But extrapolation can fail.

Eg, people often say any amount of radiation is bad, but there’s evidence that isn’t true. If you’re going to make a similar claim about alcohol, you should justify it.

“Persistent consumption above some threshold” is a radically different claim than “any amount”; and you should quantify that in both respects.