WEBVTT - Can Robots Fix Recycling? 0:00:15.356 --> 0:00:22.836 Pushkin. Imagine you're running a factory. You've got to put 0:00:22.876 --> 0:00:26.356 out a consistent, high quality product that your customers will buy, 0:00:27.076 --> 0:00:30.676 but you have no control over the raw materials that 0:00:30.756 --> 0:00:34.356 come into your factory every day. One day, half of 0:00:34.396 --> 0:00:38.396 what comes in is literal garbage. The next mixed in 0:00:38.476 --> 0:00:42.356 with your usual inputs is some random lithium ion battery. 0:00:42.796 --> 0:00:48.156 It's a fire hazard and also a surfboard weird. And 0:00:48.236 --> 0:00:50.596 you have to deal with all this stuff and still 0:00:50.716 --> 0:00:54.916 keep getting your product out the door. This is literally 0:00:54.956 --> 0:00:59.596 the way the recycling business works. Recycling plants take in 0:00:59.636 --> 0:01:03.996 a largely random, occasionally hazardous stream of stuff, a stream 0:01:04.036 --> 0:01:07.156 of stuff that changes in a pretty unpredictable way from 0:01:07.236 --> 0:01:10.476 day to day, from hour to hour, and then recycling 0:01:10.516 --> 0:01:13.716 plants have to turn that random stream of inputs into 0:01:14.156 --> 0:01:18.556 aluminum and plastic and cardboard that other companies will buy 0:01:18.876 --> 0:01:22.316 and use to make new stuff. This is why my 0:01:22.396 --> 0:01:27.076 guest today calls recycling the most demented form of manufacturing 0:01:27.116 --> 0:01:29.836 on the planet, and it's why she and her colleagues 0:01:29.876 --> 0:01:32.596 are trying to use technology to bring some order to 0:01:32.676 --> 0:01:42.276 the recycling chaos. I'm Jacob Goldstein, and this is what's 0:01:42.316 --> 0:01:44.316 your problem the show where I talk to people who 0:01:44.356 --> 0:01:48.036 are trying to make technological progress. My guest today is 0:01:48.196 --> 0:01:51.836 Rebecca who Troms. She's the co founder and CEO of 0:01:51.876 --> 0:01:55.756 a company called Glacier. Rebecca's problem is this, how do 0:01:55.796 --> 0:01:59.156 you use AI and robotics to make recycling a somewhat 0:01:59.236 --> 0:02:03.276 less demented business. If Rebecca and her colleagues are successful, 0:02:03.596 --> 0:02:06.796 they'll not only help recycling plants work better, they'll help 0:02:06.836 --> 0:02:09.516 companies figure out how to recycle more of this that 0:02:09.516 --> 0:02:11.556 they're sending out into the world in the first place. 0:02:12.716 --> 0:02:15.916 Our conversation started with Rebecca talking about the moment around 0:02:15.916 --> 0:02:18.636 five years ago when she and her co founder decided 0:02:18.636 --> 0:02:19.516 to start the company. 0:02:21.076 --> 0:02:25.196 I was, you know, even at the time, obsessed with trash, 0:02:25.356 --> 0:02:28.036 just obsessed with where does all of our stuff go? 0:02:28.196 --> 0:02:29.716 And it's one of those I call it like a 0:02:29.716 --> 0:02:32.556 Matrix moment or a red pill moment where once you 0:02:32.996 --> 0:02:35.356 realize that you've never thought about where all of your 0:02:35.356 --> 0:02:37.956 garbage goes after you put your bins on the curb, 0:02:38.316 --> 0:02:42.796 you can't unsee that, right. And so this was also 0:02:42.836 --> 0:02:44.596 around the time where there was a lot of change 0:02:44.596 --> 0:02:47.756 happening in the recycling industry. So we're rewinding to roughly 0:02:47.796 --> 0:02:51.796 twenty eighteen twenty nineteen. One cataclysmic shift for the industry 0:02:51.916 --> 0:02:55.076 is that China, who had previously been the world's largest 0:02:55.116 --> 0:02:59.636 buyer of recycled feedstock to make into new things. They 0:02:59.716 --> 0:03:03.036 basically said very rapidly, you know what we're taking in 0:03:03.116 --> 0:03:05.956 the world's recycling, but most of this is trash. People 0:03:06.036 --> 0:03:07.916 are not sorting it well enough. We're getting a ton 0:03:07.956 --> 0:03:10.556 of contamination, and we don't want to end up as 0:03:10.596 --> 0:03:14.396 the planet's landfill or incinerator. So we're going to drastically 0:03:14.436 --> 0:03:18.316 increase the bar on quality of what we're accepting. And 0:03:18.396 --> 0:03:21.356 then that caused shockwaves throughout the globe and certainly for 0:03:21.396 --> 0:03:23.956 the US, where suddenly recyclers for the first time in 0:03:23.996 --> 0:03:26.316 a long time, were like, the game is not to 0:03:26.476 --> 0:03:29.436 just crank through all this recycled material, bail it and 0:03:29.476 --> 0:03:32.676 ship it overseas. We actually need to invest a lot 0:03:32.756 --> 0:03:38.236 more in increasing the bar on quality and on purity rate. 0:03:38.436 --> 0:03:40.916 And what's even more challenging is that a lot of 0:03:40.956 --> 0:03:43.356 the backbone, historically and even to this day, is still 0:03:43.396 --> 0:03:46.836 just people standing next to conveyor belts sifting through our 0:03:46.956 --> 0:03:50.156 recycling in our trash and of course that's not only 0:03:50.316 --> 0:03:53.436 very dangerous, it's not a very well compensated job. There's 0:03:53.436 --> 0:03:55.636 a lot of hazards there. But also there's this massive 0:03:55.916 --> 0:03:57.036 sort of labor shortage. 0:03:57.156 --> 0:04:00.356 Yes, it seems like a robot friendly moment, a robot 0:04:00.436 --> 0:04:03.636 friendly environment. So like, what's your move, what's your first move? 0:04:03.796 --> 0:04:06.796 So my co founder had already at the time, he 0:04:06.836 --> 0:04:08.316 wasn't my co founder, he was just a friend of 0:04:08.356 --> 0:04:11.276 a friend. He was already pretty intent on this idea 0:04:11.316 --> 0:04:15.196 that hey, you know, automation AI, all of these technologies 0:04:15.276 --> 0:04:18.276 are so good now that for the first time, we 0:04:18.316 --> 0:04:24.276 could feasibly rapidly commercialize a purpose built industrial robot, specifically 0:04:24.316 --> 0:04:25.516 for recycling sortation. 0:04:26.196 --> 0:04:26.516 Huh. 0:04:26.516 --> 0:04:29.556 And it's not going to cost us massive amounts of capital, 0:04:29.636 --> 0:04:31.676 and we can actually do it in a matter of 0:04:31.796 --> 0:04:33.036 like a couple of years. 0:04:33.156 --> 0:04:35.916 So this idea that oh, now is the moment? Is 0:04:35.916 --> 0:04:39.276 it computer vision? Like, what is the underlying technology that 0:04:39.396 --> 0:04:42.276 made five years ago or whatever? The moment when oh, 0:04:42.276 --> 0:04:43.796 we can do this in a way that hasn't been 0:04:43.796 --> 0:04:44.356 done before. 0:04:45.116 --> 0:04:47.556 Yeah, Honestly, it's the confluence of a lot of things. 0:04:47.636 --> 0:04:49.676 So I'll break it down into sort of the computer 0:04:49.796 --> 0:04:53.196 vision piece and then also the hardware or the robotics piece. 0:04:53.996 --> 0:04:57.476 When you think about where industrial automation has come from 0:04:57.556 --> 0:05:00.836 even to this day, a lot of those technologies are 0:05:00.876 --> 0:05:06.396 operating in really well defined, truly repetitive roat environments. So 0:05:06.436 --> 0:05:08.916 think about a robot at a warehouse, and it's literally 0:05:08.996 --> 0:05:13.356 just palatizing identical boxes over and over again. And so 0:05:13.436 --> 0:05:16.516 to harken back to this idea of recycling plants as 0:05:16.556 --> 0:05:20.476 being this extremely volatile manufacturing environment, even if you have 0:05:20.716 --> 0:05:25.596 automation that's just sorting let's say aluminum cans, right, You're 0:05:25.636 --> 0:05:28.596 talking about aluminum cans that your computer vision needs to 0:05:28.636 --> 0:05:33.076 detect in infinite varieties, not just thinking about the wide 0:05:33.156 --> 0:05:35.236 variety of cans that are on the market in all 0:05:35.276 --> 0:05:38.316 of their colors and designs, but also the fact that 0:05:38.316 --> 0:05:41.356 they show up not as pristine cans but crinkled in 0:05:41.436 --> 0:05:45.156 various ways, stuffed into bags. Like, there's so much heterogeneity 0:05:45.196 --> 0:05:50.516 that even just identifying that item on conveyor belt that, 0:05:50.556 --> 0:05:53.196 by the way, has dozens of other types of items 0:05:53.196 --> 0:05:57.076 on it is already a massive challenge that only recently 0:05:57.156 --> 0:05:59.356 has been something that we can adapt to in a 0:05:59.436 --> 0:06:02.436 cost effective way. And then you can layer onto that 0:06:02.476 --> 0:06:04.956 the fact that now you're not only seeing those items 0:06:04.996 --> 0:06:08.236 with this computer vision system, but you also need to 0:06:08.236 --> 0:06:10.756 find a way to actually go and grab that material 0:06:10.876 --> 0:06:13.476 and sort it into the right location. So when we 0:06:13.516 --> 0:06:17.076 talk about advances in sort of off the shelf parts 0:06:17.116 --> 0:06:19.516 that you used to design and make your robot, or 0:06:19.556 --> 0:06:22.796 even like the gripping technology that's available, a lot of 0:06:22.836 --> 0:06:25.996 that even a decade ago, would have required an immense 0:06:26.036 --> 0:06:28.276 amount of r and D, with a much bigger team 0:06:28.516 --> 0:06:30.396 and a much higher price tag to get to the 0:06:30.396 --> 0:06:32.596 same point that we've gotten to after just a couple 0:06:32.596 --> 0:06:32.996 of years. 0:06:33.316 --> 0:06:35.276 Tell me about the first one you built. Tell me 0:06:35.316 --> 0:06:37.596 about building a prototype and putting it in the world. 0:06:37.916 --> 0:06:40.316 Man, So there are a lot of different stages to 0:06:40.356 --> 0:06:43.236 our prototyping. The first prototype, if you want to go 0:06:43.276 --> 0:06:46.156 way back, was when I hadn't even decided to start 0:06:46.196 --> 0:06:48.276 this company with my co founder yet, but I did 0:06:48.316 --> 0:06:50.996 tell him I would help him learn about the industry 0:06:51.116 --> 0:06:52.716 see if there was some sort of a business to 0:06:52.716 --> 0:06:56.276 be had, and we met in his kitchen in San 0:06:56.316 --> 0:06:59.276 Francisco where he had a little corner set up. So 0:06:59.436 --> 0:07:02.836 like very very simple. I think that it actually involved 0:07:02.916 --> 0:07:06.716 a used yogurt tub as like this rotating wheel with 0:07:06.756 --> 0:07:09.236 a piece of string tied around it, Like that's how 0:07:09.316 --> 0:07:11.916 janky we were talking, but it worked and we were like, Okay, 0:07:12.276 --> 0:07:15.036 there's something here. The first piece of equipment we actually 0:07:15.076 --> 0:07:18.516 installed into a recycling facility also tells you a lot 0:07:18.516 --> 0:07:21.556 about the constraints that these facilities are under. I was 0:07:21.676 --> 0:07:25.996 kind of thinking we had to have this super built out, sophisticated, 0:07:26.036 --> 0:07:29.436 polished thing that we've proven out to the nines in 0:07:29.476 --> 0:07:32.036 the lab, and we actually called up a number of 0:07:32.116 --> 0:07:34.956 recycling facility operators nearby and one of them was like, 0:07:35.036 --> 0:07:37.156 you know what, when you got something, just like bring 0:07:37.196 --> 0:07:40.356 it in here and try it out, because literally I 0:07:40.436 --> 0:07:43.916 remember him saying, if your robot can pick one more 0:07:43.956 --> 0:07:46.796 can than I would have gotten otherwise, Like it's already 0:07:46.796 --> 0:07:47.356 worth it to me. 0:07:47.716 --> 0:07:51.596 Just try it. Was there a rat moment, tell me more, 0:07:51.836 --> 0:07:54.076 did you see a rat? 0:07:54.076 --> 0:07:54.556 Literally? 0:07:54.796 --> 0:07:56.956 I was like, what is that not a metaphor not 0:07:56.996 --> 0:07:58.156 a metaphor a rodent? 0:07:59.276 --> 0:08:02.836 I have seen many rats. Literally, Even on that first 0:08:02.876 --> 0:08:05.156 install there was a moment where I was like literally 0:08:05.276 --> 0:08:08.716 army crawling under a conveyor belt to fasten one of 0:08:08.756 --> 0:08:11.196 the legs of the robot, and I came eye to 0:08:11.236 --> 0:08:13.276 eye with a rat, who then, of course ground a 0:08:13.276 --> 0:08:14.596 piece of food that was on the floor and then 0:08:14.636 --> 0:08:18.556 scurried away. Right, there are many friendly critters running around 0:08:18.556 --> 0:08:19.436 some of these facilities. 0:08:19.836 --> 0:08:21.916 Was there any part of you that kind of loved it? 0:08:22.156 --> 0:08:23.796 Oh, one hundred percent. 0:08:24.236 --> 0:08:26.996 So let's talk about where you are today, both on 0:08:27.036 --> 0:08:29.916 a kind of micro level, like what your robot or 0:08:29.956 --> 0:08:31.836 robots look like, and then also a little more macro 0:08:31.916 --> 0:08:34.076 of like the scope of the business. Do you have 0:08:34.156 --> 0:08:37.196 like one basic robot what to look like or you 0:08:37.196 --> 0:08:37.996 got a bunch of them? 0:08:38.276 --> 0:08:42.476 Yeah, so we have one base model of robot. We 0:08:42.556 --> 0:08:45.996 actually are already working with several dozen customers across the country. 0:08:46.036 --> 0:08:48.996 But if you imagine any sort of conveyor belt in 0:08:49.036 --> 0:08:52.436 an industrial facility, our robot think of it almost like 0:08:52.476 --> 0:08:54.636 a table, right, So it's got four legs and it 0:08:54.716 --> 0:08:57.996 kind of sits over that belt, and then the guts 0:08:57.996 --> 0:09:00.436 of the robot or the mechanisms doing the picking are 0:09:00.636 --> 0:09:03.276 kind of over the top of that conveyor system. So 0:09:03.316 --> 0:09:06.156 you've got these arms that are going back and forth. 0:09:06.796 --> 0:09:09.516 They can pick up something from the belt and then 0:09:09.556 --> 0:09:11.516 they carry it off to one of those sides of 0:09:11.516 --> 0:09:14.476 the belt. Where they actually drop it into the right location. Okay, 0:09:14.756 --> 0:09:17.036 so that's the robot. And now one other thing we 0:09:17.076 --> 0:09:19.516 haven't really talked about is this computer vision system. 0:09:19.596 --> 0:09:20.476 So that's yeah. 0:09:20.716 --> 0:09:23.916 Imagine basically like a camera with some lights to illuminate 0:09:23.956 --> 0:09:26.556 the belt sitting on this little rig that's a little 0:09:26.556 --> 0:09:29.916 bit upstream of the robot. So the material passes under 0:09:29.956 --> 0:09:33.116 the camera, the camera has a second to process or 0:09:33.076 --> 0:09:35.236 a should I say, a couple milliseconds to process what 0:09:35.316 --> 0:09:38.276 it's seeing, and then not only does it tell the 0:09:38.356 --> 0:09:40.756 robot you know, hey, there's a can coming in this 0:09:40.796 --> 0:09:43.316 location picket and then put it into this other spot, 0:09:43.676 --> 0:09:46.756 but what we're also finding is that that data as 0:09:46.796 --> 0:09:51.436 a standalone is also able to massively advance these operator's 0:09:51.476 --> 0:09:56.196 abilities to understand and optimize their facilities. So that's opened 0:09:56.276 --> 0:09:58.676 up a whole new world of use cases. 0:09:58.396 --> 0:10:01.916 Because they weren't gathering data in that kind of way before. 0:10:01.996 --> 0:10:04.796 They only sort of knew what was coming through in 0:10:04.836 --> 0:10:06.556 a very gross macro way. 0:10:07.316 --> 0:10:10.876 Right, And this gets back to you know, recycling facility operators. 0:10:10.876 --> 0:10:13.396 I have so much admiration for how they have gotten 0:10:13.476 --> 0:10:16.876 really resourceful with trying to understand their operations. But you know, 0:10:16.916 --> 0:10:18.676 to give you a sense of things. The state of 0:10:18.716 --> 0:10:21.076 the art in the industry to this day is still 0:10:21.356 --> 0:10:23.836 mostly manual audits. And when I say that, I mean 0:10:23.916 --> 0:10:26.596 imagine taking a half ton of material off your line 0:10:27.036 --> 0:10:29.956 and then literally having two to four people hands sort 0:10:30.076 --> 0:10:34.436 and categorize and wigh each item to understand what's coming through, 0:10:34.476 --> 0:10:37.036 and then assuming that that half ton is representative of 0:10:37.076 --> 0:10:39.676 like the thousands of tons coming through your facility on 0:10:39.716 --> 0:10:42.196 a yearly basis. I often tell the story of one 0:10:42.236 --> 0:10:46.076 of our early data customers, this gentleman who runs a 0:10:46.116 --> 0:10:49.596 recycling facility in California. When he met me, he was 0:10:49.596 --> 0:10:51.676 telling me that he had mounted a goat pro camera 0:10:52.396 --> 0:10:55.916 above his conveyor belt, the one that was basically supposed 0:10:55.956 --> 0:10:57.956 to be all trash leaving the facility, but he knew 0:10:57.956 --> 0:11:00.436 he was missing some good stuff, and he would spend 0:11:00.476 --> 0:11:03.076 an hour a day after work going frame by frame 0:11:03.316 --> 0:11:07.196 through some random snippet and manually tallying how many cans 0:11:07.236 --> 0:11:10.356 and bottles were on that line, and then using that 0:11:10.436 --> 0:11:12.716 to back into what he would try and change in 0:11:12.716 --> 0:11:15.156 his operation the next day, and then he would check 0:11:15.156 --> 0:11:17.716 that day to see if it changed anything. And so 0:11:17.756 --> 0:11:19.916 imagine his delight when I told him, hey, actually we 0:11:19.956 --> 0:11:22.276 can mount our own camera on there, and suddenly we'll 0:11:22.316 --> 0:11:24.676 just give you access to a dashboard. 0:11:24.556 --> 0:11:27.556 A machine will literally count everything that goes. 0:11:27.516 --> 0:11:30.556 Literally in real time. So we're seeing that, you know, 0:11:30.596 --> 0:11:32.476 the robot is this incredible foot in the door with 0:11:32.516 --> 0:11:36.476 a lot of our facility partners, but that everyone's starting 0:11:36.476 --> 0:11:39.116 to realize that, hey, actually this data can also help 0:11:39.196 --> 0:11:42.076 us understand the entire world, not just the location where 0:11:42.076 --> 0:11:43.156 the robot is sitting either. 0:11:43.636 --> 0:11:46.356 So at this point is it kind of a robot 0:11:46.396 --> 0:11:49.036 business on the front, but really you're like a computer 0:11:49.156 --> 0:11:51.716 vision data AI business. 0:11:52.116 --> 0:11:54.956 So a lot of customers come to us saying, you know, 0:11:54.956 --> 0:11:57.396 I literally had a gentleman tell me a couple months 0:11:57.436 --> 0:11:59.236 ago like I would never pay you to tell me 0:11:59.276 --> 0:12:01.436 what I already know about my trash. And I'm like, 0:12:01.516 --> 0:12:03.596 I'm not going to convince you that you don't already 0:12:03.636 --> 0:12:06.076 know everything about your trash. But you know you want 0:12:06.076 --> 0:12:08.516 a robot, Let's get you a robot. And that has 0:12:08.556 --> 0:12:10.516 sense of all the conversation where we're just sort of 0:12:10.596 --> 0:12:12.236 starting to show him this data and he's like, oh, 0:12:12.276 --> 0:12:14.516 actually I didn't realize that this was the case. The 0:12:14.556 --> 0:12:16.636 flip side is also true where someone's like, I don't 0:12:16.636 --> 0:12:18.956 know if I need a robot yet, but I'm really interested 0:12:18.956 --> 0:12:20.956 to see what I'm losing on the back end. We 0:12:21.036 --> 0:12:23.636 install that camera and then suddenly, lo and behold that 0:12:23.716 --> 0:12:25.876 data makes the case that, holy cow, I'm losing so 0:12:25.956 --> 0:12:27.716 much stuff. I don't just need one robot. I maybe 0:12:27.796 --> 0:12:30.396 need two or even three robots. Right, So it's kind 0:12:30.396 --> 0:12:33.836 of this mutually reinforcing flywheel that's been really integral to 0:12:33.876 --> 0:12:35.276 the success of our business so far. 0:12:36.076 --> 0:12:39.476 Tell me about your work with Amazon and with Colgate Palmolith. 0:12:39.676 --> 0:12:41.476 What are you doing for them with them? 0:12:41.996 --> 0:12:45.756 Yeah, So to start, maybe just I'd love to explore 0:12:45.756 --> 0:12:48.036 this idea of what is a circular economy because it's 0:12:48.036 --> 0:12:50.996 a buzzword that gets thrown out a lot, but it's 0:12:51.036 --> 0:12:54.156 really important to understand why Amazon and Coliate and their 0:12:54.196 --> 0:12:57.716 peers matter here. Right now, we are living in mostly 0:12:57.756 --> 0:13:00.076 a linear economy. In other words, someone makes a thing, 0:13:00.156 --> 0:13:02.396 we consume a thing, and then we dispose of it. Right. 0:13:02.476 --> 0:13:05.356 A circular economy tries to turn that process into a circle. 0:13:05.436 --> 0:13:07.636 So instead of throwing it out in a landfill forever 0:13:08.036 --> 0:13:10.996 or incinerating it, that material gets brought back to the 0:13:10.996 --> 0:13:13.956 front end and reused somehow to make new stuff that 0:13:13.996 --> 0:13:15.916 we can then consume. And the ideal is to make 0:13:15.956 --> 0:13:18.996 this go on forever so that we limit our resource consumption. 0:13:19.556 --> 0:13:22.316 So now that we have this growing base of recycling 0:13:22.316 --> 0:13:25.716 facilities that are you know, gathering data, that are getting 0:13:25.716 --> 0:13:28.796 a better understanding of what's coming into their facilities, what's 0:13:28.836 --> 0:13:31.556 actually being bailed and sent out to the end markets, 0:13:32.076 --> 0:13:35.756 We're working with companies like Amazon and Colgate on a 0:13:35.836 --> 0:13:37.836 number of fronts. You know, the first is even just 0:13:37.876 --> 0:13:41.556 to understand where is all of that packaging going to 0:13:41.596 --> 0:13:45.836 their credit They and several of their peers have realized 0:13:45.876 --> 0:13:49.356 that there's a paradigm shift possible now from we have 0:13:49.476 --> 0:13:52.796 designed a thing that's technically recyclable, you know, our packaging 0:13:52.916 --> 0:13:56.516 R and D engineers have made this awesome mono material 0:13:56.676 --> 0:14:02.396 HDPE toothpaste tube in Colgate's example, to now trying to understand, okay, 0:14:02.436 --> 0:14:04.596 well we made this thing that is recyclable, is it 0:14:04.676 --> 0:14:08.116 actually getting recycled? And that was a lens that we 0:14:08.196 --> 0:14:12.116 couldn't really get at scale before and so now with 0:14:12.156 --> 0:14:15.516 Glacier's technology, we're able to monitor in real time, you know, 0:14:15.716 --> 0:14:18.956 how much of these tubes are actually ending up at 0:14:18.996 --> 0:14:21.796 the recycling facility, and once they're in the facility, are 0:14:21.876 --> 0:14:23.956 they ending up in the right place, are they being 0:14:23.996 --> 0:14:26.836 sorted correctly such that they can actually be turned into 0:14:26.876 --> 0:14:29.516 new stuff, or are they ending up in the landfill, 0:14:29.556 --> 0:14:32.636 in which case you know, suddenly this recyclable tube isn't 0:14:32.716 --> 0:14:34.956 very recyclable at all. So we're starting to answer these 0:14:34.996 --> 0:14:36.516 really really critical questions. 0:14:37.276 --> 0:14:40.876 So Colgate knows whatever how many tubes of toothpaste they 0:14:40.916 --> 0:14:43.276 sold in a city, and if you are working in 0:14:43.276 --> 0:14:46.796 the recycling facility for that city, you can actually count 0:14:46.836 --> 0:14:49.716 how many tubes of that toothpaste came down the recycling 0:14:49.716 --> 0:14:51.676 line and how many tubes of that toothpaste wound up 0:14:51.676 --> 0:14:54.236 in the bin. I mean, is that the reductive version 0:14:54.276 --> 0:14:54.836 of what you're. 0:14:54.716 --> 0:14:58.236 Saying, essentially? Yeah, And where we're even seeing now, like 0:14:58.316 --> 0:15:01.836 with these rapid advances in AI and in detection, that 0:15:02.436 --> 0:15:04.676 first of all, it's no small feet to even define 0:15:04.716 --> 0:15:06.436 what is a tube and how do you tell the 0:15:06.436 --> 0:15:09.836 difference between a toothpaste tube versus a sunscreen versus a 0:15:09.876 --> 0:15:12.836 lotion tube. But now we're getting to the point where 0:15:12.836 --> 0:15:15.876 we can actually say the brand of toothpaste, it is 0:15:16.076 --> 0:15:19.196 like from all of those visual markings, and so we're 0:15:19.236 --> 0:15:21.916 just seeing this sort of Cambrian explosion of interest from 0:15:22.116 --> 0:15:24.956 a wide variety of different you know, packaging producers and 0:15:25.076 --> 0:15:29.396 brands to really start understanding this previous black box on 0:15:29.516 --> 0:15:32.556 what happens once they release this packaging into the wild 0:15:32.796 --> 0:15:33.836 for consumers to buy. 0:15:34.236 --> 0:15:38.476 So, I understand that California has a law that is 0:15:38.796 --> 0:15:41.676 in some fashion supposed to put companies like on the 0:15:41.716 --> 0:15:44.436 hook for their for their products right after they're used, 0:15:44.436 --> 0:15:49.036 to incentivize companies to have their products be recycled. Right. 0:15:49.116 --> 0:15:52.116 Is that am I characterizing that law right? And is 0:15:52.116 --> 0:15:53.276 it relevant to your business? 0:15:53.516 --> 0:15:56.436 Yes? So I believe you're referring to EPR or extended 0:15:56.476 --> 0:16:00.316 producer Responsibility laws. For those who may not have heard 0:16:00.316 --> 0:16:04.476 of EPR, it's essentially this premise that, you know, if 0:16:04.556 --> 0:16:07.356 our recycling and waste system is supposed to find a 0:16:07.356 --> 0:16:09.756 way to do something good with all the stuff throwing out, 0:16:09.796 --> 0:16:12.076 the people making all the stuff that we're throwing out 0:16:12.076 --> 0:16:14.356 should probably have some skin in the game to make 0:16:14.396 --> 0:16:18.236 sure that that stuff gets either disposed of or reused properly. Right, 0:16:18.316 --> 0:16:23.196 And so EPR laws are already in effect throughout Europe, 0:16:23.556 --> 0:16:26.476 throughout Canada, some other regions, and then they've been passed 0:16:27.116 --> 0:16:29.476 in a number of states in the US, including California. 0:16:30.036 --> 0:16:34.036 Now while EPR in the US is still in its infancy. 0:16:34.116 --> 0:16:36.316 In other words, it's been passed in a number of states, 0:16:36.316 --> 0:16:38.636 but there's a lot of hairiness to figuring out how 0:16:38.676 --> 0:16:42.636 to actually implement the system across all the producers selling 0:16:42.716 --> 0:16:46.076 into a state and all of the recyclers operating in 0:16:46.116 --> 0:16:48.036 that state. It is, I think a step in the 0:16:48.076 --> 0:16:51.476 right direction because in a lot of ways it helps 0:16:51.516 --> 0:16:54.916 to create that circle we were talking about earlier. You know, 0:16:55.036 --> 0:16:58.476 you're seeing that a lot of brands and producers are 0:16:58.556 --> 0:17:01.036 starting to take even more of a vested interest in 0:17:01.196 --> 0:17:04.156 understanding what is happening to all of their packaging, because 0:17:04.196 --> 0:17:06.676 they know that imminently they're going to need to start 0:17:06.676 --> 0:17:09.156 proving the sort of end of life outcomes for that 0:17:09.196 --> 0:17:12.356 packaging in order to you know, one, not be heavily 0:17:12.396 --> 0:17:15.316 fined and then two maybe even have a right to 0:17:15.316 --> 0:17:16.796 continue selling into that state. 0:17:17.916 --> 0:17:21.196 It seems good that Amazon and Kolgay Palmolive are trying 0:17:21.236 --> 0:17:23.356 to figure out if the things they make that are 0:17:23.356 --> 0:17:26.916 recyclable are actually being recycled, but it seems like for 0:17:26.996 --> 0:17:29.156 that sort of thing to happen at a meaningful scale, 0:17:29.236 --> 0:17:31.676 you would need laws basically, right. I mean, if the 0:17:31.676 --> 0:17:34.676 companies are just incurring the cost either out of the 0:17:34.676 --> 0:17:36.876 goodness of their heart or in the hopes of you know, 0:17:37.076 --> 0:17:41.356 generating goodwill that will lead to higher revenues, those seem 0:17:41.436 --> 0:17:45.276 like marginal cases. Are the EPR laws such that you 0:17:45.316 --> 0:17:48.436 think it will become a meaningful part of your business, 0:17:48.436 --> 0:17:50.556 a meaningful part of the world, that companies will in 0:17:50.596 --> 0:17:52.596 fact be on the hook to figure it out, or like, 0:17:52.676 --> 0:17:53.756 what do you think is going to happen? 0:17:53.796 --> 0:17:56.116 You know, I will say that early indicators are that 0:17:56.396 --> 0:17:58.756 all of these states are taking it quite seriously. So 0:17:59.036 --> 0:18:02.396 in addition to requiring a lot of these brands and 0:18:02.436 --> 0:18:06.316 producers to pay into a massive fund upfront to even 0:18:06.396 --> 0:18:09.556 just start implementing some of this movement, a lot of 0:18:09.596 --> 0:18:12.676 these states are also you know, we're seeing that some 0:18:12.716 --> 0:18:15.196 of the kind of like early deadlines and fines for 0:18:15.316 --> 0:18:18.556 non compliance are actually being upheld, which I think is 0:18:18.916 --> 0:18:21.316 a really strong signal to the market. Hey, this is 0:18:21.316 --> 0:18:24.876 something that needs to get taken seriously. Now to your point, 0:18:24.916 --> 0:18:27.036 I do think that at the end of the day, 0:18:27.076 --> 0:18:31.996 whatever flavor this legislation takes. The key to make sure 0:18:31.996 --> 0:18:36.396 that recycling is still a viable and sustainable value proposition 0:18:37.196 --> 0:18:39.836 is that there needs to be some sort of an 0:18:39.916 --> 0:18:43.796 end market for that material, right because let's say these brands, 0:18:44.076 --> 0:18:46.516 even if they're required to pay billions of dollars into 0:18:46.556 --> 0:18:51.196 this EPR system, if there's no one on the back 0:18:51.316 --> 0:18:54.836 end to receive that material that these recycling facilities are sorting, 0:18:54.916 --> 0:18:57.476 then recycling can't really happen. At the end of the day. 0:18:57.676 --> 0:19:00.556 Someone needs to buy the bail of. 0:19:00.556 --> 0:19:05.596 Plastic exactly exactly. But if they can have guarantee that 0:19:05.676 --> 0:19:09.316 there is a buyer on the other side, right that 0:19:09.316 --> 0:19:12.116 that person or that company will buy at a certain price, 0:19:12.196 --> 0:19:15.516 then suddenly they can sustain that business quite well for 0:19:15.596 --> 0:19:18.116 the long run. And so to that point, you know, 0:19:18.156 --> 0:19:20.956 one other model that's often brought up in the realm 0:19:20.956 --> 0:19:25.556 of legislation is actually minimum recycled content laws, because it 0:19:25.636 --> 0:19:28.116 kind of gets at the same issue from the other side, 0:19:28.156 --> 0:19:29.036 where you say. 0:19:28.916 --> 0:19:32.956 Basically creating demand, creating demand for a bale of recycled 0:19:32.956 --> 0:19:35.356 plastic coming out of the recycling facility. 0:19:35.036 --> 0:19:39.716 Exactly, and it kind of disentangles the market for recycled 0:19:39.716 --> 0:19:43.276 feedstock from the market for virgin feedstock, which is another 0:19:43.356 --> 0:19:45.916 great way to kind of catalyze the movement of that 0:19:46.036 --> 0:19:52.556 material throughout that recycling ecosystem. 0:19:52.676 --> 0:20:03.916 We'll be back in just a minute. What are you 0:20:03.916 --> 0:20:05.996 trying to figure out right now? What's a big thing 0:20:05.996 --> 0:20:06.876 you're trying to figure out? 0:20:07.116 --> 0:20:09.716 We are at a really exciting inflection point, a glacier 0:20:09.836 --> 0:20:14.236 because of I think two big things here. The first 0:20:14.436 --> 0:20:19.516 is just how do we scale smoothly and rapidly. You know, 0:20:19.556 --> 0:20:22.356 we've gone from a year ago we were making maybe 0:20:22.476 --> 0:20:25.276 one robot every three months, and now we have the 0:20:25.316 --> 0:20:28.116 capacity to make three to four robots per month, and 0:20:28.156 --> 0:20:30.316 we're expecting to go even faster by the end of 0:20:30.356 --> 0:20:32.796 this year in the next six months. And then the 0:20:32.836 --> 0:20:35.436 other big frontier for us, in addition to just you know, 0:20:35.476 --> 0:20:37.316