Effective Prompting in Higher Education
Best Practices
Prompt engineering involves the creation of prompts (text, audio, visual, computer code) to give precise and unambiguous instructions to GenAI tools. Various best practices exist to guarantee safe, ethical, and intended system behaviour. Prompts should be created to be straightforward, brief, and include pertinent examples. They should undergo rigorous testing to assess accuracy, detect biases, and uncover unintended consequences, with identified issues addressed through iterative refinement. Implementing protective measures, soliciting diverse feedback, and maintaining comprehensive documentation are also crucial.
Learn more below.
Useful Resoruces:
The first recommended resource is an insightful guide by Danny Liu from the University of Sydney, titled “Prompt Engineering for Educators – Making Generative AI Work for You.” This comprehensive guide offers educators a deep dive into prompt engineering, with practical examples tailored for the educational context.
Prompt engineering for educators – making generative AI work for you (Liu, 2023)
The next resource shifts our focus towards problem formulation. Oguz A. Acar’s article emphasises the significance of clearly defining and understanding the problem at hand before engaging with generative AI. This approach to problem formulation is presented as a foundational skill that complements prompt engineering, urging educators to consider the broader context and nuances of the issues they wish to address with AI. By prioritising a well-defined problem, educators can ensure that their use of generative AI is both purposeful and impactful.
A final resource is a GenAI Chatbot Prompt Library for Educators is a really practical tool. This comprehensive library offers a curated selection of prompts tailored for various educational tasks, ranging from lesson planning to administrative duties.
GenAI Chatbot Prompt Library for Educators
Final Considerations:
Prompting in GenAI is an iterative and conversational process, akin to a dynamic dialogue where each exchange refines the next. Clear problem formulation is key, as prompts can vary greatly in their effectiveness. Remember, the process involves constant refinement and feedback, making each prompt a step in an ongoing cycle of improvement.
References
Acar, O. A. (2023). Ai prompt engineering isn’t the future. Retrieved from https://hbr.org/2023/06/ai-prompt-engineering-isnt-the-future.
Liu, D. (2023). Prompt engineering for educators – making generative AI work for you. Retrieved from https://educational-innovation.sydney.edu.au/teaching@sydney/prompt-engineering-for-educators-making-generative-ai-work-for-you/.
