As mentioned here previously, the Chinese have been experimenting on genetically modifying human embryos. This would not only “fix” genetic flaws in that individual, but would also be passed down to any descendants he or she might have. Well, they’ve finally published their results:

Scientists Make History By Genetically Modifying Human Embryos

And, well, how’d those initial experiments turn out?

The resulting embryos were a total mess. After applying CRISPR to 86 embryos, 54 of the surviving 71 embryos were genetically tested. Only 28 spliced successfully, and only a fraction of them contained the replacement genetic material. What’s more, the researchers found a surprising number of unintended mutations. The scientists decided to stop the experiment at this stage.

Yeeeep.

Since the experiments were carried out on human embryos, the experiment was considered unethical – and thus un-publish-worthy – by the likes of Nature and Science. In the article linked above, several quotes are included from NYU School of Medicine bioethicist Arthur L. Caplan. He says some appropriately nasty things, but reaches a conclusion that I find dubious. Yes, there are many ethically dubious issues here… how the embryos were obtained, whether proper approvals were given b the parents, so on. But the bioethecist concludes that since the study failed on ethical grounds, “I don’t think there’s much scientifically here of interest.”

Hmmmmph.

In science, you learn from failure. The attempt to modify the embryos appears to have been a complete bust, but history is littered with science experiments that failed but led to later successes. And I find the idea that if the ethics were bad, there’s nothing of scientific interest. If someone invents a warp drive or a teleporter, and uses drunken or crazy bums as unwitting test subjects… yes, the ethics is bad, but buy howdy do I want to get a look at the test results. Similarly: if this embryo study had been a complete 100% success, the fact that the ethics was bad would not negate the scientific value.