The ZED is based on passive RGB stereovision. Unlike active depth sensors, the ZED truly replicates the way human vision works. As a rule of thumb, if you see something, the camera will most probably see, but depth accuracy will be decreased due to the weak information available in the image. Note however that the dynamic range of human eyes is far superior to the one of any camera.
If the scene is completely dark, the ZED camera will not be able to capture depth. Also, turning on and off the light will affect the quality of the depth map.