Perhaps they're referring to salmon mislabeling, a common form of food fraud. E.g. https://blogs.scientificamerican.com/food-matters/to-every-scam-there-is-a-se

Perhaps they're referring to salmon mislabeling, a common form of food fraud. E.g. https://blogs.scientificamerican.com/food-matters/to-every-scam-there-is-a-season-report-shows-salmon-fraud-prevalent-in-winter/ https://blogs.scientificamerican.com/food-matters/to-every-s... > Overall, they found close to half of the samples were mislabeled (43 percent), and rates of mislabeling were especially higher in restaurants vs. grocery stores (67 percent, vs. 20 percent). The most common form of mislabeling was farmed Atlantic salmon being represented as” wild” salmon. These trends persisted when comparing all of their salmon data (466 samples) spanning over four years. Without knowing how much salmon is consumed at restaurants via grocery stores, it's hard to validate that "most" wild salmon is actually farmed.