The reduction of information capacity of the ocean sound channel due to scattering by internal waves is a potential problem for acoustic communication, navigation, and remote sensing over long ranges. In spite of recent progress in research on acoustic signal scattering by random internal waves and the fact that random internal waves are ubiquitous in the world oceans, there is no clear understanding of how these waves influence data communication performance. The entropy decrease resulting from scattering by internal waves is an important measure of information loss. Here a rigorous calculation of the entropy is carried out using second moment transport theory equations with random sound-speed perturbations obeying the Garrett-Munk internal-wave model. It is shown that full-wave rate of entropy is of the same order of magnitude as the Kolmogorov-Sinai entropy and Lyapunov exponents for the relevant ray trajectories. The correspondence between full-wave and ray entropies suggests a correspondence between full-wave scattering and ray chaos near statistical saturation. The relatively small level of entropy rate during propagation through the random internal-wave field shows that scattering by internal waves is likely not an essential limitation for data rate and channel capacity.