This is of course nonsense. The path-integral formulation is a (mathematically less well defined) but physically equivalent (whenever one can make sense of it and calculate things with it) way to describe standard quantum theory of Heisenberg, Born, Dirac et al (1926). Often it's a nice tool to express things more elegantly than in the way with Hilbert-space objects (vectors, Statistical Operators, observables, etc.), e.g., the quantization of gauge theories and other functional methods in QFT, some aspects of scattering theory (asymptotic limits etc), and also the (semi-)classical approximation.
However, no matter in which way you express quantum theory mathematically, in quantum theory a particle is not at several places at once but there is a probability distribution for its position depending on the state it is in. The many-worlds "interpretation" is nice to think about in science-fiction story like ways but it doesn't solve anything to the problems some people still have with accepting quantum theory as the way nature works on the most fundamental level because it partially contradicts our "common sense", which however is "trained" by everyday experience with macroscopic objects (i.e., many-body systems with a huge number of microscopic constituents and electromagnetic fields), which behave according to classical physics, because that's the case for the rough "coarse grained" description which is sufficient to understand their behavior. All this can be explained with appropriate approximations for the dynamics of collective (macroscopic) observables of quantum theory, no matter whether you use the path-integral formalism or not.
