Let's think of cancers as biological errors in the growth instructions for cells. ITT, people are pointing out that ionizing radiation can cause genetic perturbations to cause cancers. Let think of that as a bit-flip in the genetic code. Then commenters posit that "if non-ionizing radiation isn't strong enough to cause genetic perturbations, it can't be responsible for sources of cancer". Biological systems respond to environmental stressors by trying to resist them and regenerate. Each regeneration genetic code is copied. Each copy has some probability of an error in it. Let's call this genetic bit rot. So it seems plausible that if non-ionizing radiation induces elevated genetic regeneration, it could cause an increase in genetic degeneracy such as cancers etc.

Also stressors are does dependent, e.g. despite earth's atmosphere being 78% by volume Nitrogen: the marginal effect on your health of marginally more nitrogen isn't stable/linear. I wouldn't recommend you increase it so much, so to some degree we need to discount opinions that "we exist in a EM/RF soup ergo an iota more isn't hazardous).

Let's also not confuse the ensemble and individual components. As RF electronics, protocols improveve, so does the minimum necessary energy. That is "the minimum necessary energy for 5G radios" <= "the minimum necessary energy 4G" isn't the comparison we should be making. What we should be assessing is going from effectsOf(2G,3G,4G) < Safelevels to effectsOf(2G,3G,4G,*5G*) < Safelevels. e.g. adding another layer of EM energy

I have no idea if the health assertions in this article are true or not but rather than unjustifiably upgrade an opinion of "I don't see how that could be the case" to "that *can't* be the case" I'll stick with the skeptics response of "I don't know".