You're technically correct -- the best kind of correct. The first thing I thought about the headline as I clicked it was "that's a rather irresponsible thing to

You're technically correct -- the best kind of correct. The first thing I thought about the headline as I clicked it was "that's a rather irresponsible thing to say" because it's generally impossible to prove that a given thing is impossible without taking assumptions as fact. In this case the assumption was that the only kinds of cancer in question were the ones we already know about, which is certainly not true. That said, the probability of a correlation, let alone causation, between cell phones and cancer is effectively zero, and for now that has to be good enough.