Background: Proximal caries is a caries that formed between one to another teeth and often can not be detected visually or with the help of explorer, so it is necessary to do radiography examination. Radiographic examination to detect caries is using bitewing techniques. Conventional radiography uses image receptors in the form of films that require a dark room in the process, using chemicals and difficult in storage or distribution. Computed radiography uses photostimulable phospor imaging plates which eliminate disadvanteges of conventional radiography and also reduce the dose of radiation received by patients. Aims: To know the is different in the accuracy results in conventional radiography and CR using bitewing techniques for artificial proximal caries detection. Method: This research used analytic observational method with cross-sectional design. Samples in the form of 40 premolar that prepared using round burs on the mesial part 1 mm deep as a simulation of proximal caries. The sample is then planted in a cast model and fixed using plasticine. Each cast model was planted with 2 maxillary premolars and 2 in mandible. The cast model is then x-rayed with conventional radiography and CR. X-rays are then seen in the viewer for interpretation of whether artificial proximal caries that have been made are seen or not. Result: Conventional radiography can detect 33 artificial proximal caries, in contrary, CR can detect 35 artificial proximal caries. The Chi-square statistic test results stated that the difference was not significant. Conclusion: There is no significant difference in accuracy between the results of conventional radiography and CR in bitewing techniques to detect artificial proximal caries.