Scientists have taught a robot how to hunt and destroy prey in a chilling new experiment. The test comes as experts warm AI could wipe out a tenth of the global population in five years. The ability to identify and zone in on a specific target will be crucial for any useful robotic technology like driverless cars, the researchers at the University of Zurich in Switzerland believe. And despite the chilling prospect of allowing a robot to mark up a target, they believe their research will prove more useful than deadly. Scientists have taught a robot how to hunt and destroy prey in a chilling new experiment.
Early terrestrial animals dragged themselves through mud and sand to take their first'steps' on land 360 million years ago. Now, scientists believe the powerful tails they used as fish were more important than previously thought. They came to this conclusion by studying African mudskipper fish and building a robot modelled on the animal. Scientists believe the powerful tails early terrestrial mammals used as fish were more important than previously thought. Scientists believe ancient creatures used the powerful tails they relied on as fish to crawl across the mud and sand as they made the move onto land, some 360 million years ago.
Some scientists are hard at work making a "kill switch" to overpower a too-strong AI and protect us, if needed. Others are specifically teaching robots how to hunt prey, also to help us. Researchers at the University of Zurich's Institute of Neuroinformatics are teaching a small, truck-shaped robot to see, track, and hunt its prey (another small, truck-shaped robot). The predator robot uses an advanced "silicon retina" to see instead of a traditional camera. This "silicon retina," which is modeled after animals' eyes, uses pixels to smoothly detect changes in real time instead of slowly processing frame-by-frame images.
Google's autonomous cars may look cute, like a yuppie cross between a Little Tikes Cozy Coupe and a sheet of flypaper, but to make it in the real world they're going to have to act like calculating predators. At least, that's what a handful of scientists at the Institute of Neuroinformatics at the University of Zurich in Switzerland believe. They recently taught a robot to act like a predator and hunt its prey--which was a human-controlled robot--using a specialized camera and software that allowed the robot to essentially teach itself how to find its mark. The end goal of the work is arguably more beneficial to humanity than creating a future robot bloodsport, however. The researchers aim to design software that would allow a robot to assess its environment and find a target in real time and space.
The applications of these lessons for the predator robot are a lot less terrifying than thinking robots are about to start hunting the human race. It's about creating software that could potentially allow a robots to both take a look at their environments and then discern a target in real time. For instance, as Tobi Delbruck, professor at the Institute of Neuroinformatics explained, "one could imagine future luggage or shopping carts that follow you." This allows the software to transcend the labels of "predator and prey" to reach levels of "parent and child," but the fundamental operating basics remain. The predator robot's hardware is actually modeled directly after members of the animal kingdom, as the robot uses a special "silicon retina" that mimics the human eye.