指导
背景(考古学)
知识管理
控制(管理)
培训(气象学)
顺从(心理学)
经验证据
心理学
实证研究
计算机科学
应用心理学
知识共享
教学设计
功能(生物学)
探索性研究
主动学习(机器学习)
认知
知识获取
德雷福斯技能获得模型
知识体系
考试(生物学)
价值(数学)
医学教育
过程(计算)
非正式学习
核心知识
作者
Qadri H. Shaheen,Abdolmajid Erfani,Moyassar Y. Mohammed,Dongyang Zhen,Qingbin Cui
标识
DOI:10.1061/jmenea.meeng-7190
摘要
This study explores the role of generative artificial intelligence (Gen-AI) as a compliance training agent by examining how interactive prompt patterns influence knowledge acquisition in the context of the Build America Buy America (BABA) Act. Amid growing interest in scalable, personalized learning tools, this experiment tests ChatGPT’s capacity to deliver regulatory education through adaptive coaching styles. The methodology integrates pre- and post-training assessments, control questions for knowledge retention, and coaching formats that range from scenario-based to interactive and gamified prompts. Results reveal that Gen-AI-driven training significantly improves compliance knowledge, particularly when prompt structures are aligned with learners’ prior understanding. Participants with low initial knowledge exhibited the greatest gains, indicating the value of real-time feedback and conversational scaffolding in supporting novice learners. Furthermore, control items confirm that learning was substantive rather than superficial, reinforcing Gen-AI’s potential for anchored instruction. However, the effectiveness of prompt types varied by knowledge tier, with structured formats outperforming more exploratory ones specifically among less-prepared participants. These findings carry meaningful implications for public-sector training: while Gen-AI training agents can extend access and personalize instruction, their design must consider learner readiness and cognitive load. The study also identifies prompt development as a core challenge, requiring balance between engagement, clarity, and instructional depth. As such, this research contributes both empirical evidence and practical guidance for integrating Gen-AI into compliance training, emphasizing the importance of pairing technological capability with pedagogical intent to ensure measurable learning outcomes.
科研通智能强力驱动
Strongly Powered by AbleSci AI