Race is non-existent

The answer to the first question is, “Yes, depending on which type of skillset we are measuring. For g, yes.”
The answer to the second question is “No” if you are talking about the actual “genes for intelligence” but “Yes” if you are talking about offspring regressing to a lower mean because of other genes carried in the gene pool of highly intelligent blacks. All outliers for various genetically-determined traits tend to have offspring that regress to the mean for the population represented by the outliers. I do not think this regression to the mean is a controversial notion.