This research presents a novel "Reasoning and Value Alignment Test" (RVAT) to evaluate the reasoning capabilities of advanced generative pre-trained transformers (GPTs), with a focus on their ability to reason in culturally complex contexts. The study addresses the limitations of current AI reasoning assessments, which often overlook cultural nuances and fail to fully automate evaluations due to the complexity of cultural contexts. The RVAT was designed to assess GPT models' reasoning abilities in alignment with local human values, emphasizing the importance of cultural sensitivity in AI development. The test involves evaluating GPT models on 200 questions across 20 domains, assessing their ability to reason with raw facts and interpretations of facts. The results indicate that while GPT models exhibit high levels of human-like reasoning, they still struggle with interpreting cultural contexts, highlighting the need for further research and development in this area. The study also explores the broader implications of the RVAT in the context of artificial general intelligence (AGI), emphasizing the necessity for interdisciplinary approaches and a deeper understanding of the interplay between GPT reasoning and cultural sensitivity. The findings underscore the importance of culturally adapted AI systems and the need for rigorous assessment to ensure alignment with both universal and local human values. The RVAT provides a framework for evaluating AI reasoning capabilities, contributing to the ongoing discourse on AI safety, ethics, and cultural alignment.This research presents a novel "Reasoning and Value Alignment Test" (RVAT) to evaluate the reasoning capabilities of advanced generative pre-trained transformers (GPTs), with a focus on their ability to reason in culturally complex contexts. The study addresses the limitations of current AI reasoning assessments, which often overlook cultural nuances and fail to fully automate evaluations due to the complexity of cultural contexts. The RVAT was designed to assess GPT models' reasoning abilities in alignment with local human values, emphasizing the importance of cultural sensitivity in AI development. The test involves evaluating GPT models on 200 questions across 20 domains, assessing their ability to reason with raw facts and interpretations of facts. The results indicate that while GPT models exhibit high levels of human-like reasoning, they still struggle with interpreting cultural contexts, highlighting the need for further research and development in this area. The study also explores the broader implications of the RVAT in the context of artificial general intelligence (AGI), emphasizing the necessity for interdisciplinary approaches and a deeper understanding of the interplay between GPT reasoning and cultural sensitivity. The findings underscore the importance of culturally adapted AI systems and the need for rigorous assessment to ensure alignment with both universal and local human values. The RVAT provides a framework for evaluating AI reasoning capabilities, contributing to the ongoing discourse on AI safety, ethics, and cultural alignment.