I assume we’re talking about quantum teleportation. What this really involves is transferring a quantum state from one particle to another – that is, changing the state of a particle that’s far away so that it matches the state of a particle that’s nearby. In the process, you change the state of the particle you started with (which is why it’s teleportation as opposed to duplication). Because two particles in the same quantum state are indistinguishable, the end result is the same as if you had somehow physically moved the particle from point A to point B.
In other words, it may be more accurate to think of it as “destroying the particle at point A while creating an identical duplicate of it at point B”, but this is essentially equivalent to teleportation.
This is all well and good for atoms, but scaling it up to macroscopic objects is basically impossible. To make the problem clear I need to outline the steps in quantum teleportation:
(1) Create entanglement between two particles that are far apart from each other. Neither of these particles is the one you wish to teleport.
(2) The sender performs an operation that breaks the entanglement between this pair, and instead entangles their half of the pair with the atom they want to teleport. In the process, the receiver’s half of the pair is also changed, but not necessarily into the state they wanted to send.
(3) The sender then measures their new entangled pair (in a particular measurement basis), and transmits this information (by classical means) to the receiver.
(4) The receiver uses this information to further adjust their particle so that it now has the state the sender was trying to send.
The problem is that if a given particle has two possible states, then a pair of particles has 4 possible states, and three particles have 8 possible states. So N particles have 2[sup]N[/sup] possible states. N is really, really large for macroscopic objects, so 2[sup]N[/sup] is ludicrously, unthinkably large. If you wanted to number each of the states of your macroscopic object, your numbers would need to have more digits than there are atoms in the known universe.
Note that for teleporting a macroscopic object, the procedure requires the sender to measure the state of a macroscopic object, transmit that state to the receiver, and then have the receiver modify the state of their macroscopic object. Even if you could somehow measure and the state of such a large system, there’d be no way to record or transmit the result of your measurement. As I said, we’re talking about more bits of data than there are atoms in the known universe – a lot more, in fact. Also note that the procedure would require you to entangle a macroscopic number of particles, and entanglement becomes increasingly fragile with the size of the system.
I think most people who work on this stuff admit it will never be possible with macroscopic objects, barring some breakthrough no one has even remotely thought of yet. Some people like to be optimistic, but I think that optimism is about on the level of me saying “Maybe someday we’ll figure out how to travel through time – but I don’t have any idea how.”