The $abc$-conjecture implies that the equation $a+b=c$ has only finitely many primitive solutions in the multiplicative semigroup generated by any particular finite set of primes.
I would appreciate any information about the status of this a priori weaker statement, and citations to the literature if any exist.