Background: Many studies have reported minor complications and disturbance of the gut microbiota after colonoscopy. Compared with air, carbon dioxide (CO2) insufflation could decrease minor complications, but its impact on gut microbiota remains unknown. Methods: Thirty-eight healthy subjects were assessed and twenty were randomized to receive either CO2 or air insufflation during colonoscopy. Neither the participants nor the staff involved in the follow-up knew which gas was used. Minor complications were assessed using symptom scores. Fecal samples were collected at eight time-points for microbiome analysis by full-length 16S rRNA gene amplicon analysis. Results: Baseline characteristics were similar in both groups. The recovery of minor complications after colonoscopy was faster in the CO2 group (the day of the colonoscopy) than in the air group (the day after the colonoscopy). There was no significant reduction in alpha diversity (species richness) of the first stool after colonoscopy in the CO2 group (115.0 ± 32.81 vs. 97.4 ± 42.31, p = 0.28) compared with the air group (123.8 ± 37.25 vs. 84.8 ± 31.67, p = 0.04). However, there were no differences in beta diversity between the groups. Linear discriminant analysis effect size (LEfSe) analysis indicated that anaerobic probiotics such as Bacteroides caccae, Bacteroides finegoldii and Bacteroides thetaiotaomicron were more abundant in the CO2 group than in the air group within 14 days after colonoscopy. On the contrary, the content of Escherichiacoli, Ruminococcus torques and Ruminococcus guavus was higher in the air group. Conclusions: CO2 is beneficial to gut microbiota homeostasis during colonoscopy in healthy subjects. The effects in patients with different diseases need to be further studied.