I showed this article to a friend of mine who did a PhD in physics at Cambridge. Here's what he had to say:
The entanglement arguments against relativity are fairly silly. From that article, looks like Scientific American is going the way of the New Scientist. :( Absolutely no mention of Everett and his MWI (editor's note: multiple worlds interpretation) in the article I notice. Even though that's the most obviously straightforward way of keeping special relativity, locality, quantum mechanics and accounting for the Aspect experiment. Basically Bell's theorem that is supposed to prove non-locality makes a subtle assumption called "counter-factual definiteness", which implies among other things that experiments have definite outcomes. Obviously it's violated by MWI. Really it's only Bohmian mechanics that's non-local and arguably not consistent with SR. Though Bohm adherents often claim that Copenhagen must be non-local by Bell's theorem, they misunderstand the fact that the wavefunction in CI is really an epistemic device and wavefunction collapse isn't thought to be a genuine physical occurrence except by a very small minority. CI when viewed in this way doesn't satisfy counter-factual definiteness either.
The article doesn't say what's new about the authors' ideas. You'd expect it to mention the transactional interpretation of quantum mechanics - it isn't news that the Aspect experiment can be explained if measurements affect things that happened earlier.
Even if these conclusions are correct, they don't mean that "Einstein was wrong".
Physical laws are approximate descriptions of reality that are based on certain explicit or implicit assumptions and any conclusions that you derive from them are valid only as long as they hold. If Einsteins assumptions are proven to be invalid in certain regimes, it does not mean that the theory of relativity is wrong, just incomplete (as we already knew it was).
For example:
Newton's laws are not wrong, they just assume that your velocity is much smaller than the speed of light. If your velocity becomes close to light, Newton's laws are no longer valid and you must use Einsteins relativity. On the other hand, Relativity assumes that Quantum effects are negligible. If your system is in the quantum regime, these assumption no longer hold and you must use different laws.
In the context of the article, it is pretty clear that the author isn't asking whether Einstein was wrong per-se about special relativity; he is asserting specifically that Einstein was partially mistaken in his criticism of quantum mechanics and his dismissal of nonlocality (in the EPR paper). On the flip side, he credits Einstein for seeing through the smoke-screen put up by Bohr about separating physics from our conceptions of reality, that is, breaking the link between physics and metaphysics.
For at least the last 70 years, physicists haven't taken the nonlocality predicted by quantum mechanics seriously enough. Even supporters of QM considered the prediction of nonlocality a liability, and tried to dismiss it as an artifact in various ways. Since much of QM couldn't be experimentally proven at the time, there was a fear that suggesting our assumptions of strict locality of action to be false would lead everyone to dismiss QM out of hand.
So, when we were finally able to empirically prove most of QM's predictions in the 80s, no one noticed that also meant nonlocality was a fact of our world, and that we should probably take the time to square that with our assumptions of spacetime geometry handed down by SR.
It now appears that to reconcile nonlocality with the things we know about spacetime geometry, we're going to have to challenge at least one other long-held assumption. One proposal suggests that not only is space nonlocal, but so is time. Another hypothesis suggests that the state of our universe at any given time is too infinitely complex to be reduced to even an infinite set of truth propositions.
The entanglement arguments against relativity are fairly silly. From that article, looks like Scientific American is going the way of the New Scientist. :( Absolutely no mention of Everett and his MWI (editor's note: multiple worlds interpretation) in the article I notice. Even though that's the most obviously straightforward way of keeping special relativity, locality, quantum mechanics and accounting for the Aspect experiment. Basically Bell's theorem that is supposed to prove non-locality makes a subtle assumption called "counter-factual definiteness", which implies among other things that experiments have definite outcomes. Obviously it's violated by MWI. Really it's only Bohmian mechanics that's non-local and arguably not consistent with SR. Though Bohm adherents often claim that Copenhagen must be non-local by Bell's theorem, they misunderstand the fact that the wavefunction in CI is really an epistemic device and wavefunction collapse isn't thought to be a genuine physical occurrence except by a very small minority. CI when viewed in this way doesn't satisfy counter-factual definiteness either.