In November of 2022, the company OpenAI made available for free public use a research release of one of their large language models called chat generative pre-trained transformer (ChatGPT).1 ChatGPT enables users to interact with the GPT artificial intelligence (AI) model in a way that is intended to be comfortable and safe. At the same time, the widespread availability of such an AI platform permanently changed the education landscape. It suddenly became possible for students to recruit the assistance of a sophisticated AI language model with only a computer and an internet connection. During the spring 2023 semester, many students at our institution did just that, prompting questions about academic integrity and the use of AI in the classroom.

The potential uses of an AI of the caliber of ChatGPT by physics students and educators cover an enormous range, including the automated generation of writing, background research on basic physics topics, detailed feedback on the content and clarity of human-written text, writing computer code, solving physics problems, and answering conceptual physics questions, to name a few.2 The challenges that the widespread use of AI brings are equally numerous, including the risk of spreading misinformation, the presence and effect of biases in the model, concerns about academic and professional integrity, ethical questions about the proper use of AI models, and questions of equity and access. To compare the adoption of AI tools to the introduction of the pocket calculator drastically underestimates the impact AI will have on education.

As a result, physics educators need to be familiar with the capabilities and limitations of AI models. AI models like the one behind ChatGPT are already developing quickly with increased use, and their presence in academia will only become more common.3 In our opinion, physics educators need to take the widespread use of AI by students seriously and study how the use of such models impacts physics education.

In addition to solving physics problems and answering conceptual questions,2 physics educators should pay particular attention to the area of physics writing. Previous work has pointed to the learning opportunities present when students write about physics concepts and experiments.4–6 The untrained use of AI tools like ChatGPT by students can easily undermine physics writing instruction by allowing them to avoid the work of engaging physics concepts, clarifying their thought, and expressing themselves in writing.

We as physics educators are in an excellent position to help students responsibly navigate this new terrain and be prepared for the quickly evolving professional landscape. However, we can only be effective educators if we understand clearly the impact of AI tools on student learning. Physics educators should collaborate to study the impact of AI tools on student writing and learning and the best way to manage its use for the benefit of education. For example, journals like the American Journal of Physics, The Physics Teacher, and Physical Review Physics Education Research should consider dedicating one issue each year to the topic of AI and physics education. Similarly, the American Association of Physics Teachers could host a regular panel discussion on AI and physics education, especially at the national meetings.

Physics educators must take seriously the use of AI by students and work to understand its impact on physics writing and physics education as a whole. If we as physics educators do not decide if and how AI should be used by physics students in the educational setting, then we leave AI to decide for us.

1.
OpenAI
, ChatGPT research preview, <https://chat.openai.com/>.
2.
J.
Wang
, “
ChatGPT: A test drive
,”
Am. J. Phys.
91
(
4
),
255–256
(
2023
).
3.
Chris
Stokel-Walker
and
Richard
van Noorden
, “
What ChatGPT and generative AI mean for science
,”
Nature
614
(
7947
),
214
216
(
2023
).
4.
Don S.
Lemons
, “
Focused writing in the introductory physics laboratory
,”
Am. J. Phys.
89
(
10
),
907
(
2021
).
5.
Kristen L.
Thompson
,
Anthony N.
Kuchera
, and
John N.
Yukich
, “
Teaching college writing from a physicist's perspective
,”
Am. J. Phys.
89
(
1
),
61
66
(
2021
).
6.
Jessica R.
Hoehn
and
H. J.
Lewandowski
, “
Framework of goals for writing in physics lab classes
,”
Phys. Rev. Phys. Ed. Res.
16
(
1
),
010125
(
2020
).