Do you mean that it was proven that observation affects the other end immediately, or is it only just a theory? If so, how it was proven, without it also being useful for sending information?
It is a theory in the sense that it is epistemic. Bob's quantum state changes depending on Alice's measurement, essentially collapsing the probability distribution to the conditional probability distribution. As science it cannot be proven, but the empirical results match the model. The paper I cited defines precise protocols to define terms like "steerable states." This is standard quantum theory, so the no-communication theorem still applies.
Is the quantum state also ontic, and in what sense? Those are open questions of interpretation.
Is the quantum state also ontic, and in what sense? Those are open questions of interpretation.