The idea that there are racial differences in IQ, even in first world countries, is not a "misunderstanding". It is very well established. Likewise, the idea that intelligence is partly genetic is also very well established. Also, the idea that race differences are partially genetic, is not a "misunderstanding". It is a verifiable fact, and no amount of saliva is going to change that.
If you want, we can go in-depth with this topic, but I would ask that I get to make a new thread for it, and that I get time to prepare my opening post and write some key replies beforehand.
This is the interesting part. We don't have that technology yet, but we will have it very soon. We can already read the human genome, but the problem is that we don't know what most of the genes do. There are already some genes that have been linked to traits such as intelligence, but the exact mechanism of how they work is still a mystery.
Already, it is almost possible to do IVF selection. It's like fertility treatments, where fertilized eggs are implanted into a woman. If we could understand the genome better, we could read the genes of the fertilized eggs to see if they possess the kind of qualities that parents would want and then implant those that do.
I haven't taken a detailed look at CRISPR, but further down the line, it should be possible to just plant whatever genes one desires into the human genome. Imagine the possibilities. We could genetically engineer geniuses that are way beyond what this world has ever seen. It is still a mystery as to how many genes in total are responsible for human intelligence, but as a ballpark figure, 150 might be close. Imagine, for example, if we could create a person who has the most beneficial variants of all of those 150 genes? Even the greatest geniuses throughout history have had perhaps 50 positive variants, and we could create people who have all 150.
There are ethical considerations to this kind of thing, and the idea that intelligence is partly genetic is a bit of a taboo. But the Chinese don't care about either of those things, so we should see interesting results in the (near?) future.