At some point in the future it might become possible to design viruses that target human genotypes with great selectivity, and at that point this will become a risk. But we are a long, long way from that future, and even then most people's circle of concern includes significant genetic variation, so very few people could release a virus designed to kill, say, all Slavic people, without killing at least a few of their own close friends and family members.

Historically, biological warfare research has focused on pathogens like anthrax that have very low human-to-human transmission risk, for two reasons. One is that, if your germ-warfare bomb starts an epidemic, it barely matters which sides of the battle lines it falls on; if it thins your own ranks less than those of the enemy, it's only because your troops (and their families dying back home) are in better condition or have better sanitary measures in place. The other is that, if you can't stop the deaths when the enemy surrenders, the germ-warfare bomb doesn't give the enemy any incentive to surrender.

Evil things do happen fairly regularly on a grand scale. But doing things on a grand scale has historically required a centralized government to organize them. That's still true for a little while longer.