Watch scientists 3D-print a human eye (What the Future)

[MUSIC] What you're looking at could be the technology blind people have been waiting for. [MUSIC] Welcome to what the future Today we've got a first in human bionics, a concept vehicle aimed at disrupting modern air travel, and robotic handiwork like we've never seen. First up from the University of Minnesota, this is time lapse video of engineers building the world's first 3D printed bionic eyeball. Now, what we're seeing is the circuits being built on a glass hemisphere that simulates the shape of an eye. It's essentially an array of receptors that turn light into electricity. This whole thing took just about an hour. Scientists say this is still early stage technology, so don't expect to see this anytime soon. But, it's a huge step toward one day helping the visually impaired. Now scientists say the next big step is finding a way to print on a soft material that could be implanted into a real eye. Next up, the TF-2 concept vehicle. The company behind it, Terrafugia, thinks it could transform on demand travel Here's the idea. This road vehicle picks you up and takes you and your bags in a detachable pod. You travel to what [UNKNOWN] calls a vertiport near-by. A vertical take-off vehicle meets you there, and carries your pod to a vertiport near your destination. Again, this is just a concept right now. But consider this, Terrafugia has the backing of its parent company Geely, which also owns Volvo. And Terrafugia is already working on a flying car. Last month it showed off the latest version of its transition. It's a combination hybrid electric road vehicle and propeller driven aircraft. Terrafugia says the first production model of the transitions will be available next year. [MUSIC] We've never seen dexterity on a robotic hand like this. It's thanks to a special artificial intelligence from OpenAI called DACTYL, most of the robots you've seen are programmed to complete tasks, like opening doors and turning valves. So they can't adapt when something goes wrong. But DACTYL actually learned to manipulate this block, here's how. Engineers show Dactyl hundreds of simulations. Then they ask Dactyl to perform a task, like putting a specific side of the block facing up. This wasn't the first time OpenAI had put Dactyl through these simulations, but this time they started messing with it. They challenged Dactyl by changing up conditions, like the size of the cube And even gravity, so [UNKNOWN] could learn to deal with the unexpected. [UNKNOWN] studied about 100 years worth of simulations to get where it was in this video. So that means robots still have a long way to go before matching the human ability to learn things relatively quickly. It's question time. Alper asked us on Facebook, how far the TF2 can fly? Now the answer is a little complicated because Terrafugia is considering two motor configurations. But engineers are predicting a range of anywhere from 315- 555 km at max payload depending on which design is chosen. [SOUND] That's gonna do it for the show, I'm Andy [UNKNOWN], thanks for watching. [MUSIC] [SOUND]