Why Johnny Can’t Prompt: How Non-AI Experts Try (and Fail) to Design LLM Prompts
By J.D. Zamfirescu-Pereira, Richmond Y. Wong, Bjoern Hartmann, Qian Yang. International Conference on Human Factors in Computing Systems, 2023
1,195 citations recorded by the source corpus.
Pre-trained large language models (“LLMs”) like GPT-3 can engage in fluent, multi-turn instruction-taking out-of-the-box, making them attractive materials for designing natural language interactions. Using natural language to steer LLM outputs (“prompting”) has emerged as an important design technique potentially accessible to non-AI-experts. Crafting effective prompts can be challenging, however, and prompt-based interactions are brittle. Here, we explore whether non-AI-experts can successfully engage in “end-user prompt engineering” using a design probe—a prototype LLM-based chatbot design tool supporting development and systematic evaluation of prompting strategies. Ultimately, our probe participants explored prompt designs opportunistically, not systematically, and struggled in ways echoing end-user programming systems and interactive machine learning systems. Expectations stemming from human-to-human instructional experiences, and a tendency to overgeneralize, were barriers to effective prompt design. These findings have implications for non-AI-expert-facing LLM-based tool design and for improving LLM-and-prompt literacy among programmers and the public, and present opportunities for further research.
Subfields: Artificial Intelligence, Safety Research, Software
View Paper: https://www.semanticscholar.org/paper/8f9e864fab09bbae4a46a2a62bb954db1a88eb3e
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Diyi Yang
- Jiaao Chen
- Chenguang Zhu
- Michael Zeng
- Omar Shaikh
- Caleb Ziems
- Werner Geyer
- Michael J. Muller
In Safety Research
- Werner Geyer
- Michael J. Muller
- Jan Marco Leimeister
- Justin D. Weisz
- Gabriela Hoefer
- Diyi Yang
- Rachel Miles
- Antonia Tolzin
Citations in the corpus
- Can Large Language Models Transform Computational Social Science? (2023, background): “In practice, users prompt iteratively until the model behavior seems locally reasonable to them (Zamfirescu-Pereira et al., 2023).”
- AI literacy and its implications for prompt engineering strategies (2024, methodology): “Zamfirescu-Pereira et al. (2023) examined the potential of prompt engineering by non-experts, using a prototype LLM-based chatbot design tool.”
- LlaVA-CoT: Let Vision Language Models Reason Step-By-Step (2024, neutral): “Additionally, instructing learning techniques for language models, including prompt tuning [65], in-context learning, and supervised fine-tuning [49], also contribute to improvements in visual reasoning capabilities.”
- Design Principles for Generative AI Applications (2024, background): “One emerging skill revolves around crafting efective natural language prompts, known as in-context learning [20, 40, 189] or prompt engineering [185, 193].”
- Automatic Prompt Optimization with "Gradient Descent" and Beam Search (2023, methodology): “In this way, we engadge the LLMs in a recursive feedback loop similar to the Socratic dialogues proposed by Zeng et al. (2022).”
- The Metacognitive Demands and Opportunities of Generative AI (2023, background): “Manual work (whether, when, how to apply GenAI in workflows) Automation strategy: understanding and adapting workflows … …”
- LLMs Get Lost In Multi-Turn Conversation (2025, positive): “Ubiquitous performance degradation over multi-turn interactions is likely a reason for low uptake of AI systems [73, 4, 28], particularly with novice users who are less skilled at providing complete, detailed instructions from the onset of conversation [87, 35].”