Our knowledge of the direct role of clouds in long-term climate change is examined in an overview of key results published over the last 15 or 20 years, along with some relevant unpublished model studies. The focus is on 1) the impact of clouds on the incoming and outgoing radiation at the top of the atmosphere, and 2) the two-way interaction of clouds with other variables of the climate system—i.e., the cloud/climate feedback problem—as revealed by climate model simulations. A common framework is established for comparing results from different investigations. The total effect of clouds on radiative fluxes at the top of the atmosphere—specifically, the difference in flux between average conditions and cloud-free conditions, often called cloud forcing—has been derived from earth radiation budget measurements by several investigators using various data sources and methods. There is general agreement that the annual global mean effect of clouds is to cool the climate system, but there is significant disagreement on magnitude, with the two investigations based on recent satellite data indicating a range from 17 to 27 W/m2. Cloud sensitivity, which represents the differential response of top-of-the-atmosphere fluxes to changes in cloud cover parameters, is a critical factor in cloud feedback. The observational data at this point in time, however, are not sufficient to yield more than rough estimates. Two estimates, of the sensitivity to cloud amount, show wide discrepancies. To be useful, future studies of sensitivity will have to separate different cloud types. Sensitivity of clouds to cloud condensation nuclei raises the issue of a more direct role of clouds in climate change, where aerosols associated with SO2 emissions can ultimately lead to brighter clouds and a reduction in solar heating. On cloud feedback in climate simulations, there are wide discrepancies amongst models. Investigators find, in general, that the sign of cloud feedback is positive when one allows the spatial distribution of clouds to vary in response to climate change. When one additionally takes into account the effect of changes in cloud properties, that result is less certain. The conclusion is: 1) clouds may have a strong influence on climate change, but 2) we are far from knowing the magnitude, and even the sign, of this influence.