Crafting Effective Prompts: Thinking with Machines Not Just Talking to Them

This chapter develops a conceptual and practical foundation for crafting effective prompts in the context of generative artificial intelligence (Gen AI) use in academia. Moving beyond the view of prompts as simple instructions, the chapter positions prompting as a form of cognitive framing that shapes how large language models (LLMs) interpret tasks, generate outputs, and support scholarly work. It introduces different types of prompts – informational, instructional, creative, and conversational – and demonstrates how each aligns with distinct academic purposes such as knowledge retrieval, teaching design, research ideation, and human–AI interaction. Drawing on emerging research in prompt engineering and human–AI interaction, the chapter highlights how specificity, context, and structure influence output quality. It further examines the role of iterative prompting, showing how refinement, constraint and dialogic engagement can transform AI from a passive tool into an active thinking partner. At the same time, the chapter critically engages with the limitations of Gen AI, including risks of superficial coherence and epistemic unreliability.

Chapter Goal: Provide an approach to writing initial prompts