00:00Neuralink exists to bridge the gap between humans and computers and we've
00:05developed a robotic surgery process to deliver this groundbreaking technology
00:09to as many people as quickly as possible.
00:12Here at Neuralink, the robot is critical to our surgical process. After the
00:17surgeon has created the opening in the skull and lined up the brain to the
00:22robot, the robot's role is to grab threads off of the implant and insert
00:27them into the brain. The reason we need a surgical robot is these thin flexible
00:32electrodes are almost impossible for a surgeon to place like one by one.
00:37Imagine placing like a single strand of hair into jello and then having to do
00:42that hundreds of times. In the early days this was like very much a research
00:47project and we didn't really have much focus on like usability or like mass
00:51production. We went through many many iteration cycles improving actuators, the
00:56optics, the needle design and eventually we got to a point where we were like
01:01internally convinced that this had some like major legs to it and we could go to
01:06our first human surgery. After our first successful surgery it was a like huge
01:11weight off our shoulders. It was like all right it worked for one person how can we
01:15bring it to thousands or millions more people in the future.
01:18We are constantly learning new things. We always look at what could be better
01:22about this robot. Nobody actually goes to engineering school trying to make five
01:27percent improvements to a product. We all studied this so that we could make the
01:3010x jumps and when you start thinking about how do you make a product 10x
01:34better your solution space opens up entirely. This next generation of the
01:39robot is faster, it's easier to use and it's more scalable. This is really exciting
01:44for us because it allows us to reach more people into more regions of the brain
01:47across the world. The components here are the NC just where the needle is, there's
01:54the implant that's held on by the implant holder, there's the implant arm and one
01:58interesting thing about the implant arm is this is a previous generation implant arm
02:02and so we shrunk this whole thing down to this which is huge achievement. Then
02:07there's other components like the sensors we've got cameras over here, here, there's in
02:13total eight cameras. There's a OCT system which stands for optical coherence
02:17tomography. This sensor allows us to see through the surface so this robot has to
02:22move in five axes to reach all the regions of the brain. If you imagine the
02:26the cranectomy has to be here but it for a different region of the brain the
02:30cranectomy might be here so the robot has to adapt to that. It's important to
02:34reach more regions of the brain because every problem has a specific location in
02:38your brain. If you want to solve motor problems it's a different region. If you
02:42want to solve vision problems it's a different region. We're never going to
02:45stop iterating. We're never going to stop making progress. We want to solve
02:49every problem that originates in the brain. Long term we're looking at ways to
02:53increase the depth and accessibility of the robot and then also continuing to
02:57streamline the surgical process. This is an exciting paradigm shift in how
03:02humans approach medicine. We're taking engineering principles and applying it to
03:07a field that historically has not had them applied. How do you make a specific
03:12step more repeatable, more reliable, faster? When you think about the future of the
03:17surgery you have to think about the whole door-to-door experience. The robot is
03:21one part of that experience. We have to think from first principles how to solve
03:25each step and how to combine them together to provide an amazing experience
03:29that compares to LASIK surgery. We know that this approach works. Now we get to the
03:34exciting part. How do we build the best product off of that base architecture? The
03:38robot itself is not in its final form. We want to go faster. We want the threads
03:44to go deeper and help hundreds of thousands more people. I think we're
03:47gonna look back and our grandchildren are gonna think it was crazy that somebody
03:51could get in a motorcycle accident and never walk again. That's what we're
03:55working towards here. And it's all going to be because this robot was able to
03:59insert threads that no human could imagine doing on their own.
