Semaglutide sure does seem to cause thyroid cancer in rats, but the mechanism isn't known, and nobody has found evidence that this generalizes to humans.

In Jan 2027, I'll resolve this YES or NO based on my assessment of the available evidence. "A nontrivial rate" here means "at least one extra thyroid cancer case per 10,000 patient-years".

I won't try very hard to rule this out, but: semaglutide simply making people die less of heart disease/diabetes and instead live longer so that their thyroid cells have more time to go haywire doesn't count. The semaglutide has to be ~directly messing up those thyroid cells.