5 Suggestions for Writing Good AI Prompts… According to Chat GPT

As the AI buzz continues, you’ve probably started to hear things like ‘AI Prompt Engineering’. There’s a bunch of good and emerging resources out there. And I thought it’d be fun/interesting to hear suggestions from AI itself. In this case ChatGPT. Thought this was a pretty good answer from what I’ve read. Enjoy.

Creating AI prompts can be a great way to stimulate interaction with a language model. However, doing so effectively requires some planning and understanding of the AI's capabilities and limitations. Here are the top 5 suggestions for writing AI prompts:

1. Be Specific: The more specific you are with your prompt, the better the AI will understand and respond. If you're vague or ambiguous, the AI may not give you the result you're hoping for. For example, instead of asking "tell me about history", ask "Can you provide an overview of the American Civil War?"

2. Context Matters: AI models like GPT-4 respond based on the context provided in the prompt. If you are providing a scenario, make sure the scenario is explained in sufficient detail. For example, instead of saying "Write a dialogue", say "Write a dialogue between a scientist and a child about the process of photosynthesis".

3. Use the Right Format: If you want the AI to respond in a certain format, specify that in your prompt. If you want a list, start your prompt with "List...". If you want a story, start your prompt with "Write a story about...". The format you ask for can guide the AI's response.

4. Establish the Desired Tone: If you want the response to be in a specific tone or style (humorous, academic, conversational, etc.), specify that in your prompt. You can say something like "In a conversational tone, explain quantum physics".

5. Manage Expectations: While AI models like GPT-4 are highly sophisticated, they do have their limitations. They won't always understand complex human emotions, humor, or subtext. They also won't be able to provide real-time or updated information beyond their last training data (in this case, up until September 2021).

Finally, remember to iterate and refine your prompts based on the AI responses. Experimentation will often yield the best results.

Next
Next

The Human Factor: The Overlooked Risk with Rapid Technology Changes