People are well aware of the glamorous fields of physics: cosmology and particle physics. These are the disciplines that concern the frontiers of the very large and the very small respectively. I don't care so much about those. The frontier that I'm most interested in is the most abstract one: complexity. Physics has traditionally tackled problems that were either simple or could be made simple. Progress was made after the 1980s with the rise of solid state physics and associated attitudes towards emergence, but now? With the advent of statistical learning, advances in nonlinear dynamics and so on? Physicists are starting to tackle some insanely complex systems. Not to mention of course that computers are getting more powerful with time as per Moore's law, so simulations are really coming into their own as useful scientific approaches. Imagine the computational physicist of the 2050s, imagine the tools that she might have at hand to solve problems like the physics of life, or perhaps the phase diagrams of extremely heterogenous materials, or so on...
I can't help but be extremely excited! Here's to hoping that humanity makes it that far. :)