AI-Enhanced Experiments shows researchers how to use large language models alongside reproducible R workflows to design, generate, and validate multi-version stimuli and short chat-based interventions while explicitly addressing validity, ethics, and documentation. Participants leave with practical prompt templates, R scripts, and validation routines to produce scalable, preregistration-ready experimental treatments for political behavior, public policy, and related social science research.
People increasingly use frontier AI models for health advice, but via different access modes (e.g., ChatGPT, ChatGPT Health, APIs) with varying settings. Here, we find systematic differences across access modes. Because evaluations typically rely on APIs while consumers interact through chatbot interfaces, these discre...
Yuan Pu, Yew-On Chang, Fu-Rong Jia et al.· 0 citations
Artificial intelligence (AI) is transforming scientific practice. What started as a passive data-analysis tool has become something closer to an active partner, now capable of generating hypotheses, planning experiments, and coordinating multi-step workflows. This manuscript explores how large language models (LLMs) an...
Graziella Orrú, Pietro Pietrini, Giuseppe Sartori· Forensic Sciences Research· 0 citations
This paper maps key evaluation challenges and proposes three frameworks specific to these tasks and offers design teams a practical roadmap to incorporate continuous, evidence-based testing, promoting greater rigor and accountability in AI-assisted design practice.
Current limitations both of LLMs development and implementation are discussed from a unique perspective : crossing the technical expertise and field research from social sciences, present paper demonstrates how the problems faced by models transcends pure technicality and echoes on user’s level. Scaling and architectur...
Anna Shcherbakova· International Journal of Blo...· 0 citations
