Biological weapons can be divided into two categories: infectious and noninfectious. Infectious agents infect a host, reproduce, and then spread to another one or more new host. Noninfectious agents do not.
Noninfectious biological agents are similar to chemical weapons. You can stop both of them with an airtight suit and a breathing apparatus. The cost of protective clothing and the associated training is a rounding error in the budget of a modern military. Meanwhile, noninfectious biological agents are expensive to produce and maintain. Operationally, the military implications of noninfectious biological agents are basically identical to the military applications of chemical weapons. They both revolve around about keeping small particles from touching you, especially the inside of your lungs.
In Why Don’t We Use Chemical Weapons Anymore? Bret Devereaux explains why chemical weapons do not make sense for any war involving strong states. Basically, chemical weapons are not used against strong states because they cost more to deploy than to defend against. Chemical weapons are not used by strong states because a strong state can accomplish area denial and destruction of the enemy better via nuclear and conventional weapons. Chemical weapons are used only by weak states against weak enemies. Chemical weapons can be considered obsolete for the purposes of modern warfare. Noninfectious biological agents can be considered obsolete too because noninfectious biological agents are operationally similar to chemical weapons.
It is not unlikely that future noninfectious biological weapons will used for assassinations, but I don't see how they would be a game-changer. The future of biological warfare revolves around the use of infectious agents against civilian populations.
We can divide the use of biological agents into tactical applications, strategic applications and terrorist applications. Tactical weapons are used against military targets. Strategic weapons are used by a state against civilian targets. When weapons are used by a non-state actor against a civilian target, we call it terrorism. It is true that particular weapons are optimized for tactical, strategic or terrorist use. However, weapons can be used for applications beyond their intended purpose. Using a tactical weapon strategically makes it a strategic weapon. Using a strategic weapon tactically makes is a tactical weapon.
Infectious biological weapons are insignificant as a tactical weapon against a strong state. They do not differ significantly from chemical weapons and noninfectious bioweapons in this respect. Tactical bioweapons are unimportant to near-future grand strategy between strong states. We may observe limited tactical use of infectious bioweapons by weak states against other weak actors.
For a strong state, the only important use of a biological weapon is strategic. In particular, a strong state may consider using a biological weapon against the civilian populations of an enemy state. There are advantages and disadvantages to the strategic use of biological weapons.
Disadvantages of Strategic Biological Weapons
First-strike weapons exist to destroy the enemy's power to retaliate.
Weaponized pathogens make lousy first-strike weapons because they take a long time to do damage. They do not instantly destroy the enemy state's power to retaliate. This can be somewhat mitigated if a state can deploy a bioweapon covertly. This is not an issue for terrorist groups.
Second-strike weapons exist to retaliate after surviving an attack. Submarine-based nuclear missiles are the quintessential second-strike nuclear weapon because they can be fired after your nation has been annihilated by a nuclear attack.
The biggest disadvantage of a strategic infectious biological weapon is that pathogens don't respect human borders. The cheapest way to create a hard biological border is by inoculating your own population. Inoculating a nation's population against a biological weapon cannot be done in secret. If done in public, it is likely to provoke a conflict.
Nuclear weapons do respect borders, relatively-speaking. If you have reliable second-strike nukes then they are superior to biological agents as a strategic weapon.
Advantages of Strategic Biological Weapons
However, constructing nuclear weapons is expensive and requires access to uranium. It is hard to hide a nuclear weapons program. Attempting to enrich uranium paints a target on your capitol. As bioengineering technology gets cheaper and more powerful, it may make sense for a non-nuclear state to develop infectious biological weapons as last-ditch strategic deterrent.
When human beings develop infectious biological weapons, these weapons will be tiny and nonradioactive. As biotechnology democratizes, more and more states will have the potential to create an infectious biological weapon. Attribution will become increasingly difficult. Future bioweapons are thus similar to present-day cyberweapons.
Biotechnology is following a trajectory similar to Moore's law. Eventually, a small illegal organization will have the potential to create a pandemic disease. The optimal disease would be airborne, with a long infectious incubation period followed by death. A biological agent is easy to sneak through customs. The attack surface of potential viruses is too broad to vaccine everyone preemptively. The only defense is rapid response.
The exponential advance of biotechnology works in the favor of attackers and defenders. It makes testing and vaccination easier. But there are many viruses (not to mention bacteria) we do not have vaccines for. We cannot count on the rapid development of an effective cure of vaccination against an engineered pandemic bioweapon. The only reliable defense is rapid response, contact tracing and (if possible) testing.
Though terrorist groups are most willing to deploy a biological weapon, the difficulty of attribution means that state actors may do the same thing covertly. Pandemic response infrastructure calibrated against deliberate biological attack should therefore be put in place several years before it becomes possible for a state actor to engineer a pandemic bioweapon. There are many ways for a pandemic bioweapon to get loose.
When such a system is created it will change the calculus of bioweapon borders. Infectious biological weapons do become constrained by borders, but stop being effective against strong states. This drops infectious bioweapons into the into the purview of weak states alongside chemical weapons and noninfectious bioweapons.
In a perfect world, an active bioweapon defense system would be equally effective against state actors and non-state actors. However, a state actor launching a covert bioattack against the civilian population of an enemy state is unlikely to restrict warfare to the domain of biology. Such an attack would likely combine all covert domains of war including, especially, cyberspace. Addressing the future of covert biological warfare is therefore predicated on an analysis of future cyberwarfare.
What about coordination?
It may be easy to get the major world powers (China, USA, Russia) to denounce the use of infectious biological weapons. But arms control treaties only work when weapons are big, visible and expensive. Infectious pathogens are tiny and invisible. Genetic engineering is getting cheaper fast.
You could use the pre-existing differences between your population and the targets. In nation states, you could make the virus decide by presence of absense of certain DNA markers. Or you could make the virus lethal/harmless for people who drink milk / eat pork / consume lots of corn syrup...
It will not be perfect, but in a war, certain losses are often considered acceptable.
Of these ideas, genetic markers seem like the best way to go. They're discrete and hard for a target nation to emulate. You don't even need 100% coverage to keep your population (mostly) safe. The most important thing is to keep R0<1 in your nation and R0>1 for the enemy.
The covert release of an infection agent in multiple locations simultaneously along with a long and symptomless period of incubation and transmission would render any tracing and testing strategy ineffective.
Future? That's been the go-to biowar tactic for 3000+ years.
I mean in contrast to noninfectious toxins like ricin of the Bush-era WMD program.
I think the reference is to the well-accepted cases of intentional (along with likely unintentional) introduction of smallpox to Native American populations, and at least some amount of bubonic-plague-era disposal of bodies near to enemy camps with the intent to spread it.
I don't think it's been used much in the modern era - the populations worth attacking are generally advanced enough to require more direct and selective means. I expect this will continue for quite some time - it'll pick up in terrorism a bit, but organized forces with significant civilian populations to protect will realize that such weapons are too likely to escape the enemy.
There's probably a middle ground of "semi-infectious" weapons, which can be controlled easily, but not without doing a bunch of damage before it's understood. Or that infect, for instance, only a specific strain or growing condition of wheat.
Infectious bioweapons can be importantly different from chemical weapons because a single soldier screwing up his protection may infect his whole unit or even more if r is high and incubation period is long.