Background and objective: Hypoxic exercise exacerbates periodic breathing in otherwise healthy, awake humans. Interactions between sleep, exercise and hypoxic exposure have not been fully elucidated. Methods: Fourteen men were confined 10 days to a simulated altitude of 4175 m (FIO2 = 0.139; PIO2 = 88 mm Hg). They were randomly assigned to an exercise intervention of 2 × 60-min cycle exercise/day at 50% of their hypoxia-specific peak power output (exercise, n = 8), or they completed no exercise (control, n = 6, random order). Sleep and breathing were objectively assessed via full polysomnography on night 1, after 14-h acute exposure (N1), and again on night 10 (N10). Results: The exercise group spent more time in light sleep than control on N10 (95% confidence interval (CI): 8.5-15.0%; P = 0.013) and experienced more stage shifts (CI: 13-44; P = 0.023) on both nights compared with control. The exercise group experienced more apnoea-hypopnoea (AH) events per hour compared with control (CI: 1-110; P = 0.046); AH events that were associated with night desaturations were also higher on N1 (exercise: 397 ± 320, control: 124 ± 205, P = 0.047) and N10 (exercise: 375 ± 229, control: 110 ± 138, P = 0.028, CI: 49-489 total events; P = 0.020). The length of hyperpnoea was increased from 12.8 ± 2.2 s on N1 to 14.6 ± 2.7 s on N10 (P = 0.008), and thus, total cycle length also increased (P = 0.002) in both cohorts. Mean pooled duty ratios were 0.68 ± 0.02 on N1 and 0.69 ± 0.02 on N10 (group effect P = 0.617). Conclusion: Daily, moderate-intensity exercise in normobaric hypoxia equivalent to 4175 m exacerbated AH events, and negatively affected sleep architecture in exercisers compared with matched controls.