Maybe I should've asked you the question I just asked on another post instead: as someone interested in minimizing x-risk, who should I support for President? Or better yet, who has a good compilation of candidates' records on x-risk-related issues, so I can make my own decision?

Glad to hear!

I don't think there is much publicly available on this topic besides Koblentz's work (also check out his 2003 article in International Security). The "strategy of conflict" as it pertains to bioweapons is something we thought about, but we don't discuss it much in our paper. Some thoughts:

Historically bioweapons research has focused on diseases that are not transmissible person to person like Tularemia, Anthrax, Q Fever, and Botulism. If you dump a bunch of anthrax spores from an airplane over a city, you would kill a lot of people (I recall seeing a study where they estimated that if you dumped anthrax over a large city you would kill ~200,000 people) even though it's not transmissible person to person.

Japanese Unit 731 used Plague, which is transmissible person to person, but used it well behind enemy lines on enemy cities to limit collateral damage. This made this sort of weapon more of a poor man's strategic bombing campaign that could wreak havoc on civilian populations by dumping swarms of plague infested fleas on enemy territory. There is a lot of uncertainty on the actual numbers, but Unit 731 may have killed more civilians in China through these sorts of attacks than the US killed Japanese civilians in Japan from the nuclear bombs and firebombings (the exact numbers are hard to know in large part because it's hard to attribute civilian deaths directly to the released bioweapons because the weapon is meant to present itself as a natural epidemic, as we mention in the paper). There is evidence that the Japanese did sustain some collateral damage from their plague attacks against China during the war: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1200679/.

Starting a pandemic transmissible Person to Person in the enemy's country is probably a bad idea because it could turn into a global pandemic and there is significant collateral damage to your own country. Though, it should be noted that using other WMDs like nuclear weapons also has significant, and possibly similar scale, collateral damage. One of the dangers of future development in biotechnology is that it could make biological weapons more attractive as a weapon.

Determining strategic behavior and implications of using weapons that impose significant collateral damage to the country using them and to non-combatant countries is actually an active research topic of mine.