Science

The Same A.I. That Tags Your Facebook Pics Can Also Stop Poachers: Video

Ivory smugglers and poachers have practically free rein of the 12,000 square miles of the Serengeti. Only 150 park rangers are tasked with watching over an area roughly the size of Belgium, making it relatively easy for illegal hunters to slip in and out undetected. But with the help of thousands of artificial intelligence-powered cameras, there’s a chance to slow down the slaughter of animals using the same facial and object recognition tech that spots your friend’s faces on Facebook.

Ecologists have already planted a vast number of camera traps to unobtrusively observe wildlife. Last week, a group of four environmental and tech entities announced a joint effort to alert authorities when poachers enter wildlife reserves using a similar array of motion-sensing cameras. The project could save up to 25,000 animals yearly, according to estimates. The plan is to deploy the cameras and their companion tech in 100 reserves over the course of 2019.

These aren’t just normal cameras. Environmental non-profit RESOLVE and chipmaker Intel teamed up to create TrailGuard A.I. cameras, that will act as stand-in sentinel when park rangers aren’t around. When a TrailGuard spots what it thinks may be a poacher, it will alert whatever rangers are nearby. The anti-poaching effort was funded by Leonardo DiCaprio Foundation and National Geographic Society.

The Intel chip that will power the TrailGuard A.I. cameras.

RESOLVE

”By pairing AI technology with human decision-makers, we can solve some of our greatest challenges, including illegal poaching of endangered animals,” said Intel’s Head of A.I. for Social Good Anna Bethke in a statement. “[This] technology enables the camera to capture suspected poacher images and alerts park rangers, who will ultimately decide the most appropriate response.”

The sensors will all run on Intel’s Movidius Myriad 2 VPU, a chip that will actively look for people or vehicles when a camera is triggered. Images will only be sent to rangers when either of those two elements are captured. This will notify authorities in real-time and eliminate the need for them to sort through hundreds of empty photos or pictures of animals.

TrailGuard A.I. is one of the many efforts to leverage modern technology to aid conservation efforts. Chinese computer scientists have begun feeding A.I. Google Maps satellite images to detect looting of ancient tombs. While research led by the University of Washington has used to genetic testing to identify ivory that originated from smuggling rings.

An example of the cameras detecting poaching activities in the wild.

RESOLVE

Both of these efforts and TrailGuard A.I. are examples of how readily available data can be used to protect the world’s marvels. They also evidence of how A.I. even fully replace humans, but augment our potential.

Constant monitoring of vulnerable areas are tasks best left to A.I.’s endless data processing abilities. While final decisions and swift action are best left to human judgement. The combination has the potential to safeguard vulnerable, yet valuable places.

Related Tags