An introduction to QBism with an application to the locality of quantum mechanics. C. A. Fuchs, N. D. Mermin and R. Schack in Am. J. Phys. 82:749 (2014).  What the paper says!?

This is an introduction to QBism by one of its most ardent promoteur. It describes in general terms how it solves various problems, including the claim that «“Quantum nonlocality” is an artifact of inappropriate interpretations of quantum mechanics.»

It is highlighted how in the Bayesian viewpoint (in particular from the Dutch-book approach to deriving probabilities), probability and thus quantum mechanics is intrinsically a single-user theory:

It makes no sense to impose Dutch-book coher ence on a combination of probability assignments made by more than one agent. In this sense probability is a “single user theory”

Some of the most satisfying clarifications are neatly formulated:

The notorious “collapse of the wave-function” is nothing but the updating of an agent’s state assignment on the basis of her experience.

It also touches upon the problem of reality:

An agent dependent reality is constrained by the fact that different agents can communicate their experience to each other, lim ited only by the extent that personal experience can be expressed in ordinary language. Bob’s verbal representation of his own experience can enter Alice’s, and vice-versa. In this way a common body of reality can be constructed, limited only by the inability of language to represent the full flavor—the “qualia”—of personal experience.

quantum mechanics itself does not deal directly with the objective world; it deals with the experiences of that objective world that belong to whatever particular agent is making use of the quantum theory.

It also tackles the central difficulty that probability-1 does not refer to concrete, external reality, and say that this is a major conceptual difficulty of the theory:

That probability-1 (or probability-0) judgments are still judgments, like any other probability assignments, may be the hardest principle of QBism for physicists to accept.

Later they link this to EPR (and say this is where lies their "mistake", not like Bohr's complain on "not perturbing a system"). This point of view is fully enforced by the Wigner's friend paradox, which indeed shows (in its modern, post-Wigner formulation) that probability-1 for a given observer can be "undone" and yield interferences precluding the perfect knowledge as synonym of reality.

This makes QBism intrinsically local, since it all concerns one single person, who is timelike by essence:

QBist interpretation, cannot assign correlations, spooky or other wise, to space-like separated events, since they cannot be experienced by any single agent.

The observer is special in QBism:

In QBism, the only phenomenon accessible to Alice that she does not model with quantum mechanics is her own direct internal awareness of her own private experience.

Here there is an interesting comment on other agents:

The personal internal awareness of agents other than Alice of their own private experience is, by its very nature, inac cessible to Alice, and therefore not something she can apply quantum mechanics to.

Links to Copenhagen interpretation:

Quantum Bayesianism—agrees with Bohr that the primitive concept of experience is fundamental to an under standing of science.

Unlike Copenhagen, QBism explicitly takes the “subjective” or “judgmental” or “personalist” view of probability,5–9 which, though common among contemporary statisticians and econo mists, is still rare among physicists: probabilities are assigned to an event by an agent10 and are particular to that agent. The agent’s probability assignments express her own personal degrees of belief about the event.

This means that reality differs from one agent to another. This is not as strange as it may sound.