News & Newsletters

Digital Learning's Nadine Stewart recently attended a workshop on the CIDI framework for AI prompt engineering and shares the key takeaways.

“Conversational AI tools such as ChatGPT and Copilot are increasingly becoming more integrated in our working and our day-to-day lives, helping us with tasks as diverse as coding to giving recommendations and language learning.

However, what we get out of these tools is arguably only ever as good as what we put in. If we improve the quality of our prompts, also known as ‘prompt engineering’, we can help AI tools to help us and in turn we will inevitably receive more accurate, relevant, and focused responses.

The CIDI framework provides a clear, structured approach to writing AI prompts. The basic elements are:

• Context: Provide background to set the scene. What’s the situation?
• Instructions: Explain what you want the AI to do.
• Details: Add any specifics you need in the output (tone, format, audience, etc.)
• Input: Include any source materials or examples you'd like the AI to work from or refer to.

Although admittedly a little more time consuming, we might reap the benefits of creating more intentional, detailed prompts – particularly if you are using AI repeatedly for similar activities or tasks.

For anyone interested in learning more and seeing some examples of the CIDI framework in practice there is a short LinkedIn Learning course (13m 6s).”

Happy prompt engineering!