Artificial Intelligence Driven Physics Assignments using Context Prompts
Abstract
Generative artificial intelligence (Gen AI) is increasingly shaping how students access information and engage with physics content, yet many instructional approaches still assume a pre-AI classroom. In this paper, we describe a practical framework for integrating Gen AI into physics instruction through the use of context prompts: instructor-designed prompts that define the role, task, and boundaries of the AI interaction. We focus on AI-supported exploration assignments implemented in an asynchronous introductory physics course enrolling approximately 70 students. In these activities, students used AI as a guided tutor to explore selected topics, ask follow-up questions, request explanations and analogies, and connect physics concepts to their personal interests and everyday experiences. We argue that this structure can help move students beyond answer-seeking and toward more dialogic and personally meaningful engagement with course content. We also discuss reflection as a natural extension of the same framework, while distinguishing it from the implemented exploration activities described here. Finally, we highlight practical considerations that emerged in use, including uneven access to AI platforms, grading workload, the need to teach students how to engage AI conversationally, and the growing importance of AI literacy in physics education.
Related
- Advancing Nuclear Physics with Machine Learning and Artificial Intelligence
- From Prompts to Physical Laws: A Generative AI Workflow for Engineering Physics Education
- From Local Context to Climate Science: Low-Cost Experimentation in the Initial Training of Physics Teachers
- From the Physics of Society to a Sociology of Artificial Agents
- Using artificial intelligence for clinical trial quality assurance in a multi-institutional neurovascular-sparing prostate radiotherapy trial
BibTeX
@misc{rodriguez2026artificial,
title = {{Artificial Intelligence Driven Physics Assignments using Context Prompts}},
author = {Miguel Rodriguez and Peter Wulff},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.22578},
eprint = {2609.22578},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.22578},
note = {CC0},
}
RIS
TY - UNPB
TI - Artificial Intelligence Driven Physics Assignments using Context Prompts
AU - Miguel Rodriguez
AU - Peter Wulff
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.22578
UR - https://arxiv.org/abs/2609.22578
AB - Generative artificial intelligence (Gen AI) is increasingly shaping how students access information and engage with physics content, yet many instructional approaches still assume a pre-AI classroom. In this paper, we describe a practical framework for integrating Gen AI into physics instruction through the use of context prompts: instructor-designed prompts that define the role, task, and boundaries of the AI interaction. We focus on AI-supported exploration assignments implemented in an asynchronous introductory physics course enrolling approximately 70 students. In these activities, students used AI as a guided tutor to explore selected topics, ask follow-up questions, request explanations and analogies, and connect physics concepts to their personal interests and everyday experiences. We argue that this structure can help move students beyond answer-seeking and toward more dialogic and personally meaningful engagement with course content. We also discuss reflection as a natural extension of the same framework, while distinguishing it from the implemented exploration activities described here. Finally, we highlight practical considerations that emerged in use, including uneven access to AI platforms, grading workload, the need to teach students how to engage AI conversationally, and the growing importance of AI literacy in physics education.
ER -