Cognitive Analyzer Prompting


Cognitive Verifier Prompt: Basic


Teach Your AI to Fact-Check: A Guide to Cognitive Verifier Prompts

Purpose: Ensure the AI’s output is factually accurate or logically consistent.

Template: “Generate [output]. Then, verify the accuracy by [method of verification].”

Example: “Generate a summary of the latest research findings on climate change. Then, verify the accuracy by cross-checking key facts with reputable sources.”

Purpose: Check if the AI’s output fully addresses all parts of the prompt.

Template: “Describe [task or topic]. After completing the description, verify that all essential aspects of [task or topic] are covered.”

Example: “Describe the steps involved in making French toast. After completing the description, verify that all essential ingredients and cooking steps are covered.”

Purpose: Check if the AI’s output fully addresses all parts of the prompt.

Template: “List possible solutions to [problem]. After listing, verify each solution’s relevance to the specific context of the problem.”

Example: “List possible solutions to reduce urban air pollution. After listing, verify each solution’s relevance to the specific context of large metropolitan areas.”

Purpose: Ensure the AI’s output is consistent with previous inputs or known data.

Template: “Provide a response to [situation]. Then, verify that the response is consistent with [related information or previous responses].”

Example: “Provide investment advice based on current market trends. Then, verify that the advice is consistent with historical data on market performance during similar economic conditions.”

Purpose: Check if the AI’s response shows any unintended bias or assumptions.

Template: “Explain [concept or policy]. After explaining, verify that the explanation is free from cultural, gender, or personal biases.”

Example: “Explain the importance of equal education opportunities. After explaining, verify that the explanation is free from cultural, gender, or personal biases by checking if all examples and language used are neutral and inclusive.”

Purpose: Assess if the AI’s response is practically applicable in real-world scenarios.

Template: “Propose a solution to [problem]. After proposing, verify the practicality of the solution by considering [real-world constraints like cost, time, resources].”

Example: “Propose a solution to increase public transportation usage in small towns. After proposing, verify the practicality of the solution by considering constraints like budget, population density, and existing infrastructure.”

Purpose: Ensure the AI’s output comprehensively addresses complex or multi-part questions.

Template: “Address the question [complex question]. After answering, verify comprehensiveness by ensuring all parts of the question have been responded to adequately.”

Example: “Address the question: What are the impacts of remote work on employee productivity, company culture, and urban development? After answering, verify comprehensiveness by ensuring all parts of the question have been responded to adequately.”

Disclaimer!

LLM like ChatGPT, Gemini can provide incorrect and inaccurate outputs. Always double check the output before you use it.

Have a question?

If you have any other queries, feel free to drop a comment.

Learn More !

Experiment directly with ChatGPT and Google Gemini.
Want to learn more about effective prompts to get the best out of GenAI and LLMs?


Discover more from Debabrata Pruseth

Subscribe to get the latest posts sent to your email.

What do you think?

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Scroll to Top