Background <p>The rapid advancements in artificial intelligence (AI) technologies are fundamentally transforming mathematics teaching processes and offering new pedagogical opportunities within instructional environments. However, the effective use of these technologies is closely related to mathematics teachers’ levels of knowledge, awareness, attitudes, and skills regarding AI. The purpose of this study is to examine the relationship between mathematics teachers’ AI literacy and AI anxiety, to conduct an in-depth analysis of their perceptions regarding the integration of AI into mathematics education, and to evaluate the effects of variables such as watching AI-related films, technology use, and age on this process.</p> Methods <p>This study employed a mixed-methods design. In the quantitative phase, a predictive correlational model was employed, while in the qualitative phase, a case study approach was utilized. Data were collected from 251 mathematics teachers working in various regions of Türkiye. The quantitative data were analyzed using a range of statistical analysis techniques, whereas the qualitative data were evaluated through content analysis.</p> Results <p>The findings indicate that mathematics teachers’ levels of AI literacy are above average, whereas their levels of AI anxiety are below average. A significant and negative relationship was found between AI literacy and AI anxiety. Furthermore, the level of technology use in mathematics instruction was identified as the strongest predictor of both AI literacy and AI anxiety. The results also revealed that mathematics teachers’ most prominent anxiety is that the excessive use of AI tools may weaken students’ independent thinking and problem-solving skills. In addition, anxiety regarding the potential weakening of the teaching role and the possibility that AI could replace teachers were also noteworthy.</p> Conclusions <p>Professional development programs should encompass not only the fundamental technological features of AI but also its pedagogical contributions to mathematics instruction. Mathematics teachers should be provided with opportunities to observe how AI supports key instructional processes such as differentiated instruction, formative assessment, and conceptual visualization. Furthermore, training modules should aim to develop teachers’ abilities to critically evaluate AI-generated mathematical content in terms of accuracy and pedagogical appropriateness. Through such targeted training, teachers can enhance their AI literacy and create safe and pedagogically meaningful digital learning environments for their students.</p>