Can Ai Detect Smuggled Sea Cucumbers?

Scientists hoping to extremity nan illicit trade of marine wildlife person a caller instrumentality to spot seahorses, shark fins and oversea cucumbers hidden successful luggage. The tool, which uses artificial intelligence, could beryllium deployed astatine airports to bolster wildlife enforcement efforts, nan researchers say.

Wildlife trafficking is simply a awesome industry: astir nan world, immoderate $20 cardinal successful works and animal products are sold illegally each year, according to nan International Criminal Police Organization (INTERPOL). That includes marine species—such arsenic sea cucumbers, seahorses and shark fins, which are illegally harvested and sold for imaginable medicinal uses aliases arsenic food. Many of these wildlife products walk done airports and often spell undetected, biology advocates say.

In nan new study, which was published connected Sunday successful nan diary Frontiers successful Ocean Sustainability, researchers trained an AI algorithm connected hundreds of three-dimensional x-ray images—the benignant of imaging already utilized successful airports—of 68 dried shark fin, seahorse and oversea cucumber samples. Across hundreds of images, nan algorithm correctly identified these samples 92 percent of nan time, pinch a mendacious affirmative complaint of astir 13 percent.

On supporting subject journalism

If you're enjoying this article, see supporting our award-winning publicity by subscribing. By purchasing a subscription you are helping to guarantee nan early of impactful stories astir nan discoveries and ideas shaping our world today.

Examples of discovery scans of marine wildlife, including seahorse (top left), shark fins (top right), oversea cucumber (bottom left), and shark fin (bottom right). The bottommost 2 scans were inserted into images of bags.

"Marine wildlife trafficking: usage of AI algorithms for nan autodetection of shark fins, seahorses and oversea cucumbers" by Pirotta et al., successful Frontiers successful Ocean Sustainability, June 7, 2026

“Never successful my profession would I deliberation AI would beryllium specified an important portion of my research,” says Vanessa Pirotta, lead writer of nan study and a wildlife intelligence astatine Macquarie University successful Australia. X-ray imaging “enables america to look successful and astir luggage and message items—this intends we tin usage this tech to understand really group whitethorn alteration their trafficking efforts complete time,” she says.

The algorithm, she adds, is aimed astatine “building our discovery capacity” and is not intended to switch “manual quality inspection” aliases “biosecurity dogs.”

From here, Pirotta hopes to deploy a type of this exertion successful airports.

“The adjacent measurement is moving toward making these algorithms progressive astatine beforehand lines astir nan world—this is apt to thief capable those gaps regarding occurrence and support enforcement efforts,” she says.