The wind-energy industry is growing rapidly and its growth is expected to undergo a major spurt in the immediate future due to major global efforts in curbing the greenhouse gas emissions. The increased attention on harnessing wind power has led to the development of larger and more flexible wind turbines. For example, turbine-blade diameters have increased from a modest 15m to a phenomenal 120m in the last 30 years. However, with such giant structures and increased flexibility, design issues hitherto unforeseen have surfaced: for example, complex oscillation modes, structural damage due to edge-wise vibrations, multi-stall operations, etc. Accurate predictions of different types of instabilities in newer and even larger new generation turbine blades require a thorough understanding of the flow field and the fluid-structure interaction mechanisms. Wind-tunnel testing of such systems is prohibitively costly, but numerical simulations can provide a suitable platform to resolve the overall dynamics and offer valuable physical insight. However, the accuracy of a numerical study depends on the fidelity of the mathematical models used to represent the physical problem. The accuracy of the analysis will demand accurate mathematical solvers resolving the dynamics of the structure, unsteady flow field and the related fluid-structure interactions, and accurate modeling of the wind-field and relevant system parameters based on observations. Moreover, the presence of measurement noise can also affect the accuracy. Unfortunately, these conditions can rarely be met in practice. Modeling the wind-field during the operating conditions of a wind turbine is not trivial and is often corrupted by noise. Many modern wind farms are constructed as off-shore facilities and subjected to extreme conditions where the records of load data are limited. The associated flowfield can also be inherently random as a result of turbulence modeling. This highlights the importance of projecting the problem in a stochastic framework where the uncertainties enter through inadequate knowledge of the physical model and the inherent randomness of the loading. A probabilistic analysis enables (a) quantification of the propagation of system uncertainties into the response and (b) immediate identification of the most likely states of the system. Even though probabilistic treatment of uncertainties has been receiving research attention for the last 25 years or so by the structural dynamics community, application of this technique to computational fluid dynamics (CFD) and fluid structure interaction (FSI) problems is a recent trend, the reason being resolving the unsteady fluid dynamics and FSI behavior accurately by solving the Navier-Stokes equations within a deterministic framework itself involves complexities and requires large computational resources. However, with the recent advent of fast and cheap computers, the interest in approaching these problems in a probabilistic framework has gained momentum. In this chapter, a review of recent probabilistic attempts in aeroelasticity and some possible future trends are discussed. © 2010 by Nova Science Publishers, Inc. All rights reserved.