Artificial intelligence (AI) is now a linchpin of communication. In principle, an AI response is dependent on the linguistic design of prompts, which is a research priority nowadays. This study touches on linguistic structures of prompts in human-AI discourse: the syntactic, lexical, and pragmatic. The main purpose of this analysis is to examine how such linguistic elements impact the quality, relevance, and coherence of AI-generated responses. A mixed-method approach was adopted to combine corpus analysis and qualitative linguistic inquiry based on 250 prompts collected from 60 AI users at the University of Bisha, Saudi Arabia. The analysis included sentence type, clause complexity, lexical specificity, and discourse cohesion. AI outputs were evaluated to assess clarity, accuracy, and completeness. Findings show cross-model tendencies in how certain linguistic features of prompts are associated with response quality. Prompts in the form of clear imperatives and interrogatives tend to create more context-sensitive responses than vague or declarative prompts. AI interpretive performance stems from explicit lexical cues, cohesive markers, and precise syntactic construction. Both linguistic form and AI users’ prompt literacy mediate effective human-AI interaction. The study, documenting cross-model tendencies rather than mechanisms tied to specific LLM architectures, provides a linguistic approach to effective prompt design, which is a contribution to applied linguistics and language education. It also bridges theoretical linguistics and AI regarding how human language structure shapes machine understanding in AI communicational discourse.
Ghazwan Mohammed Saeed Mohammed (Sat,) studied this question.