Potential of ChatGPT in Mathematics Education

The adoption of technologies such as ChatGPT, one of the powerful chatbots supported

by artificial intelligence, in higher education institutions can positively affect various

academic activities as it will make it easier to respond to individual student needs

(Dempere et al., 2023). The purpose of this study is to examine the effect of ChatGPT on

the achievement of pre-service teachers in mathematics education. Discussions of new

technologies such as ChatGPT from an educational perspective are common as they often

disrupt traditional practices and require teachers to adapt to their potential benefits and

drawbacks (Qadir, 2022). Hwang and Tu (2021) stated that the application of AI in

mathematics education is still in its infancy, with researchers focusing on using AI

technologies to solve basic problems in mathematics courses. It is thought that the results

of the study will contribute to the expanding literature on whether ChatGPT can be used

in mathematics education. The study was conducted with 65 pre-service primary school

teachers studying at a state university within the framework of a pre-test/post-test design

with a control group. ChatGPT 3.5 was used because it was free of charge and easy to

access. An Achievement Test consisting of two open-ended questions and screenshots of

all interactions of pre-service teachers in the learning process with ChatGPT were used as

data collection tools. In the implementation phase of the study, while the teacher provided

the instruction in the control group, ChatGPT alone assumed the role of the teacher in the

experimental group. The data obtained were analyzed using the SPSS 18.0 program and

interpreted by the researchers. According to the results of the study, it was concluded that

teaching by the teacher increased the achievement of pre-service teachers more than

teaching with ChatGPT. It is thought that this situation may be due to the quality of

ChatGPT's answers to mathematics questions, the fact that most of the pre-service

teachers encountered ChatGPT for the first time and had little experience in using it, the

pre-service teachers' lack of confidence in ChatGPT's answers, and the unsatisfactory

ChatGPT answers. For this reason, it is necessary to emphasize that assigning a teacher

role to ChatGPT in the education and training process and conducting this process only

with ChatGPT is not effective.

Keywords: ChatGPT, mathematics education, AIEd, artificial intelligence