla ChatGPT experience, you can simply turn it off before you begin. You can also edit your instructions at any time as your needs change.

This feature moves ChatGPT from a one-size-fits-all tool to a truly personalized assistant. It saves time, reduces frustration, and ensures the AI’s output is consistently aligned with your specific needs and preferences. So go ahead, dive into your settings and teach your ChatGPT some manners. Or tell it to be a sarcastic pirate. The choice is finally yours.