On a typical mission, Kimberly Warner would go to a restaurant. After browsing the menu, she always picked a seafood dish. She made casual conversation with her dinner companions and took mental notes of her meal. When the waiter wasn’t looking, she would snatch a marble-sized, sauce-soaked sample and slip it into her purse. Mission accomplished.

As far as covert operations go, seafood sleuthing isn’t the most glamorous work—but that doesn’t bother her. “I feel like a real detective,” says Warner, the lead scientist in a 2012 study on mislabeled fish by Oceana, an ocean conservation group. The fruits of Warner’s stealth proved disheartening: Oceana reported that 33 percent of fish it tested in American restaurants and markets were impostors, products of seafood fraud.* At these volumes, such fraud, committed knowingly or not, is swindling Americans out of up to $25 billion annually.

Who stands to gain from passing off low-grade seafood as upscale entrees? To avoid an anti-dumping tariff of about 65 percent instated for low-priced imports, foreign seafood businesses use fraudulent species to throw off inspectors; one case of Asian catfish posing as grouper saved the perpetrator more than $60 million in tariffs. And once these shipments make it past customs, they can go on to fetch large amounts of money. Catfish can be sold as grouper, for example, for as much as four times its typical price.

While this puts restaurants at risk—unwittingly selling a fraudulent product is never good publicity—the harms of seafood fraud fall disproportionately on consumers. Besides spending more for a lower-value product, they may be paying for the health risks that come with unchecked fish. Federal officials at the National Oceanic and Atmospheric Administration (NOAA) have discovered FDA-banned antimicrobial agents and high levels of mercury in imported grouper. The Oceana report also found that 84 percent of white tuna sampleswere actually escolar—a fish that causes digestive problems—in disguise.

Short of building a home laboratory, there’s not much consumers can do to make sure that they’re buying the fish they’re paying for. They can try to stay sensitive to prices that are too good to be true, or they can buy seafood from a member of the Better Seafood Bureau, a trade organization that documents and reports fraud along the seafood supply chain.

They might also seek out the 10 percent of fish that isn’t imported—as it turns out, our domestic fisheries tend to follow the rules for seafood labeling. “We have some of the better managed fish and show evidence that some species are being rebuilt after getting overfished,” says Warner.

In his 18 years as a forensic scientist at NOAA, Trey Knotts has seen more than 2,000 catfish filets in disguise. Recently he has been seeing more mislabeled species coming in at up to half a million pounds per shipment. “There are individual companies importing millions of pounds of mislabeled fish,” he says.

Meanwhile, less than 1 percent of imported seafood is inspected for mislabeling. This is not for lack of trying. Knotts says NOAA has fewer than 100 agents who do inspections. “If you consider the coastline of the U.S., it’s a massive amount of territory to be covered by 90 agents,” he says. (Thanks to diminished federal funding, this number is growing smaller still.)

So what’s being done about our leaky seafood system? Warner’s study exposed the magnitude of the problem, and now the Obama administration is taking steps to combat fraud. One of the most ambitious proposals this year will be for federal agencies, such as NOAA and the FDA, to work together more closely on seafood-fraud investigations. And with a new set of tools on the way—including Grouperchek, a tested device soon to be available for commercial use later this year—the government and the industry appear closer than ever to fighting seafood mislabeling.

Researchers unanimously blame foreign fisheries for seafood fraud. Ninety percent of our fish is imported from countries with loose aquaculture laws, such as Thailand, Indonesia, Canada, China, Ecuador, and Vietnam. Some seafood from these countries may come mislabeled from unregulated fish farms. The most notorious offenders are Thailand and Vietnam, a country with polluted fish farms along the Mekong River teeming with Asian catfish ripe for export.

With so much fraud beginning overseas, U.S. regulators are turning their attention to foreign imports. Last December, a presidential task force made its first recommendations for how the government should fight seafood fraud. It advocated for more thorough international collaboration on labeling, and described an urgent need for better seafood tracking starting this year. The fish industry currently suffers from very little oversight and a non-transparent supply chain, says Warner. “What we ask for is electronic traceability that follows fish [through the chain],” she adds.

Seafood substitution violates the Magnuson-Stevens Act, a 1976 law that regulates commercial fisheries and controls overfished species such as speckled hind, a critically endangered fish that Oceana frequently found labeled as grouper. The law also covers the overfished Atlantic halibut, which is sometimes sold as Pacific halibut, its similar-looking but eco-friendlier cousin. Continued unregulated fishing can lead to the depletion of fish populations and the collapse of entire marine ecosystems.

This is not just an American headache. “There are over 100 studies where we have found seafood-fraud incidents. Studies have been performed in 29 countries, in all continents, except Antarctica. It’s a global problem,” Warner says.

One of the repeat offenders in the Oceana report was grouper—or, more accurately, the 66 varied species of fish that the FDA legallyrecognizes as falling under the loose term "grouper." Though the U.S. already grants the seafood industry this leeway, some fisheries will still try to pass off yet other species as grouper, including Asian catfish, perch, bream, weakfish, and king mackerel. “There are upwards of 400 species that could be mistaken for grouper,” according to Knotts.

And it’s not because groupers are just so spectacularly humdrum that they blend in with the other fish. Seafood fraud is most prevalent with filleted seafood, products where heads, tails, and other identifying features are removed from each fish. For decades, DNA sequencing has been the only tool that can make sense of anonymous fish parts.

And DNA sequencing is precisely what keeps Knotts’ lab at NOAA busy every day. By sequencing a specific gene, such as Cytochrome B, he can accurately map a seafood sample onto a digital tree of known species. If a fish labeled grouper is actually Asian catfish, its Cytochrome B sequence will be far removed from the grouper branches. But DNA sequencing takes at least 12 hours, which is too much time for most wholesale buying decisions. The seafood industry needs a faster way to differentiate between species, right on the dock.

One option that might hold promise is Grouperchek, a portable, commercial seafood identifier—for one species, at least. Theunassuming device was designed by Robert Ulrich and John Paul, scientists at the University of South Florida, and it looks like an external hard drive plugged into a laptop.

The small machine can test four samples at a time. The simple process is different from traditional DNA sequencing—it can’t identify the species of a specimen. It can, however, from a small sample, verify whether a fish is grouper. And it can do so in 45 minutes.

This portable technology can be used at all points of the supply chain, which is good news for everyone, from distributors to restaurants to consumers. “[Grouperchek] has widespread potential,” says Knotts, the NOAA scientist. “It can fill in some of the inspection gaps and can up the amount of seafood you can actually inspect.” This means that regulators can check species at auction markets, while suppliers and processors can verify their products, adds Louis Petrin, founder of Arrow Scientific, an Australian food-testing supply company. “Even pet food companies that pay a premium for certain [fish] species can [now] simply check the large frozen blocks they receive,” he says.

Grouperchek’s creators hope to make it commercially available by July this year, and are working on devices that would also verify red snapper, shrimp, tuna, and halibut. Even with these improvements, however, Grouperchek will have its limits. In order for NOAA to press charges against seafood fraudsters, who face penalties of fines andjail time, the government still needs forensic scientists like Knotts to provide evidence that will hold up in court. There, the agency will have to identify which fish was substituted. So DNA sequencing, that slow but definitive process for identifying fish, isn’t going away any time soon.

But court cases rarely happen, as the odds are in favor of the wrongdoers. NOAA law-enforcement agents must stay on top of the latest seafood schemes in order to make the decision of whether to bring a batch of grouper to forensics; an ill-founded suspicion means lab time wasted and refunds issued to the importer. “This is a game of cops and robbers,” says Knotts. Once a scheme is figured out, the offenders move onto the next ruse. “It’s a shifting game for people who had gotten in trouble for their grouper-catfish scheme. It’s a constant challenge for us to try to react from the scientific end in a way that can be applied in a legal-enforcement context,” he adds.

The constant contact with mislabeled fish has made Knotts wary about seafood fraud even when he’s away from the lab. Very rarely does he bring his own dinner in for testing, but his work has made him more suspicious, he says. “Every time you see grouper on the menu, you go ‘hmm.’”