I don't think the goal is a moral panic. There are three takeaways from this:
1 - The AI has figured out a way of determining race that could be detrimental to medical studies where race needs to be anonymises. I won't get into this as I have a loose understanding of the role race and bias plays in medicine
2 - The AI has identified a novel distinguishing factor between races that humans so far have been unaware of. This would be significant.
3 - The AI is cheating via an unforeseen mechanism