In general relativity, gravity is a distortion of spacetime due to mass. Its effects travel (if that's the right word) at the speed of light. In the SM all 3 other known interactions are mediated by exchange of particles.
A major unsolved problem is the belief that there should be a way to unify these, perhaps by some kind of quantum gravity, and perhaps by means of a gravitational exchange particle, hypothetically dubbed the "graviton".
In a region of extreme mass, a distortion in spacetime has an intuitive meaning, so it's easy to understand how a black hole can exist (gravity doesn't "leave" anywhere so it never has to pass through the EH). But in a theory where gravity is a result of exchange particles, how can that exchange occur, once an event horizon exists? Would a discrete gravitational boson have to have no self-interaction, conveying a kind of gravitational immunity, or would event horizons have an exclusion carved out ("nothing can leave except gravitons")?
Theories of quantum and exchange-particle gravity have been developed for decades, so this issue must have arisen and been considered in them.
How do current proposals for quantum gravity and "gravity based on an exchange particle" get around this apparent problem, or prevent it arising?