AI spots smuggled sea creatures with 92% accuracy

Illegal wildlife trade in the ocean is worth billions of dollars yearly, but items like shark fins and seahorses are hard to spot in luggage. Researchers at Macquarie University in Australia built an AI algorithm that can identify smuggled sea creatures with 92% accuracy.

They adapted airport X-ray CT scanners and trained the AI to spot shark fins, seahorses, and sea cucumbers. The team scanned 298 samples, including items wrapped in foil or hidden in toys. The AI got 95% accuracy for shark fins, 96% for seahorses, and 86% for sea cucumbers.

The false alarm rate is 13%. While not perfect, it could help customs fight smuggling. But 3D CT scanners are expensive and not available at all airports, so the system is just a backup, not a replacement for human checks.