1) Black heterosexual women are more likely to have sex with black heterosexual men, who are in turn more likely to have HIV (according to the same data). The black male population has fewer cases than the females, but this might be reconciled by noting that male-to-female transmission is more likely than the reverse.
2) Perhaps the bigger risk factor for heterosexual women is not being black, but being poor. TFA does not include a breakdown by socioeconomics. Or perhaps, the risk factor for women is being part of a community with a high rate of male imprisonment.
Related to the above, it seems that male-to-female transmission during sex is more likely then the reverse in high-income countries like the US, but this does not hold in low-income countries. So that raises another question of why. See http://www.aidsmap.com/Estimated-risk-per-exposure/page/1324...
There are a whole bunch of possibilities, and I don't claim to know which are most true, but presenting a simple graph that says black women get HIV a lot seems a little...not helpful?
Edited to add: I dug up a study from three years ago by the very same CDC which specifically says it's poverty more than race that matters. So why are they publishing charts that highlight race and ignore poverty? See here: http://www.cdc.gov/nchhstp/newsroom/povertyandhivpressreleas...