By that same logic, I can walk into a hospital, go to someone on life support who can't currently "survive on their own", and unplug their systems, thus "withdrawing the support they require to live". If they die, is it really any different from killing them any other way?
To make the analogy even more accurate, let's say you are in a coma and will come out of it in 9 months. You can't survive without support now, but we know for a fact that you will come out of the coma after the 9 months have passed and will then be able to do so. Is it fine for me to shut down your life support half way through?
in the specific case of pregnancy, the support is given at significant personal cost and risk, which is not captured in either of your analogies.
However, I do acknowledge that there are exceptional circumstances, such as the birth posing a health risk to the mother, both mother and child being likely to die if delivery is attempted, etc. That would be the case of "not being able to afford it" in my analogy, or the significant personal risk you mention (at least I assume that's what you mean, as normal pregnancy is nowhere near a significant risk, with the current maternal death rate being 1 in 10,000). In those cases, I'd be willing to agree with you.
this is not necessarily as hard of a moral conundrum as it seems. some people draw a distinction between deliberately killing a being that can survive on its own and withdrawing support that a being requires to live. this is somewhat analogous to the trolley problem: is it worse to kill than to allow to die?