We present the results of hydrodynamic simulations of the interaction between a 10 Jupiter mass planet and a red or asymptotic giant branch stars, both with a zero-age main sequence mass of 3.5 $M_odot$. Dynamic in-spiral timescales are of the order of few years and a few decades for the red and asymptotic giant branch stars, respectively. The planets will eventually be destroyed at a separation from the core of the giants smaller than the resolution of our simulations, either through evaporation or tidal disruption. As the planets in-spiral, the giant stars envelopes are somewhat puffed up. Based on relatively long timescales and even considering the fact that further in-spiral should take place before the planets are destroyed, we predict that the merger would be difficult to observe, with only a relatively small, slow brightening. Very little mass is unbound in the process. These conclusions may change if the planets orbit enhances the stars main pulsation modes. Based on the angular momentum transfer, we also suspect that this star-planet interaction may be unable to lead to large scale outflows via the rotation-mediated dynamo effect of Nordhaus and Blackman. Detectable pollution from the destroyed planets would only result for the lightest, lowest metallicity stars. We furthermore find that in both simulations the planets move through the outer stellar envelopes at Mach-3 to Mach-5, reaching Mach-1 towards the end of the simulations. The gravitational drag force decreases and the in-spiral slows down at the sonic transition, as predicted analytically.