The discrimination of chicken spoilage is extremely important in order to guarantee the quality and safety of chicken before consumption, and requires an efficient and reliable method. We aimed to find a rapid and accurate method to discriminate the level of spoilage of chicken, and have developed a gas Fourier transform infrared spectroscopy (G‐FTIR) methodology based on analysing volatiles from chicken samples. In this method, the volatiles of chicken samples were first measured by the G‐FTIR spectrometer. After that, a chemometrics method combining principal component analysis (PCA) with support vector machine (SVM) was developed to encode the G‐FTIR spectra. With the combination of G‐FTIR spectra and the chemometrics method, a relationship between the spoilage levels of chicken and the G‐FTIR spectra of volatiles was established. As a result, the spoilage levels of chicken samples were accurately distinguished with a prediction accuracy of 100%, for samples stored at both normal (20 °C) and cold (4 °C) temperatures. The results obtained were correlated with the routine total volatile basis nitrogen (TVB‐N) method for validation. In addition, The G‐FTIR detection process for each sample could be performed within 3 min, requiring no chemical reagents. The G‐FTIR methodology performed well for the rapid discrimination of chicken spoilage, which could also be extended to the detection of other spoiled foods.