He says numbers, not primes, so it's trivially correct. Though that doesn't directly imply the conjecture being true for sufficiently large numbers, since prim

He says numbers, not primes, so it's trivially correct. Though that doesn't directly imply the conjecture being true for sufficiently large numbers, since primes become increasingly sparse as one moves up the number line... it seems more like an appeal to intuition than a real argument, to me.