Time constraints are cited as a barrier to regular exercise. If particular exercises can achieve multiple training functions, the number of exercises and the time needed to achieve a training goal may be decreased. It was the objective of this study to compare the extent of trunk muscle electromyographic (EMG) activity during running and callisthenic activities. EMG activity of the external obliques, lower abdominals (LA), upper lumbar erector spinae (ULES), and lumbosacral erector spinae (LSES) was monitored while triathletes and active nonrunners ran on a treadmill for 30 min at 60% and 80% of their maximum heart rate (HR) reserve, as well as during 30 repetitions of a partial curl-up and 3 min of a modified Biering-Sørensen back extension exercise. The mean root mean square (RMS) amplitude of the EMG signal was monitored over 10-s periods with measures normalized to a maximum voluntary contraction rotating curl-up (external obliques), hollowing exercise (LA), or back extension (ULES and LSES). A main effect for group was that triathletes had greater overall activation of the external obliques (p < 0.05), LA (p = 0.01), and LSES (p < 0.05) than did nonrunners. Main effects for exercise type showed that the external obliques had less EMG activity during 60% and 80% runs, respectively, than with the curl-ups (p = 0.001). The back extension exercise provided less ULES (p = 0.009) and LSES (p = 0.0001) EMG activity than the 60% and 80% runs, respectively. In conclusion, triathletes had greater trunk activation than nonrunners did while running, which could have contributed to their better performance. Back-stabilizing muscles can be activated more effectively with running than with a prolonged back extension activity. Running can be considered as an efficient, multifunctional exercise combining cardiovascular and trunk endurance benefits.