When interacting with AI models such as ChatGPT, the way you phrase your prompts plays a big role in the quality of the responses you receive. To get the most out of these advanced models, it’s essential to understand some critical concepts in prompt design: Ambiguity Reduction, Constraint-Based Prompting, and Comparative Prompt Engineering. These techniques help to tailor your prompts for specificity and clarity, eliciting more valuable and accurate responses from the AI.

Ambiguity Reduction

Ambiguity reduction is crucial when designing prompts for AI models, including large language models (LLMs) like ChatGPT. An ambiguous prompt can lead to various interpretations, potentially resulting in an unanticipated or undesirable output. By clearly specifying the context and desired format of the response , you can guide the model toward more valuable and precise outputs.

Constraint-Based Prompting

Constraint-based prompting is a method to guide AI responses by defining explicit conditions or requirements in the prompt. For instance, if you want a list of all prime numbers under 100, instead of asking, “What are prime numbers?” you could ask, “Can you list all prime numbers less than 100?” Providing specific constraints  encourages the model to produce a more focused and applicable response.

Comparative Prompt Engineering

Comparative prompt engineering asks the model to compare or contrast multiple entities or concepts. This technique can help evaluate a model’s understanding, knowledge recall, and ability to discern differences and similarities between various items. A comparative prompt might be: “Compare the principles of classical and quantum physics.”




Ambiguity Reduction

Clarifying prompts to narrow down possible interpretations and guide the AI model towards desired outputs.

Constraint-Based Prompting

Defining explicit conditions or requirements in the prompt to focus the AI’s response.

Comparative Prompt Engineering

Asking the AI model to compare or contrast multiple entities or concepts to evaluate its understanding and discernment.


Implementing these strategies can significantly enhance the utility of AI models in answering prompts and fulfilling specific tasks. By reducing ambiguity, setting explicit constraints, and leveraging comparison, you can prompt the model more effectively and receive more meaningful responses. Remember, prompt engineering is an iterative process, so continually refining your prompts will lead to better results.

Practice and application

The table below provides a progression of three prompts for each strategy: Ambiguity Reduction, Constraint-Based Prompting, and Comparative Prompt Engineering. Try out one or all of these prompts with ChatGPT and observe the difference in the responses you receive. Remember, a more refined prompt often leads to a more satisfying response.

As you examine the responses, apply the RACCCA framework for evaluating AI responses: Relevance, Accuracy, Completeness, Clarity, Coherence, and Appropriateness. This exercise will help you master crafting effective prompts and better understand the nuances of interacting with language models like ChatGPT. Happy experimenting!



Initial Prompt

Improved Prompt

Optimal Prompt

Ambiguity Reduction

Tell me about climate change.

Describe the effects of climate change.

Discuss the impact of climate change on global agriculture, including specific examples of affected regions.

Constraint-Based Prompting

What is a black hole?

Explain the formation of a black hole.

Describe the process of a star collapsing to form a black hole, detailing the stages and physical changes involved.

Comparative Prompt Engineering

Explain AI.

Describe the different types of AI.

Compare and contrast narrow AI, general AI, and superintelligent AI, providing examples and practical applications for each.


To learn more about building your ChatGPT prompting skills, check out the CareerCatalyst course, AI Foundations: Prompt Engineering.