[Theoretical physicist Juan] Maldacena's idea thrilled physicists because it offered a way to put the popular but still unproven theory of strings on solid footing — and because it solved apparent inconsistencies between quantum physics and Einstein's theory of gravity. It provided physicists with a mathematical Rosetta stone, a 'duality', that allowed them to translate back and forth between the two languages, and solve problems in one model that seemed intractable in the other and vice versa (see 'Collaborative physics: String theory finds a bench mate'). But although the validity of Maldacena's ideas has pretty much been taken for granted ever since, a rigorous proof has been elusive.
In two papers posted on the arXiv repository, Yoshifumi Hyakutake of Ibaraki University in Japan and his colleagues now provide, if not an actual proof, at least compelling evidence that Maldacena’s conjecture is true.
Well, hold on just a second here. We go on to learn that the "evidence" is just a mathematical model. And don't get me wrong, a mathematical model is an important thing in physics; it may be an important step toward String Theory making some bona-fide falsifiable predictions. But I'm reminded of something Sean Carroll often says: mathematics only tells us the consequences of axioms; it can't tell us which of those axioms actually correspond to reality. That's why physics is an empirical endeavor – if you just "solve" your way to the secrets of the universe, there'd be no need to build multi-billion-dollar particle colliders.
The fact that this "evidence" is really nothing more than a mathematical conjecture is really driven home by this rather huge caveat nonchalantly tacked into the tail-end of the article:
Neither of the model universes explored by the Japanese team resembles our own, Maldacena notes. The cosmos with a black hole has ten dimensions, with eight of them forming an eight-dimensional sphere. The lower-dimensional, gravity-free one has but a single dimension, and its menagerie of quantum particles resembles a group of idealized springs, or harmonic oscillators, attached to one another.They're cool ideas, but until they produce something testable, that's all they are. But at least we still have this: