Since Bruce Willis' retirement the issue of whether or not the controllability of climate models could save us at the last minute, has considerably strengthened. The core of these models is the Navier-Stokes equations, a quite intricate non-linear PDE system for which prediction is already difficult. However, John Von Neumann in the 70's, and later Jacques-Louis Lions in the late 80's, made public their intuition that controlling this system, that is driving the state from a given initial data to a targeted final state in finite time, could be achieved, possibly with a technically realistic forcing term. The mathematical analysis of this problem has started with the analysis of the heat equation in the 90's, then of the incompressible Euler equations, followed by some partial results on the Navier-Stokes system. Still issues due to boundary layers, robustness and cost are quite largely in front of us.