Amid the growing interest in generative AI technologies like ChatGPT for educational purposes, this research seeks to better understand Chinese EFL learners' acceptance and usage of them for L2 writing purposes. This study conducts an investigation by establishing a structural equation model that incorporates the Technology Acceptance Model (TAM) and Task Technology Fit (TTF) theories. The survey involving 304 university students in China discovered the important roles of the perceived usefulness and perceived ease of use in explaining Chinese EFL learners' attitude, which then influenced their behavioral intention to accept generative AI instruments as an English writing assistant. The other major finding revealed that task technology fit significantly affected their behavioral intention and made an impact through the meditation of attitude. By examining these constructs, the research presents fresh perspectives about the effective integration of generative AI in language education, contributing empirical evidence to promote theoretical understanding and practical applications in L2 English writing instruction.