It's honestly just baffling to me that people find it surprising that consistent, low doses of poison may cause cancer.
That's not to say this doesn't cause cancer. Whether or not it does is irrelevant to the point that plant biology is not human biology — hell, other mammals' biology isn't human biology (see: cats and chocolate, grapes, macadamia nuts, and onions) — and making inferences based upon this hypothesis is not particularly likely to result in correct assessments.
Further, "the dose makes the poison". Alcohol is a poison. Oxygen is a poison in high enough doses. Is weed killer a poison? In some arbitrarily high dose, certainly. In some arbitrarily small dose, no. In the dose you're likely to encounter on the surface of produce? Possibly, perhaps even probably. But again, making inferences based on the fact that it kills weeds is faulty thinking. By that logic, one would be forgiven for avoiding onions at all cost; after all, they're cat poison.
> "the ambivalent quality of pharmakon is more than purely a matter of ‘wrong drug, wrong dose, wrong route of administration, wrong patient’. Drugs, as is the case with anti-retroviral therapy, have the capacity to be beneficial and detrimental to the same person at the same time."