Although exercise promotes beneficial effects in diabetic patients, some studies have questioned the degree of their importance in terms of the increase in total energy expenditure. In these studies, the decrease of physical activity levels (PAL) was referred as “compensatory effect of exercise”. However, our aim was to investigate whether aerobic exercise has compensatory effects on PAL in type 2 diabetes patients. Eight volunteers (51.1 ± 8.2 years) were enrolled in a supervised exercise programme for 8 weeks (3 d · wk−1, 50–60% of VO2 peak for 30–60 min). PAL was measured using tri-axial accelerometers in the 1st, 8th and 12th weeks. Biochemical tests, cardiorespiratory fitness, anthropometric assessment and body composition were measured in the 2nd and 11th weeks. Statistical analysis was performed using non-parametric tests (Friedman and Wilcoxon, P < 0.05). We found no significant differences in PAL between intervention periods, and participants spent the majority of their awake time in sedentary activities. However, the exercise programme generated a significant 14.8% increase in VO2 peak and a 15% reduction in fructosamine. The exercise programme had no compensatory effects on PAL in type 2 diabetes patients, but improved their cardiorespiratory fitness and glycaemic control.