This study explores the perceptions of second language (L2) educators on the surge of Large Language Models (LLMs) like ChatGPT, and their potential impact on language education. We surveyed over 100 L2 instructors, asking questions about their ideas for AI-proofing assignments, their policies, and their perceptions of how this tool will impact the profession. Data was collected through an anonymous survey and analyzed with a mixed-method approach through a constructivist lens. The results show mixed feeling: ranging from excitement for the pedagogical potential of ChatGPT to fear for potential academic dishonesty and job security in some. The results analysis provides insight into L2 educators use and policies of ChatGPT. The quantitative data highlighted that instructors' levels of excitement or concern about the adoption of ChatGPT in language education correlate with their personal experience with the tool, in line with Bax's (2003) normalization framework. Those with prior interaction with ChatGPT showed more enthusiasm for its educational potential than those without. Interestingly, this study found no significant difference in attitudes across different age groups or years of teaching experience. The qualitative data show that L2 educators anticipate wide use of ChatGPT in their students' assignments, yet a large majority (90%) is confident in their ability to identify students use of Chat GPT and Google Translate in their work. This study also reviews the current use and policies regarding MT and ChatGPT, highlighting a significant number of instructors prohibiting their uses. The results also highlight different ways instructors have been AI proofing their assignments and how they are currently using ChatGPT in their professions. In conclusion, this study advocates for a proactive use of these tools emphasizing the importance of adopting innovative pedagogies, a student-centered classroom approach, and certain ethical considerations.