At airports across the globe, X-ray CT scanners are about to get smarter—and thousands of illegally trafficked seahorses, shark fins, and sea cucumbers won't stand a chance. Scientists at Macquarie University have developed an artificial intelligence algorithm that detects these smuggled marine creatures in airport baggage with 92% accuracy, a breakthrough that could help protect some of the ocean's most vulnerable species from one of the least-known but most damaging wildlife crimes.
When wildlife trafficking comes to mind, many picture rhino horns or baby orangutans destined for black-market pet trade. But the illegal trade in marine life—worth billions annually—deserves the same urgency. Shark fins destined for soup, seahorses dried for traditional medicine, and sea cucumbers stripped from the ocean floor for profit move across borders with alarming ease, hidden in luggage and parcels that slip past existing detection methods. The damage is staggering: these trafficking routes drain fragile ocean ecosystems, and animals smuggled alive risk becoming invasive species in foreign waters. Yet quantifying the harm has been nearly impossible, simply because we lack the tools to catch the crime in progress.
Dr. Vanessa Pirotta and her team at Macquarie University saw an opportunity. Airports already use X-ray CT scanners to detect explosives and biosecurity threats—3D imaging machines that create detailed digital pictures of baggage contents. The researchers realized these same scanners could be repurposed to identify trafficking. They trained a neural network algorithm on 298 scans derived from 20 sea cucumber samples, 30 seahorse samples, and 18 shark fin samples, many sourced from actual wildlife trafficking seizures. To mimic real-world smuggling tactics, they scanned items wrapped in tin foil, concealed in children's toys, and hidden among legitimate luggage contents.
The results are striking. The algorithm achieved 95% accuracy in detecting shark fins, 96% for seahorses, and 86% for sea cucumbers, with a false positive rate of just 13%. These numbers suggest the system could automatically flag suspicious shipments that currently evade detection, disrupting trafficking networks and building legal cases against smugglers. "The trade of wildlife is cruel and unethical," Pirotta said. "For many, this may be the first people have heard of illegal trafficking of marine wildlife."
But Pirotta is careful not to oversell the technology. The algorithm works for three species; thousands of others are trafficked globally. The system complements rather than replaces traditional methods like sniffer dogs and human inspectors. And not every airport can afford 3D CT scanners; many rely on older 2D imaging technology. Manually reviewing every false positive remains necessary work. "AI is not a silver bullet for detection, nor a replacement for human and sniffer dog detection," Pirotta acknowledged.
Still, in the battle against marine wildlife trafficking, a 92% detection rate represents real progress. It's a tool—imperfect but powerful—that could intercept shipments that would otherwise vanish into the supply chain. As ocean ecosystems face mounting pressures from overfishing, climate change, and pollution, cutting off the illegal trade in endangered marine species matters deeply. The algorithm won't solve wildlife trafficking alone, but it offers something the ocean desperately needs: a fighting chance.