The technique involved is called “gain of function.” Scientists take a known virus and inject it into, say, a ferret. Then they wait for the ferret to get sick, harvest the virus from the ferret’s body — which may now have mutated — and inject it into a second ferret. Then they do that again with a third ferret, and a fourth, and so on, with the virus continually mutating, until finally it mutates into a form that can infect another ferret naturally, through proximity instead of injection.
(Viruses that are transmissible among ferrets tend to be transmissible among humans, Newsweek notes.) They’re not trying to engineer a deadly pathogen via this “animal passage” process so much as they’re trying to predict what a deadly pathogen might look like if it emerged naturally among animals in the wild.
The U.S. program to fund research like this is called, not coincidentally, “PREDICT.” In theory this will lead to antiviral drugs derived from the lab-mutated viruses so that we have a weapon that’s ready if/when a deadly novel virus emerges in nature, but so far it seems like there haven’t been any practical results from it.
The risk from this technique is obvious. By hurrying along the mutation process, scientists are breeding pathogens right there in the lab that may prove lethal *and* infectious. If there’s a lab accident and someone gets infected before knocking off work for the day…