This may be true in the U.S., but worldwide it’s almost certainly not.
There is a very good correlation between percentage of population that is lactose intolerant and the age of onset: the larger the percentage the earlier the onset.
That makes sense. Lactose intolerance is universal in mammals because all mammals who use lactose in their milk are designed to lose the ability to make the lactase enzyme at about the age of weaning. Those human populations who have that version of the gene also tend to start becoming lactose intolerant at about the age of weaning.
It’s those populations with the greatest percentage of the mutant gene that never sends out the stop signal that have so-called adult-onset lactose intolerance.
Even though the mutation is a simple one to a single gene on chromosome 2, researchers have found some 43 variations to that gene, presumably indicating that the mutation took place independently over and over again and became prevalent through natural selection. (The mutant gene is dominant.)
Why do those populations have adult-onset LI? Again presumably, they have mixed gene sets that delay the stop signal. I’m not familiar with any solid research explaining it, though.
It’s this adult-onset pattern that’s weird, but because it is so overwhelmingly the case with the white European-descended populations in the U.S. that lactose intolerance was actually called adult-onset LI until researchers in the 70s started doing studies with every population on earth and found out the real truth. Now it’s simply called primary LI.
It’s equally a myth that people with lactose intolerance can’t handle any milk. Many studies have shown that most people, and especially children, can handle a glass of milk with no symptoms.