Abstract
Psychological scale development has traditionally required extensive expert involvement, iterative re-vision, andlarge-scalepilottestingbeforepsychometricevaluationcanbegin. TheAIGENIE Rpackageim-plementstheAI-GENIEframework(AutomaticItemGenerationandValidationwithNetwork-IntegratedEvaluation), which integrates large language model (LLM) text generation with network psychometricmethods to automate the early stages of this process. The package generates candidate item pools usingLLMs, transforms them into high-dimensional embeddings, and applies a multi-step reduction pipeline— Exploratory Graph Analysis (EGA), Unique Variable Analysis (UVA), and bootstrap EGA — toproduce structurally validated item pools entirely in silico. This tutorial introduces the package acrosseight parts: installation and setup, text generation, embeddings, item generation, the full AI-GENIEpipeline, the GENIE pipeline for researcher-supplied items, advanced prompt engineering, and fully lo-cal operation. Two running examples illustrate the package’s use: the Big Five personality model (awell-established construct) and AI Anxiety (an emerging construct). The package supports multipleLLM providers (OpenAI, Anthropic, Groq, HuggingFace, and local models), offers a fully offline modewith no external API calls, and provides the GENIE() function for researchers who wish to apply thepsychometric reduction pipeline to existing item pools regardless of their origin. The AIGENIE packageis freely available on CRAN at https://CRAN.R-project.org/package=AIGENIE.