Interrogating the Machine: Integrating Hercule Poirot's “Little Grey Cells” Deductive Framework into AI Literacy and Critical Prompt Engineering in Education

Fauziah Nur Aisyah Rosyidah (Universitas Negeri Malang)

Proceedings Series of Educational Studies
2026: International Conference on Educational Management and Technology (5th ICEMT 2026)

Generative artificial intelligence has entered classrooms faster than the verification habits required to use it safely. This study examines the mismatch between adoption, model reliability, and student verification behaviour, and derives a cognitive scaffold from an unexpected source: the deductive procedure of the fictional detective Hercule Poirot. A systematic literature review of studies published between 2021 and 2026 was combined with descriptive time-series analysis of sixty-one months of global relative search-volume data for four query strings. The search data show that interest in operating generative systems reached the normalised ceiling of the index, whereas interest in governing them within schools remained at the measurement floor, returning zero in fifty-five of sixty-one months. The reviewed literature converges on three findings: high adoption among students, persistent model hallucination on complex academic queries, and low rates of systematic cross-checking against primary sources. From this convergence the study derives a four- stage framework detailed observation, contradiction identification, alibi verification, and logical conclusion that repositions the chatbot as a witness under examination rather than an authority. The framework is conceptual and awaits empirical validation.