He isn't really a string theorist, more a general relativist.
There's a difference between being fundamentally untestable, like the proposition 'God exists', and being untestable due to technological issues. We can't test plenty of things in QCD yet but it's still science. String theory does say things about phenomena we can test. For example, string theory says specific things about the behaviour of gravity, specifically that gravity should be carried by massless particles and obey the Einstein field equations at large distances. If string theory had spit out any equations other than the Einstein field equations it would be immediately falsified.
There's no 'laws' against multiple universes.
As pointed out, your statement is unscientific because it's baseless. And your $$10^{500}$$ number comes from string theory, which you don't consider science either. And the $$10^{500}$$ number isn't how many universes string theory says exist. I've already explained it in this thread, it's the number of possible solutions to particular equations space-time structures in string theory have to obey, just like cubic equations have 3 solutions. There's infinitely many different metrics which obey the Einstein field equations but that doesn't mean GR predicts infinitely many universes. Furthermore, while there's $$10^{500}$$ possible vacuum states which solve the relevant equations under a naive counting that doesn't mean they are all different physical systems. Many of them are just rewritten versions of one another under dualities, in a more complex but conceptually similar notion to changing coordinates in GR doesn't change the physics.
You have absolutely no grounds to say "2 universes is not allowed, there should be more" or to use the $$10^{500}$$ number from string theory in that manner.
Furthermore, if you talk about 'alternative universes' in the context of 3 dimensional slices (ie D3-branes in string theory) of some larger space then if those can have signatures. Branes near to one another can exchange gravitational effects so you can see a gravitational signature of material not in our universe. Or the branes can even collide, causing a very obvious signature in the sky. In fact people have even constructed cosmological models where our 3 dimensional 'slice' is formed in the decay products of some things much larger and higher dimensional colliding in a 'larger universe'. String vacuum dynamics and brane cosmology are considered as interesting areas of research for cosmologists because they naturally incorporate components otherwise shoe horned into cosmological models like inflaton fields, reheating, flatness and dark energy. Even in models where it is obvious it isn't going to be how the real world works they provide useful insight into otherwise difficult to examine processes.
String theory has done that in other areas. Gluon-gluon processes are some of the most dominant processes in the LHC right now but they are difficult to do using Feynman diagrams. Within string theory someone (Witten) developed a method to resum pure massless gauge field processes, which led to a method called MHV. The string theory origins then fell away and left a method no one would have otherwise constructed. It's a completely different way of computing gluon dynamics from Feynman diagrams. Or an even bigger contribution to our understanding of gauge theory is gravity/gauge duality. Motivated by string theory examples it's now seen as something independent of string theory, something which string theory provides examples of but which can be explored even if someone killed string theory tomorrow. It's told us a lot about how gauge theories work in strongly coupled regimes like those at the centre of neutron stars, ie quark-gluon plasmas.
As always Reiku, get your facts right before opening your mouth.
