Over the next decade, we can expect that some imaging of extrasolar planets will be possible with the high-resolution LET and present 8 m class telescopes. But it will be limited by sensitivity and contrast ratio to self-luminous planets of the nearest young stars. When the Giant Magellan Telescope (GMT) comes on-line, it will have because of much larger light grasp and sharper PSF, the potential for imaging many new planets as well as, for the first time, imaging planets of known msini. It will also be capable of starting atmospheric studies through spectrophotometry. The full angular resolution of the GMT (that of D=24 m filled aperture) will be exploited with coronagraphy and nulling interferometry. The new coronagraphic technique of phase apodization being pioneered at the MMT will enable very high contrast at angular separations ≥3λ/D. To reach the highest contrast levels, the AO system is being designed not to minimize wavefront error, but to shape the corrected wavefront so as to cancel speckles in the search region. Interferometric measurements of complex amplitude in the focal plane make this possible, regardless of whether the speckles originate from errors in diffraction or phase. New control algorithms are being developed to minimize the decorrelation time as well as the intensity of residual speckles, so that they average out to the smoothest possible background halo. In this way, detections at 1.65 μm at the 5σ level of planets at 10 -8 contrast at 50 mas separation should be possible. The low background AO system of the GMT, made with its deformable secondary, will allow also high contrast imaging with high sensitivity at 5 μm, down to 100 mas separation.