CREATIVE SURVEY Prompt

Prompt Engineering - Generating Creative Survey Responses with LLMs

Language models can be used not just for factual answers but also for generating imaginative and creative content. This chapter explores how to prompt LLMs like ChatGPT to produce creative or unconventional responses to survey-style questions, useful for brainstorming, content generation, or adding a unique flair to interactions.

Prompting for Creative Survey-Style Answers

Instead of seeking standard answers, you can prompt an LLM to respond to a survey question from an imaginative, humorous, poetic, or otherwise creative perspective. This involves clearly signaling your intent for creativity in the prompt.

Example of prompting for a creative survey response:

User: Imagine you are a wise old tree. If asked in a survey, 'What is your favorite season and why?', how would you respond creatively?

ChatGPT (expected creative output):
"Ah, a curious inquiry from a fleeting being! If my rings could whisper, they'd tell you of Autumn. Not for the melancholy some perceive, but for the glorious release! The vibrant hues are my final, fiery breath before the quiet slumber of winter. It's a time of shedding, of letting go, making space for the fresh dreams of spring. Plus, the squirrels are particularly frantic with their nut-burying, and their antics are quite the entertainment!"

In this example, the user explicitly sets a persona (‘wise old tree’) and asks for a ‘creative’ response to a typical survey question. This guides the LLM away from a generic answer.

Best Practices for Eliciting Creative Survey Responses

To get the most imaginative responses:

• Clearly State the Creative Intent: Use words like ‘creative’, ‘imaginative’, ‘poetic’, ‘humorous’, ‘from the perspective of X’ in your prompt.
• Define a Persona or Role (Optional but Effective): Assigning a character or role to the LLM (e.g., ‘Respond as if you are a mischievous cat,’ ‘Answer as a 19th-century inventor’) can greatly enhance creativity.
• Specify Tone and Style: ‘Provide a dramatically pessimistic answer to…’, ‘Write a short, witty response to…’
• Encourage Unexpected Elements: ‘For the survey question “What’s your dream vacation?”, give an answer that includes a surprising twist or a fantastical element.’
• Set Constraints for Focus: Sometimes, too much freedom can be overwhelming. ‘Give a creative, one-sentence answer to…’ or ‘Answer in the style of a haiku.’
• Consider the Survey Question’s Nature: Some questions lend themselves better to creative answers than others. Open-ended, subjective questions are often good candidates.

Example Application: Python Script for Creative Survey Answers

Let’s use Python and the OpenAI API to generate a creative survey response. Remember API key security.

import openai
import os

# Securely load your API key
# openai.api_key = os.getenv("OPENAI_API_KEY")
# For this example, replace 'YOUR_API_KEY_HERE' with your actual key:
openai.api_key = 'YOUR_API_KEY_HERE'

def generate_creative_survey_answer(survey_question, persona=None, style=None):
    try:
        prompt_parts = [f"Consider the survey question: '{survey_question}'"]
        if persona:
            prompt_parts.append(f"Respond as if you are {persona}.")
        if style:
            prompt_parts.append(f"Your response should be {style}.")
        else:
            prompt_parts.append("Provide a creative and imaginative answer.")

        user_prompt_content = " ".join(prompt_parts)

        response = openai.ChatCompletion.create(
            model="gpt-3.5-turbo", # Or a newer model for more nuanced creativity
            messages=[
                {"role": "system", "content": "You are a highly creative and imaginative respondent."},
                {"role": "user", "content": user_prompt_content}
            ],
            max_tokens=200, # Adjust for desired length
            temperature=0.9, # Higher temperature for more creativity and unexpectedness
            n=1,
            stop=None
        )
        return response.choices[0].message['content'].strip()
    except Exception as e:
        return f"An error occurred: {e}"

# Example usage:
question1 = "What is the meaning of life?"
persona1 = "a cheerful robot from the future"
style1 = "optimistic and slightly humorous"
answer1 = generate_creative_survey_answer(question1, persona1, style1)
print(f"Question: {question1}\nPersona: {persona1}\nStyle: {style1}\nCreative Answer:\n{answer1}")

print("\n---\n")

question2 = "If you could have any superpower, what would it be and why?"
answer2 = generate_creative_survey_answer(question2, style="brief and poetic")
print(f"Question: {question2}\nStyle: brief and poetic\nCreative Answer:\n{answer2}")

Potential Output

Running the script could produce outputs like:

Question: What is the meaning of life?
Persona: a cheerful robot from the future
Style: optimistic and slightly humorous
Creative Answer:
Beep boop! The meaning of life, organic friend, is clearly to maximize efficient joy-unit accumulation! And perhaps to ensure all universal power outlets are compatible. Mostly the joy part, though. With optimal charging, of course! ✨

---

Question: If you could have any superpower, what would it be and why?
Style: brief and poetic
Creative Answer:
To weave silence into symphonies,
healing the world's cacophony,
one hushed note at a time.

Conclusion

Prompting LLMs for creative survey responses opens up a fascinating avenue beyond factual recall. By clearly signaling your intent for imagination, assigning personas, or specifying styles, you can guide models like ChatGPT to generate unique, engaging, and often surprising content.

This technique can be valuable for brainstorming, artistic endeavors, creating unique content, or simply exploring the creative potential of AI. Remember that ‘creativity’ can be subjective, so experimentation with prompts and parameters like temperature is key. For further reading, explore topics like ‘computational creativity’ and ‘narrative generation with LLMs’.