How do you fight an insect that inflicts billions of dollars of crop damage every year, an invader that has spread to every inhabited continent and has a talent for evolving resistance to every new chemical designed to poison it? Maybe—just maybe—by releasing more of the same insect.

This summer thousands of moths will be taking to the sky over a cabbage patch in upstate New York. Bred in a laboratory, the male bugs carry a gene designed to kill progeny they sire with wild female moths feasting on the cabbage. This genetic time bomb is an emerging tool for keeping agricultural pest populations in check, made possible by ever-improving methods for editing DNA.

Over the coming months, before winter kills the moths, researchers at Cornell University will monitor how well their modified pets compete against wild cousins for mates. If all goes well, this small-scale experiment could be a step toward commercializing the genetically engineered diamondback moths, made by a British company called Oxitec.

“We might imagine male moths being released on a given field once, twice, three times a week, to mate with wild females and prevent the pest from ever becoming a problem,” says Neil Morrison, who leads the moth project at Oxitec, which has also been tweaking other six-legged nuisances like olive flies.

For Morrison the laboratory lepidopterans are the newest weapon in a long-running sexual battle against insect pests. Half a century ago scientists began sterilizing the screwworm to drive down births of this flesh-eating cattle parasite. Lacking genetic tools to render the critter infertile, they blasted it with radiation to deliberately inflict random genetic damage (an idea inspired by human mutations blamed on the atomic bomb). Released by the millions, sterilized screwworm flies dominated the outdoors mating scene, eradicating the pest from the United States and saving countless cows from agony.

“This technique has been very successful, but it’s like using a sledgehammer to sterilize the insect,” says Anthony Shelton, a Cornell entomologist who tried to sterilize moths with radiation but found that the required dose rendered them too weak to fly. “Genetic engineering allows us to be much more precise.”

Oxitec has already tested its deadly DNA in mosquitoes set free in the Cayman Islands, Malaysia, Brazil, and Panama.* "I've always been very relaxed about this approach, and they’ve done a fantastic job of illustrating that these insects are safe," says Peter Atkinson, a geneticist at the University of California, Riverside, who was not involved in the moth project. "The moths aren't passing the gene they carry on to a next generation out in the wild because there is no next generation."

The gene harbored by the altered moths seems to have little effect on the males being released. But daughters that inherit this gene in the wild should die. That’s because the gene produces a toxin in females, unless it is switched off by feeding the moths regular doses of tetracycline. In the lab, where researchers provide the antibiotic, multigenerational colonies of male and female moths can be raised. In a cabbage field, where insects can hardly pop over to a pharmacy for the drug, newborn females never make it to adulthood.

“From a biological standpoint, it’s a reasonable technology,” says David Riley, a vegetable entomologist at the University of Georgia who was not involved in the Cornell project. “Sterility is not something that persists very well in nature, though, so you’d have to release quite a number of moths to control populations in the field.”

Two years ago Shelton pitted Oxitec’s moths against unaltered males in outdoor cages. The modified bugs proved to be just as long-lived, and almost as good at finding mates. Encouraged by the results, Shelton sees the open-air release scheduled to take place in New York in the coming weeks as the next step in testing whether the insects could be a viable alternative to chemical insecticides.

“I’ve had growers write to me in China, in India, in Southeast Asia, with photographs of destroyed fields, asking for help,” Shelton says. “There’s a desire to try a different kind of technology.”

The diamondback moth was once a relatively insignificant pest, only one of many that munched on cabbage, broccoli, brussels sprouts and other Brassica vegetables. Its rise to infamy began in the 1940s, after World War II spurred the development of synthetic pesticides. Most insects were no match for DDT or other new chemicals—but the diamondback revealed a talent for survival. It has evolved resistance to every new generation of poison invented since.

“There’s not an insecticide made by man that we know of that can stand up to the diamondback moth,” Riley says. Buoyed by the death of its competition, the diamondback has spread across the world.

Many farmers are likely to welcome the new tool, if Shelton’s experiments go well. “I think using the insects is an awesome idea,” says New York grower Anthony Piedmonte. He already releases wasps, ladybugs and other insects into his fields to prey on the moths that eat his cabbage. But some New York growers have a different perspective. The Northeast Organic Farmers Association (NOFA) denounced the U.S. Department of Agriculture (USDA) for issuing the permit that allows the moth experiment to take place, concerned that such methods could undermine their ability to sell organic produce if dead larvae wind up in their crops.

In making its decision, the USDA said it had considered possible risks involving the gene itself—and a fluorescent protein that helps researchers track the moths—and had checked for toxicity as well as the possibility of allergic reactions. The agency also said it had found little risk of environmental impact. Though the moths are not expected to travel far from the New York release site, the USDA required the experimenters to install traps at the edges of the cabbage patch and farther away. “We did put in some conditions around monitoring the site to check for unexpected dispersal,” says Alan Pearson, a branch chief in the USDA’s Animal and Plant Health Inspection Service.

But the traps are not designed to catch every moth, and that worries Andy Fellenz, an organic farmer and NOFA staff member. Diamondback moths are not an enormous problem on his farm; his small amounts of cabbage do not attract significant numbers of the pests, and because he does not use chemical pesticides, resistance is not an issue. Regulations that govern organic farming prohibit from him from using genetically engineered insects, while potentially penalizing him if larvae from such insects used by his neighbors turn up in his crops, he says.

“I’m not dead set against genetic engineering, but if there’s contamination with genetically engineered materials beyond a certain percentage, my crop can’t be sold as organic,” Fellenz explains. “Whose responsibility is it to compensate organic farmers for this?”

For Shelton, the decision to move forward with the moths means weighing the risks—which he says the data from lab tests and outdoor cage tests so far suggest are minimal—against the known risks of the old technologies. “Insecticides have been a very valuable tool for pest management,” Shelton says. “But with the pressure to decrease the amount of insecticide use, we are going to need some other tools.”

*Editor's Note (8/25/17): This sentence was edited after posting. The original incorrectly stated that Oxitec released mosquitos in Florida, not Panama.