One issue is that adaptive optics on telescopes have not yet become so sophisticated that they work well in the shorter visible wavelengths, much less the UV. To correct the image at a given wavelength you need to adjust the surface of your deformable mirrior to a small fraction of that wavelength... this is easier to do with longer wavelengths.... extending this to shorter wavelengths is something that is being worked on... and is only a matter of time and money...

In the IR the largest ground based telescopes can currently outperform the Hubble- however, in shorter wavelengths, Hubble still has the advantage.

I believe that in the next decade or two space telescopes in the near IR and Visible will be viewed as inefficient- space observatories will most likely concentrate on the shorter and longer wavelengths- meanwhile the near IR and visible will be handled well by current and planned ground based observatories- planning is already underway for 30 meter scopes and larger- and, this is a plug for my current subject of interest: most of the largest existing scopes and all future large ones will use Laser guide stars...