The prevalence of antibiotic-resistant bacterial infections has raised great social concern and even aggravated the burden of public healthcare systems worldwide. Simple, rapid, highly sensitive and specific detection of antibiotic-resistant bacteria is urgently needed for guiding proper antibiotic treatment. Herein, we demonstrate a novel strategy to specifically distinguish and detect antibiotic-resistant bacteria using CRISPR/Cas9 induced isothermal exponential amplification reaction (IEXPAR). Through the specific recognition and cleavage of CRISPR/Cas9 system, the antibiotic-resistant genes can be digested into two short segments with free 3ʹ-OH end. Using one of the cleaved DNA as target to design IEXPAR template (Xʹ-Y-Xʹ), highly efficient exponential amplification of IEXPAR can be triggered, resulting in the direct detection of antibiotic-resistant genes under isothermal condition with high accuracy and rapidness. While for antibiotic-sensitive bacteria, CRISPR/Cas9 system cannot cleave any DNA due to the absence of antibiotic-resistant genes and no subsequent amplification reaction occurs which ensures the high specificity of our method. Further applying our method to the detection of real biological samples, the specific discrimination of antibiotic-resistant bacteria from antibiotic-sensitive bacteria is realized even with large amounts of DNA interferents. In consideration of the distinct advantages of rapid, high specificity and facile operation, this CRISPR/Cas9-induced IEXPAR assay may serve as an efficient and convenient tool for the accurate detection of antibiotic-resistant bacteria in complex biological samples. • A CRISPR/Cas9 induced isothermal exponential amplification reaction is developed. • CRISPR/Cas9 can specifically recognize and cleave antibiotic-resistant genes. • One of the cleaved genes can trigger IEXPAR using simply designed template. • Antibiotic-resistant bacteria can directly discriminate from other bacteria. • This method has applied to the detection of antibiotic-resistant bacteria in biological samples.