One of the most common frustrations with AI is having to repeat the same context, formatting rules, and background information in every new chat. ChatGPT Custom Instructions solves this by allowing you to set permanent preferences that the AI remembers across all conversations. Whether you are a coder, a content creator, or a student, mastering this feature will significantly boost your productivity.
Step 1: Access the Custom Instructions Menu
To get started, log in to your ChatGPT account on a desktop or mobile device. Click on your Profile Icon or name (usually located in the bottom-left or top-right corner) and select Custom Instructions from the settings menu. If you are using the mobile app, you can find this under Settings > Personalization > Custom Instructions.
Step 2: Define Your Background (The 'Who Are You?' Section)
The first text box asks: "What would you like ChatGPT to know about you to provide better responses?" This is where you provide context so the AI doesn't have to guess. To optimize your productivity, include the following details:
- Your Profession/Role: (e.g., "I am a Senior Software Engineer specializing in React.")
- Your Goals: (e.g., "I use AI to brainstorm marketing copy and debug code.")
- Your Skill Level: (e.g., "Explain technical concepts to me like I am an expert, not a beginner.")
- Preferred Tools: (e.g., "I prefer solutions that use Python or Google Sheets.")
Step 3: Set Your Response Preferences (The 'How Should It Respond?' Section)
The second box asks: "How would you like ChatGPT to respond?" This section dictates the AI's tone, length, and style. To make the AI more effective, use these productivity-focused prompts:
- Tone: "Keep responses professional, concise, and objective."
- Formatting: "Always use Markdown headers, bold key terms, and use bullet points for lists."
- Detail Level: "Don't apologize for being an AI. If you don't know an answer, say so. Avoid fluff and filler words."
- Actionability: "Whenever providing code, include comments and a brief explanation of the logic."
Step 4: Implement 'Chain of Thought' Requirements
To improve the accuracy of ChatGPT's reasoning, add a requirement in the response settings that encourages step-by-step logic. For example: "When solving complex math or logic problems, explain your reasoning step-by-step before providing the final answer." This reduces the likelihood of AI hallucinations and ensures higher-quality output.
Step 5: Save and Test Your Profile
Once you have filled out both sections, ensure the "Enable for new chats" toggle is turned on and click Save. Open a new chat and ask a simple question related to your field. For example, if you set your persona as a developer, ask "How do I optimize a loop?" You should notice that ChatGPT immediately uses your preferred language, style, and technical depth without any additional prompting.
Step 6: Iterative Refinement
Custom Instructions are not permanent. As your workflow changes, you should periodically update these fields. If you find ChatGPT is being too wordy, go back and add: "Limit all initial responses to under 200 words." If you are switching from a coding project to a creative writing project, update your background info to reflect your new goals. This flexibility is the key to maintaining a high-speed AI workflow.
💡 Pro Tip: Keep your software updated to avoid these issues in the future.
Category: #AI