#033 – IS THE DRONE INDUSTRY IN A DECLINE? LILY’S DOWNFALL, PARROT’S LAYOFFS, & THE LATEST DRONE INDUSTRY NEWS WITH RECODE’S APRIL GLASER
[00:00:06] IAN SMITH: Before we get started with this episode I want to say from the bottom of my heart thank you for listening. I absolutely love producing this show for you. One thing some listeners will note is that I’ve had a grand total of two different sponsors during the life of this podcast. Unbeknownst to anyone else only one of those deals actually involved any money.
[00:00:28] I’m not going to lie. Money does help some pocket change to take cover recording and production expenses would go a long way to ensure I can continue doing this for as long as possible. With that said please do check out the podcast’s new Patreon account, if you like this show you can choose to donate one dollar a month to the podcast and also earn some really cool rewards like an invitation to join the show’s supporter only private slack group that I’ll be creating where we’ll speak about the drone topics that you want to hear about and get more one on one interaction.
[00:01:02] So go ahead and check out what you can do to help at Patreon.com/dronespodcast Thanks a bunch. Back to the show.
[00:01:20] [Introduction]: Welcome to CommercialDrones.FM, the podcast that explores the commercial drone industry. The people who power it and the concepts that drive it. I’m your host Ian Smith.
[00:01:31] IAN SMITH: Good evening everybody and welcome to Commercial Drones.Fm. I’m here in San Francisco with April Glaser who covers robots drones artificial intelligence and other smart machines for Recode and prior to joining recoat she held a journalism fellowship at Wired magazine where she produced podcasts and reported on the cultural legal and business aspects of autonomous technologies. Her writing has also appeared in Slate, Gizmodo and Back channel. She’s got a degree in philosophy from Temple University as well so welcome to the show. April thanks so much for being here.
[00:02:08] APRIL GLASER: Thanks Ian. Great to be here.
[00:02:10] IAN: Yeah so can you tell us first of all like what is recoat. I mean some people might not know what Recode even is?
[00:02:16] APRIL: Sure. Yeah. Maybe it’s a bit smaller of a publication but amongst folks here in the Bay Area and Silicon Valley, it’s definitely an influential publication because we report on the business of technology. But really the nitty gritty stuff and media as well. So when people are leaving when money is changing hands when there is an epic failure we’re usually the first to break those stories.
[00:02:43] Our executive editor is Kara Swisher who is one of the most, if not the most celebrated technology journalists in the history of technology journalism. She started Recode a few years ago and it was recently acquired by Vox.com. And so we’re kind of the tech business publication with Vox Media.
[00:03:05] IAN: So Vox also owns I think the verge and some other things. Ok cool so. So how did you get into. So you’re you’re a journalist Recode. You report on drones, robots autonomous vehicles all this technology, but how did you get into that? I mean was this something that was kind of suggested to you? or you were like oh hey I really like these drones. I like these. These you know this technology these robots like what’s the story behind that?
[00:03:31] APRIL: Well I’ve always spent my life focusing on technology policy and kind of the legal implications of technology. But as a kind of a geek I was always kind of a hardware geek. Did hardware hacking in different things and had been following the robotics industry for a while. You know a member of a hacker space people are always building cool stuff.
[00:03:52] When I saw that this opening was coming up I had been going to academic conferences on robotics and and meeting people who are in the drone industry and for the past couple of years and had just developed a very sincere fascination with how quickly it was moving with the idea of flying robots. This kind of uncapitalized air space and what it could mean and the fact that it was in the making now. I thought that this was a time that I could jump in and really report on an industry as it’s forming. And you know I’ve been at recoat now for about four months and it’s moving quickly already since I’ve been there and it’s just a very exciting time to be following robotics and drones.
[00:04:35] IAN: Yeah you. I was just looking through in preparation of the show in four months you’ve written quite a few articles. Actually it was like over 140 or something like that so nicely done.
[00:04:46] APRIL: Thank you.
[00:04:47] IAN: Putting in the work.
[00:04:47] APRIL: I think and to the people who are listening you guys are doing a lot of work and I feel like there’s just an interminable amount to write on and I’m you know sincerely interested in all of it so I have definitely been typing away.
[00:05:01] IAN: Well, we’re very glad to read your story. So you were recently at CES 2017 in Las Vegas Nevada. Consumer Electronics Show, historically obviously most people know about it but it’s you know a lot of technology is announced there, most of it usually being consumer.
[00:05:20] And there is a drone area at the show? like what’s the deal? I mean I want to talk about your experience at CES. Like maybe you can give us a brief synopsis like what did you see there and what was interesting?
[00:05:32] APRIL: A lot of dudes. I’l say that first and foremost is that there are…
[00:05:38] IAN: Typicaly.
[00:05:38] APRIL: Definitely, a lot of men at CES and not as many women but hopefully every year that will get better. But no there is a huge drone section.. I wasn’t there the year before. But what I heard is that last year the drone area was actually quite bigger and this year it was more consolidated. And perhaps that’s a reflection on the industry. But there were all kinds of amazing demonstrations and it was really fun to just see everybody show off their best technology the the stuff that’s not out yet that’s going to be out soon.
[00:06:09] People who are actually geeks in this space get to quiz engineers and it’s a really kind of exciting environment with so many things happening there and it’s not just drones at CES, yes of course there’s also a lot of cars and and robotics and you know sound systems so it’s really just a place for 200,000 geeks to descend on Las Vegas which is already kind of an intense place to be.
[00:06:34] But the drone area was really exciting because you know they set up all these cages where drones are flying around in these sets and doing all these demos. Of course DJI had an amazing booth. Right now it’s not necessarily that legal to fly drones over people in populated areas particularly inside of a building but they had found a way to do it. They set up this amazing kind of clear sealing area where you could look up in this kind of glass cage and see drones flying overhead.
[00:07:08] IAN: I think I did you post a picture of that on Twitter maybe.
[00:07:10] APRIL: I tweeted that, I think one of my favorite things though yes was actually Qualcomm’s demonstration, there not a drone maker but they have they make processors right. They’re a chip maker. And they’ve been working on their Snapdragon drone platform which is this kind of combined machine learning and autonomous flight platform. So as the drone flies it can actually name what it’s looking at in the room.
[00:07:38] And so it’s not using GPS to navigate and I thought that was really cool so they had this kind of set up that mocked up the inside of a warehouse. And it was it was interesting to me because jeepneys doesn’t work inside. But there are a lot of applications where you’re going to want to fly drones indoors you know. And so that’s where machine learning really makes sense with autonomous flight because it was learning what it needed to avoid.
[00:08:01] You know as it saw something gets it oh that’s a ladder that’s a barrel and avoided those things and then actually was able to map out the room and fly faster as it learned to the room around it. And I thought that technology was really impressive and and really indicative of of the type of artificial intelligence that we’re going to be seeing on board drones in the future.
[00:08:20] IAN: Yeah that’s going to be really interesting. I remember that Qualcomm announcement that was pretty yeah I think everyone kind of took that as like OK this is where this technology is headed because these processors which power – I mean they’re flying computers. I think it took that from one of your stories actually. But they’re yeah they’re just flying computers and it all depends on the processing power on board. So it was it was indicative of what’s to come for the drone industry once everyone gets their hands on those chips.
[00:08:48] No yeah. The DJI thing actually. So that was their booth. What you’re saying. So they were flying drones above people but in like some type of transparent like, there was like a transparent barrier between the thought the drone and the people and so if it failed it wouldn’t knock something on the head?
[00:09:06] APRIL: Yeah I just thought it was really cool. And and that’s that’s what they had. I mean DJI obviously just makes some of the best drones on the market just in terms of their engineering. And it’s just really neat to see a bunch of drones flying over your head in a glass box. You know just really as as someone who’s enthusiastic about this technology I totally geeked out at that booth.
[00:09:30] There was also a Polaroid actually had a number of drones I think. Yeah which I wasn’t aware of until now. But they had about eight, I think a couple of them are ready to market but they had a 8 drone set up and I guess many more are going to come to market in the next year. Kind of taking their photography expertise onboard but also they have this amazing consumer facing brand. You know and so I think we should expect to see drones come out of companies we didn’t necessarily think were going to at.
[00:10:01] It was great to see GoPros’ Karma which I hadn’t seen yet I had only heard reports of because it was recalled so quickly after it hit market.
[00:10:08] IAN: They were there?
[00:10:09] APRIL: GoPro was there and at the CES, Nick Woodman made an announcement that the Karma is going to launch again in February. So apparently they figured out whatever the battery malfunction was that was causing the aircraft to fall from the sky which is not what we want. They fixed that apparently and they’re going to try to come back to market. I was surprised to see that it was a lot bigger than I thought it was going to be.
[00:10:34] IAN: The actual drone.
[00:10:35] APRIL: Yeah yeah. Whereas the Mavic I don’t know if you’ve seen like the comparable one. It’s quite small. And the the Karma’s small too, it just requires its own carrying case. You can’t like throw it in your backpack. That type of thing and so it was it was interesting to finally see that because I had reported on it there was a lot of hype around it. And of course then there was a huge snafu with the battery malfunction where they had to recall all of them. The hardware is hard but making hardware that flies is even harder. You know so.
[00:11:06] IAN: Absolutely. About the size difference on the Mavic and you know I thought on the announcement or the launch events for DJI and GoPro for both of their drones. Nick Woodman you know theirs was a week earlier – the GoPro announcement and he pulled out a backpack with a drone in it and then whenever DJI did theirs I thought it was kind of like a tongue and cheek jab is how I took it. When Michael Perry was on stage and he pulled the Mavic from his back pocket and not even a backpack.
[00:11:38] APRIL: Yeah.
[00:11:39] IAN: That was funny. We all know, now that you mention GoPro, I’m not that surprised they were there of course they have an entire other line of products besides the Karma. And you know we all know that GoPro makes some good cameras that can actually make some pretty good launch videos for certain products.
[00:11:57] APRIL: GoPro makes some amazing camera. I mean I think what you’re alluding to is the recent case that has come to fore with with the Lilly drone that never came to fruition. And we learned that in their video which had really created a lot of hype a couple of years ago and in fact generated 34 million dollars in presale orders.
[00:12:19] Some 60,000 pre-sale drones were sold of course not. The drones weren’t sold because the drones didn’t exist yet. But but in that video that they shot that that got all that excitement going on and was responsible for getting all those sales as well as something like 15 million in Series A you know – Venture capital funding a good video can go a long way. That was shot with a GoPro as well as a DJI Inspire.
[00:12:48] IAN: And that was kind of a dirty little secret. Would we say as such, I mean you you just wrote an article so we’re going to talk about some of the articles that you’ve written in the past month which are all super interesting there’s so much going on right now in the industry just in January. It is ridiculous. But the whole Lily thing.
[00:13:10] Like OK I have strong opinions on this. I never fell into the hype. I can honestly say that I didn’t think the platform was that good. I was actually shocked at the the reception that it received because I just knew from experience or I thought I knew I guess I was somewhat correct but that was really hard to do. It just seemed like like a something that you just the technology wasn’t even capable of doing that yet. And so anyways. Yeah you wrote this article. Now we’re getting sued.
[00:13:41] APRIL: They’re getting sued by the district attorney’s office for the city of San Francisco and they’re getting sued for corrupt and fraudulent and misleading behavior. It’s a consumer protection lawsuit because the D.A. is alleging that they had lead customers on to think that this drone existed and that it was going to exist and and that it was very manufacturable and the DA had amassed evidence over the past few months in their investigation what they launched or what they went public with the same lawsuit and public the same day that Lilly announced it was closing which was just a very traumatic day for the drone world.
[00:14:26] But yeah and they basically said that Lilys you know put a video out that misled customers and think that they were going to get a product that never existed and that Lily wasn’t making and that that wasn’t going to come to market as well. They alleged that Lily had broken an FTC Federal Trade Commission law that says and this is kind of a preorder law if somebody or somebody prepays for a product and they don’t get it within 30 days of the promised delivery date then the you know quote unquote manufacturer is supposed to give them the option to get a full refund back or they have to consent to a further delay.
[00:15:06] And Lily never gave any of their preorders an option to consent to further delay they just said oh it’s going to be another six months. Bear with us or it’s going to be another three months. They just kept announcing further product delays and never never gave the option to for people to get their money back. That said you know if people had requested it perhaps they would have fulfilled.
[00:15:30] I’m not sure of the details of like you know specific cases but there were supposed to give that option explicitly and people had to consent her also had to issue that refund. So that was part of the lawsuit as well. A lot came out in that lawsuit actually.
[00:15:43] IAN: One of my favorite things is a quote from Antoine Balaresque, the co-founder of Lily, and it says ‘I’m worried that a lens geek -It’s an email. It’s I guess it’s case files or something.
[00:15:58] APRIL: Some discovery from the lawsuit. Exactly.
[00:16:01] IAN: It says, “I’m worried that a lens geek could study or images up close and detect the unique GoPro lens footprint. But I’m just speculating here. I don’t know much about lenses but I think we should be extremely careful if we decide to lie publicly.”
[00:16:17] I don’t know how careful they were in the end, but they got caught and doing some really shady stuff.
[00:16:27] APRIL: You know it’s interesting to think about what this will do you know for the industry and for trust and drones. I mean you know this comes at a time, the same month we’ve seen Parrot drop a substantial amount of its workforce that was dedicated to drones. You know late last fall rather 3DR announced it was no longer going to be making consumer drones and kind of moving away from drone manufacturing. Lily was something people were really excited about, another drone manufacturer that that wasn’t able to bring their product to fruition.
[00:17:03] You know this really comes at a time when the drone industry should be developing a lot of hype and people should be getting really excited about these products and also not to mention GoPro again their drone failed within weeks of it coming to market they had to do a complete recall. I mean it’s just one after the other after the other and everyone seems to want to drone I mean I think the FAA said over 600,000 drones were registered last year. You know this is just happening so quickly in terms of the demand. But but, we’re not seeing the companies really deliver. So it’s a very interesting market right now for sure.
[00:17:40] IAN: And in the meantime DJI is kind of sitting pretty in the corner just hands folded on the lap, kind of watching this unfold which I’m sure is great from a business standpoint but I’m sure that you know they you know deep down you have to have competition and it’s great to have all this competition so it’s not you know Lily doing all this stuff. Yes bad business practices.
[00:18:04] It would have been really cool to see that kind of platform exist and I think the closest it’s ever existed is the actual – the Mavic pro from DJI which is really interesting.
[00:18:14] APRIL: I mean that’s the thing is that so Lily announced this product in 2015 by the time they announced it and by the time that they failed to bring it to market they announced it was going to happen. The Mavic came out which has just a lot of the features that were advertised in the Lily, right?
[00:18:31] You know one thing that you mentioned you saw the video and I’d actually be curious to hear what to you didn’t seem like was going to actually work or why you were a skeptic about the video because I’ve been speaking to engineers for the past week or so as I’ve been writing my report, just trying to dissect where they were lying or where they were over promising.
[00:18:51] One of the points was that people made is that Lili’s drone had a kind of a cylindrical shape which is the most drag inducing shape and you wouldn’t you wouldn’t see that on a drone. Another one was that their drone could pop out of water go in and pop out of water. That takes a lot of propulsion power and the size of their drone just wouldn’t be able to carry a battery load that that could not that drones can’t go in and out of water that’s certainly possible but that takes a lot of power to break that surface you know.
[00:19:21] IAN: Did you ever actually like hold one? like did you get to feel like? Cause I remember there was.
[00:19:25] APRIL: Did you?
[00:19:26] IAN: No I didn’t but I know that there was one or some. There was one. There was one prototype.
[00:19:31] APRIL: One.
[00:19:31] IAN: That I know about, that I heard about and I saw a picture.
[00:19:34] APRIL: OK.
[00:19:34] IAN: On Facebook and some shop and some gadget stores somewhere here in California and to me – my first thing when I first saw it was that there’s no way that that thing is going to fly because the the materials itself look like it was just so darn heavy, and it didn’t look practical at all for like a flying machine. I don’t know I just had this feeling I guess I mean I definitely wasn’t looking at the aerodynamic.
[00:19:59] APRIL: Sure.
[00:19:59] IAN; Shape of it and everything because you know worse shape things have flown. I’ve seen them.
[00:20:05] APRIL: That’s true.
[00:20:06] IAN: But you know this is just an insane story. I mean you’ve got Lily off the heels of of 3DR and then you know Parrot. So we were kind of talking about Parrot. So they laid off a third of their drone division which is about 290 employees. Is this, do you think because of DJI?
[00:20:29] APRIL: I do think that companies just can’t compete with DJI and it’s because DJI. Well first of all it’s because drones are very hard to make well. You know Parrot obviously was making good drones but you know 3DR also had problems with their manufacturing. Their drone wasn’t delivered with the gimbal that it was supposed to have and other things that are incredibly essential for you know a drone that’s supposed to be for aerial imaging. You know Lily didn’t come to market.
[00:20:57] DJI is just- their engineering is so good. They’ve been in the kind of drone and recreational, aviation, kind of hobby, RC space for something like over 10 years. And so you know just now drones are starting to become popular but they have all of this time to fail and experiment and really perfect their engineering you know. And it’s just incredibly hard to keep up with them. They also manufacture in the same place where they design in Shenzhen and that type of kind of vertical style of integration where they get to iterate right there on the spot. I think also makes them quite hard to compete with.
[00:21:38] They also don’t necessarily have the arrogance that we see a lot of these Venture Capital funded companies have I would say. Just because you know they didn’t. They don’t launch their new drones. And maybe because they don’t have to, because they’re already so successful with these just incredibly shiny videos and make all these big promises. They actually just have that the drone is there and ready.
[00:22:00] And so yeah they’re incredibly hard to compete with that there’s I think a constellation of reasons why Parrot also which was you know probably the second most formidable kind of competitor to DJI is starting to inch out of the consumer drone space and they’re going to focus on commercial operations which makes sense because commercial drones are a huge market and it’s growing as people discover more applications and consumer drones are also a huge market. But there’s a lot that can go wrong there. You know that again DJI the a big competitor with Mavic the GoPro karma couldn’t stand the sky. It’s just I mean there’s a lot of different reasons why they keep emerging. A lot of it’s the follies of their competitors.
[00:22:48] IAN: Yeah for sure. So I’m taking this from your article by the way. But they missed their quarterly sales estimates by 15 percent. So like you said they’re going to switch their focus to commercial, but consumer drones for Parrot generated more than 80 percent of its 60 million dollar revenue.
[00:23:05] APRIL: Yeah.
[00:23:06] IAN: In Q4 of 2016 which is just insane. I mean drones are big business. This is a commercial drones.Fm podcasts, like we need the figures to back all this up and justify. We need proof and that’s some proof in pudding, but even despite that Parrot looking to I guess leverage more of their company’s sensefly which specializes in commercial more industrial grade drones which is a space that has big big margins. I know for a fact some massive margins especially on fixed wing aircraft. If you look at fixed wing aircraft they are expensive as hell they are hard to do. But if Parrot doubles down on that – I think they can really make a nice run at DJI, because DJI just doesn’t have anything fixed wing, I don’t know
[00:23:54] APRIL: And that’s the thing is in the commercial kind of industrial drone space there’s not a clear dominant player right now this is definitely a place where someone or a company can emerge as a dominant player not only because, there’s not a DJI to compete with, but also because there aren’t the regulations that the consumer industry has to grapple with. You know we have Part107 now. We know the rules. You know it can an industrial operation isn’t trying to – I mean delivery is like the smallest subset of these applications, you know they’re not trying to buzz through busy city streets necessarily or trying to you know fly over people at night. They want to work during the day when other people are working. They just don’t have the same regulatory hurdles that that delivery drones will have.
[00:24:43] And especially in that consumer drones will have also they’re also going to be operated by professionals. So they’re a lot less – a lot less potential for bad bad press.
[00:24:54] IAN: Yes.
[00:24:54] APRIL: You know DJI has been experiencing this over the past year. You know people buy their drones and can put anything on them and fly them over anywhere and DJI doesn’t have control over what people do with them. With commercial operations you know that your your PR is a lot safer like kind of what you’re going into in terms of potential bad headlines that could emerge out of your drone.
[00:25:19] Another thing that’s that you know with DJI to think about is I think at least this is what Chris Anderson told me and I put this in my story. I had a chance to speak with him. He’s the founder of 3D Robotics which was also a competitor to DJI and the consumer space but is now focusing like Parrot focusing on its commercial applications and kind of industrial software. But but he said that DJI in 2015 dropped prices of their consumer drones by 70 percent in about nine months. And he said he had never seen anything like that before and that was very difficult to compete with.
[00:25:59] And whether it was DJI that dropped the price or the people that were selling them. However that worked out. The prices dropped to a point where competitors just couldn’t compete. Parrot has a really good chance to I think do well in the commercial space because like DJI they’ve been doing drones for a lot longer than most other folks. And that experience goes a long way because like I said hardware is hard, making hardware fly is even harder and Parrot actually you know they make good aircraft.
[00:26:29] IAN: They do.
[00:26:30] APRIL: So they have a very large fleet of commercial drones. I think over 10 Sensefly which is their commercial subsidiary has a lot of different products available for many applications but, they don’t have the software. And as you know there’s a huge ecosystem happening right now where people are building all kinds of software applications.
[00:26:50] IAN: Yes DJI is definitely playing to win. I mean you can see that but also you know one thing that Parrot will inevitably run into is DJI just launched an Enterprise division DJI Enterprise.
[00:27:01] SAARAH: They did.
[00:27:01] IAN: And so they are not going to sit on their laurels of course. And as soon as they see some more traction I mean, it was just so obvious that it seemed like right after the Karma launch, that our Mavic pro launch is actually going to be a week later because, if you look at the timeline when GoPro announced the launch announcement date. Then DJI had time to you know – after the leaks of the Mavic on social media and stuff and the pictures, to adjust their launch announcement day to be after GoPros’ and I’m sure that they took all the strategic information from that to you know doctor up there their announcement really quick can make any last second changes if they needed to.
[00:27:42] But in the end it looks like the hardware failure of the Karma kind of helped them too. But.
[00:27:48] APRIL: To DJI’s credit I did hear that the timing of the karma announcement and the Mavic announcement was incidental.
[00:27:56] IAN: Really?
[00:27:56] APRIL: Hard to know, I will say that a lot of products try to come out in time for the holidays.
[00:28:02] IAN: That’s true.
[00:28:02] APRIL: Actually DJI was unable to get the Mavic, they ran into a snafu with their production as well.
[00:28:10] IAN: Yes, that’s right.
[00:28:10] APRIL: But I think handled it very responsibly. But definitely there were some back orders and people didn’t get their Mavic under the tree in time for Christmas.
[00:28:15] IAN: We can’t forget that.
[00:28:16] APRIL: So I think that’s important. Like even they also you know had a little issue that they had to access out and people are now starting to get them and I’m hearing all kinds of reports of people flying them and seeing them.
[00:28:27] IAN: We can’t forget that. And also we can’t forget like it wasn’t always so easy for DJI it still isn’t easy. They went through so many I mean if anyone remembers what a flyaway is like we hardly nobody really talks about that anymore.
[00:28:41] The kind of thought is that if you have a flyway these days like you know something really really rare happened inside the aircraft with either you know some weird software or hardware bug. But they used to be just known as like the company. Oh don’t buy a Phantom is going to fly away and now it’s like it’s literally like the best drone that you can buy for so many applications.
[00:29:02] So earlier you had mentioned delivery. So one really cool story that you just wrote recently, very recently. It might have been the latest one that you did was “Robots delivering food to doorsteps” We’re gonna switch from the sky. We’re going to come back down to Mother Earth, down to the ground. So what’s the deal with this? Little robots delivering food to people? What’s going on?
[00:29:26] APRIL: You know I like going from sky to ground because I do imagine a type of ecosystem where drones might drop a package down and then like in a central location because drones can’t really drop anywhere and then a robot will pick it up and take it to like an outdoor ground robot. Which I’m about to knock about and then they also have indoor robots that take things to say like in a hotel room or in an office. And there could be just a complete automated, you know delivery pathway.
[00:29:56] IAN: How depressing.
[00:29:56] APRIL: I know, that’s the thing.
[00:29:57] IAN: When you say it like that, it’s like it sounds so smart but.
[00:30:00] APRIL: I don’t know, yeah – I still use a French press and basically only own a phone charger so. I get that it’s depressing. But it is something that makes sense particularly because on demand startups, have also been dropping like flies it’s an incredibly difficult business model to make work because of worker turnover, because of just having to stockpile things. –
[00:30:28] IAN: When you say on demand?
[00:30:28] APRIL: Like Mealstar, so there was Sidecar which was a competitor of Uber that’s no longer there, Spoonrocket was a meal delivery startup that’s no longer around. There was a I think it was called Homejoy which was a housecleaning startup that was on demand no longer around.
[00:30:45] IAN: Oh, wow.
[00:30:45] APRIL: There’s just a lot of reasons why this business model doesn’t work. When it comes to On-Demand delivery though, Caviar is one, Postmates isn’t profitable actually, even though they have a ton of funding. You know it’s very hard to make these these things profitable.
[00:30:59] Robots, and I say this with a bit of a cringe actually makes sense. Right? Not to take the place of people but it’s actually like people just aren’t doing these jobs, they don’t want to do them. They don’t do them, they do them slowly, it’s very hard to get that one hour delivery turnover that Amazon is trying to do or even two days that they’ve really got to set up to want things so quickly.
[00:31:26] So what I wrote about though is is a technology called or not a technology rather but a company called Starship Technologies and that were started by a couple of founders from Skype and you know who knows if this will really take off. You know these things come and go all the time. But they did just get $17 million in seed funding and that matters. But what they’ve developed is a little ground delivery six wheeled robot to go from say a restaurant to your house and it’s autonomus, it’s supposed to go without people. And I guess you know you put the takeout in the robot if you’re at the restaurant and you you put in your phone app the place where it’s going to go in the house.
[00:32:09] And of course it wouldn’t really work at an apartment or if it you have to come downstairs can’t go your – it can’t go upstairs. And then you come down and you pick it up out of this kind of glorified cooler on wheels that’s driving by itself. And it’s there and you already paid you know through your phone. There’s a lot of assumptions that will need to actually play out for this to work where I used to. I’m from Nashville but I lived in Philly for eight years which is a place where I could just imagine these robots getting spray painted and kicked over and stolen. I mean .
[00:32:43] IAN: Delivering delicious cheesesteaks.
[00:32:45] APRIL: Or delivering delicious cheesesteaks, right. They’re not big enough to hold a pizza, I think that’s important to remember that’s a big delivery market. But there’s definitely I think they’ll work in some places and not in others and maybe they’ll work in others eventually but it’ll definitely take some kind of cultural acclimation for people to really be like OK I’m not going to steal this.
[00:33:06] IAN: This is my friend.
[00:33:07] APRIL: I’m going to kick this over because it’s a stupid thing on the street you know. Like I kick everything. But no, but they are now delivering in Redwood City which is a city in Silicon Valley just south of here as well as in Washington D.C.. The catch is that they will be autonomous eventually. Apparently all the tech is there they’re ready to go. I’ve seen them before you said you’ve seen one but right now they need a handler.
[00:33:36] IAN: A – handler.
[00:33:36] APRIL: So basically I would want this job.
[00:33:38] IAN: All I’m thinkin about is the Thanksgiving Day dog show, where like the handlers are like coaxing the dog to like prance very like –
[00:33:46] APRIL: You’re thinking correctly. Yeah actually, so there’s actually somebody walking alongside the robot as it makes its delivery at least for now. The idea is eventually they won’t have to. They’re not required to a handler by law, I asked if this is part of the regulation I said no they can go off on their own.
[00:34:04] IAN: Oh, that’s interesting.
[00:34:05] APRIL: But right now they want to collect data. They want to, in case something terrible happens they want to make sure you still get your take out. You know they’ve been delivering in London for a few months now since last year. So right now they have a deal with Doordash which is delivery on demand service here in California and they have a deal with Postmates which is a very popular service and they’re piloting these in D.C. and in Redwood City.
[00:34:33] We’ll see how they go. I mean I’m going to be tracking this. It’s really interesting to me just to think and kind of to put my like sci-fi hat on and get kind of excited about weird futures having robots sharing the sidewalk with me you know delivering things around it also kind of bums me out, Just because I like people. But like I said we do see this kind of on demand startup space. Just be an incredibly difficult thing that people actually want so it’s a business model that makes sense in theory like you know I want stuff to be brought to me right now. You know yesterday, I’m hungry. I’m a 30 year old impatient millennial.
[00:35:14] But but when it comes to the economics of it it’s just really hard to make work. So maybe bringing in you know a little, you know robotics thinking actually makes sense in this space. And you know what we’ll see how it goes. But this just started at the beginning of this week. I know it started. It started Wednesday was when this news launched.
[00:35:37] IAN: The whole societal kind of acceptance of this reminds me of like the Google Glass thing I guess we’ll. It’s a little bit different because people – well some people got probably got hit in the face because of it. Poor robot, I hope it doesn’t get hit but I saw one of these in person it’s actually a lot bigger. I think what the dimensions is to about 2 feet tall and weighs about 40 pounds empty. And it kind of briskly no, not briskly like kind of strolls along the sidewalk at a –
[00:36:03] APRIL: It’s walking speed.
[00:36:03] IAN: At a four mile per hour pace so, cool. I’ll take a robot delivery any day. I think that’s really cool and it’ll be fun to see who’s going to do the first aerial – like you mentioned, aerial drone to you know ground base terrestrial drone handoff delivery back over to the house and I’m sure there’s going to be some companies really trying to work with Starship Technologies on that. Awesome Well shifting focus a little bit.
[00:36:36] This is this has been in the news like mega mega lately. The FAA has issued the largest fine ever. $1.9 million fine down to $200,000. So it was going to be 1.9 million, down to $200,000. What’s the deal behind this fine? You wrote a storyon this.
[00:36:57] IAN: So I have a question standing with the FAA. Hopefully though they’ll answer my information request soon. But I want to know why the fine dropped by 90 percent. That’s a crazy drop. In October they issued a fine to a company called Skypan International based out of Chicago, for conducting what they said were 65 illegal unauthorized, reckless actually was the word that they used. Flights over Chicago and New York City, over the course of I think between 2012 and 2014.
[00:37:27] IAN: Two years of.
[00:37:27] APRIL: Two years of reckless flying and you know this is the most congested airspace in the country. It’s very important that especially as we’re figuring these technologies out and the regulations out that pilots and operators, kind of get to know each other. When I talk about pilots, I mean like people who are flying airplanes and actually manned aircraft, you know don’t become afraid of drones. It’s just a very sensitive time. And I think for the industry so it’s definitely we don’t want to see this kind of behavior now if this is if this is an industry that’s going to to actually get as big as as people are projecting it will.
[00:38:05] APRIL: But SkyPan was threatened in October with A $1.9 million fine. Earlier this week the FAA announced that the fine has been dropped to 200,000 substantially. I will let you know when I find out why? But they are now required to pay that fine as well. They have been threatened with a $150,000 fine, if they violate another FAA rule this year.
[00:38:33] And they’re also required to make three PSAs, like Public Service Announcements about drone safety with the FAA. I don’t know if they’re going to be like ‘hey don’t do what we did.’ But essentially that’s their slap on the wrist. And like I said it’s such a dramatic number change a 90 percent drop. 1.9 million dollars would be a huge fine and really send a message to to the industry that the FAA is very serious about making sure drones are safe and that they’re following the rules as they come out. And of course the rules are still being written so it’s very confusing and understandable.
[00:39:11] One thing to remember is that Skypan had actually violated the FAA rules before they the Part107 rule came out in August. So this was at a time when they had to get a special waiver to conduct these flights, right. So they couldn’t. I mean of course now they couldn’t fly them over these congested cities right now anyway. But this was before there was as much clarity as there is now. Even so they were they were violating the rules, they were breaking the law. $200,000 is still a lot of money.
[00:39:45] IAN: Yeah.
[00:39:45] APRIL: For sure. 1.9 million.
[00:39:47] IAN: It’s better than 1.9 million for sure. But I mean the FAA It’s all about safety. It’s all about safety. So there’s been questions like, how will the FAA reinforce you know these rules? or enforce the rules rather, and it looks like you know they’re kind of sending a message here like 1.9 million would have been a message a 200,000 is still a message.
[00:40:10] APRIL: It would cripple a lot of new companies for sure.
[00:40:13] IAN: Oh yeah, you wouldn’t be doing business. Actually that’s interesting that they did they did not revoke their operator certificates like allowing them to continue to operate their aircraft so that’s interesting. Please keep us in the loop on that and update us.
[00:40:28] The next kind of story that we’ll talk about we’ve got just two more that we’re going to touch on, these paper airplanes. This is really cool so I saw this story coming across social media. I didn’t really read it too much. I thought paper airplane drones OK. That’s you know kind of like yeah paper paper and you stick a motor in it and then it’s a drone. But I was just reading your story on it it’s actually fascinating. Tell us about these these paper airplane drones that could one day save our lives.
[00:40:57] APRIL: Sure. So. This is thanks to a grant from DARPA, the kind of experimental arm of the military that also helped to develop autonomous driving technology which has led to a whole new industry that’s now burgeoning. They gave a grant to a startup or not a startup or other research space here in San Francisco called other lab. And to develop this disposable drone technology and these drones are really neat. They’re made out of cardboard there. And the idea is that they can carry medical supplies to places that have poor road conditions or that people really shouldn’t go.
[00:41:39] And the inspiration for this I speaking to one of the engineers her name is Star Simpson and she was saying that the idea is that like imagine during the Ebola outbreak there were so many places where they couldn’t get vaccines fast enough. And also people couldn’t go there. And so a drone just made a perfect sense. But also they didn’t want to. It makes sense to have one that just goes one way and that’s disposable and one that anybody can can can create and fold up.
[00:42:06] So what they’ve developed is these kind of cardboard drones that come in this flap packaging that anyone can fold together and assemble and then package with a vaccine or medical supplies and then whoever the like. They have a small computer inside so they could be programmed then to fly to a certain location and they don’t need to be sent back because they’re so cheap to make and they are disposable so there are it’s a one way drone. They’re not drone is maybe not the right word. There certainly are autonomous aircraft. So I would put them under that category. But but they’re gliders.
[00:42:44] IAN: They don’t have a motor –
[00:42:46] APRIL: They don’t have a motor, they glide –
[00:42:47] IAN: Like that’s the kicker, that’s why this is so crazy, it’s a one way –
[00:42:49] APRIL: They have sensors that actually change the surface controls of the aircraft so they can, you know change their wings or their rudders to go – so that way they actually land in the exact correct place. But they’re super light, they don’t have a motor, they’re deployed from another aircraft like a helicopter or another like airplane or something like that. And then they just go to their destination and they’re one way.
[00:43:16] So because there one way they can go twice as far as another drone that you come back you don’t need a comeback. It’s a pretty fascinating technology and you know and it’s again I’m thinking about you know autonomous aircraft in places where they can really solve a real need you know and and really help people. Thinking about all of the technical possibilities of you know small aircraft and what they can carry without you know on piloted aircraft. It’s pretty exciting that they’re they’re thinking in these terms and that they’re really thinking of emergency response as well as just incredibly cheap material. I mean you know it doesn’t have to be super fancy it doesn’t have to be produced in a factory and all sourced with all these crazy parts.
[00:44:06] This is a cardboard kit that you just fold together tape up put put the medicines on and then whoever you’re working with will program it to fly to the right place and right. It’s a research project right now. They’re looking they’re actually working with a I guess like a biotech start up about making the the cardboard material, actually making it out of mushroom, like mycelium type filaments or something like that.
[00:44:34] IAN: Edible?
[00:44:34] APRIL: No so that way it can –
[00:44:34] IAN: OK, just biodegradable –
[00:44:34] APRIL: So that way it can biodegrade. But it would be made out of a mushroom based material.
[00:44:40] IAN: Wow.
[00:44:40] APRIL: Like a mycelium based material. So that’s pretty cool in all the different spaces that they’re collaborating with to make these disposable aircraft. One thing that’s pretty neat about this is that it actually comes from a DARPA program that is all about kind of disappearing technology. It’s DARPA’s vanishing programable initiative. And so they’re also working on like a type of glass that’s like shatters and dissipates. Xerox is working on that. So it’s neat that DARPA’s working on all this technology that disappears. It’s like disappearing, evaporating, disposable tech.
[00:45:21] IAN: Dissplsable tech.
[00:45:21] APRIL: From a science perspective it’s fascinating. I mean talk about – obsolescence.
[00:45:29] IAN: Frozen bullets.
[00:45:29] APRIL: Yeah it’s also a little terrifying. But this we see a health application so that’s that’s pretty neat.
[00:45:34] IAN: I’m fascinated I’m going to be following that very closely. The last headline here it is, “The U.S. government showed just how easy it is to hack drones made by Parrot, DBpower and Cheerson” That’s kind of alarming.
[00:45:52] APRIL: Yeah, So you know drones are computers and they’re often connected to the internet and anything that’s connected to the internet can be hacked. It’s vulnerable and drones are no exception. The issue with the difference with drones is that they’re flying and have propellers that can fall out of the sky when they’re hacked, right. And so security when it comes to drones and actually like having information security and encryption in terms of you know between the operator and the aircraft as well as the open Wi-Fi network that a lot of these drones operate on is incredibly important.
[00:46:28] So the FTC the Federal Trade Commission had been researching this all last year. They kind of spent many months looking into the security of drones and they presented their research in October. I submitted a FOIA request to find out exactly what drones they were referring to, because when they presented their research. They said. Oh well we have these three drones that we were able to hack into and two of them we could take complete control of the flight path, we could turn off the aircraft, and cause them both to fly from the sky.
[00:47:04] On all of them, they were able to take the video feed over and see what the drone was seeing and not let the operator know that this was happening. And I was like oh I want to know which companies did this. And so like a journalist – like I do all the time I submitte FOIA request and found out that the three drones, one was from Parrot which is we were talking about before a very popular drone company it was their AR Drone Elite quadcopter.
[00:47:30] Another one was from DBpower, less popular but certainly they make a lot of consumer drones. These are all kind of under $200. All of these drones we’re talking about so these are the kind you’d get off the shelf and give as a gift to your nephew or niece or whatever. And that was the Hawkeye 2 from DBpower. And then Cheerson another drone company their Onecase CX10W was hacked as well.
[00:47:55] And yeah, they were just able to show that was – another thing they were able to show was that all the smartphone apps that came with the drones gave absolutely no notification when somebody was hacking into the video feed so you know you have no idea this is happening. And it really just I think illustrated particularly the fact that they could commandeer two of them and take over the flight path and make them fall to the ground. How important security is and encryption is to think about when it comes to these drones because they were able to do this by just sitting at a computer and hacking into the computer system. And they were you know not near that. They don’t have to be right next to it. You know they’re just they’re they were just hacking.
[00:48:43] IAN: These are going to be big issues coming up. I was listening to something on NPR that was a guy, it was basically kind of like a Ted talk and some guy was talking. He’s a researcher and he specializes in cybersecurity and you know the internet of things. It’s not just drones that this is going to be happening, to its autonomous cars. There were cases where autonomous driving vehicles could be taken over or not even like fully autonomous ones like semi auto, you know optionally piloted vehicles if you can say that and that brakes could be activated. You could presumably just like you know drive the car right off the road or into another car by hacking into it.
[00:49:28] And so there are a lot of unsecure technologies that we’re bringing into our homes and in our lives and that are going to be flying above us that are not very secure. And so this is going to be a big issue coming up. This is something you mentioned before what we before we started recording that you’re going to be starting to pay really close attention to yourself too.
[00:49:50] APRIL: Yeah I think with particularly with drones it’s it’s going to matter a lot because you know unlike. No other you know smart devices like my toaster, my fridge which certainly I don’t want hacked. A drone can fall from the sky and really hurt somebody. And also because this is an industry that has a like I said I’m fascinated by this industry it has a lot of potential to grow these snafus at this point, in the industry. It’s such a nascent space really matter and can really affect things.
[00:50:22] Also people are afraid of drones in many ways right. And I think privacy is probably the biggest thing people are afraid of. We did a Twitter poll where something like a little over a 1,000 maybe 11,000 people participated, asking you what’s their biggest fear with drones, and privacy was the biggest one that they were concerned about, falling from the sky wasn’t.
[00:50:42] IAN: Wow.
[00:50:43] APRIL: Well certainly if more drones fall from the sky that’s going to be an issue.
[00:50:47] IAN: True, true.
[00:50:47] APRIL: So definitely keeping an eye on this. That said the things that drone makers can do to ensure the security of of these devices isn’t that complicated. You know they could simply encrypt the links you know between the operator and the aircraft and lock the Wi-Fi down on the drone. Do these kind of basic things. Just one example. I’ll give it I don’t know if you remember this I think it was in November of last year, the New York Times website went down, Recode’s website went down. Twitter went down and it was because of a huge botnet that was created from hundreds of thousands of IoT devices in people’s houses.
[00:51:29] One hacker was able to, because they’re all connected to the Internet kind of weaponize the hundreds of thousands of mostly security cameras from this one company that uses this chip maker or this particular chip from China. The kind of the ubiquity of this particular chipset or this component set allowed for the hacker to create a huge botnet from all of these devices in people’s homes or around the world. Actually send a DDoS attack which is when they just send so many requests to overload the servers of this particular hosting company that serves the websites for these really large companies like Twitter like Spotify went down, New York Times went down.
[00:52:16] So you know when it comes to security of Internet connected devices whether they’re flying or whether we just have them plugged into the wall that’s going to be an increasingly important thing to look at.
[00:52:28] IAN: Wow, we’re in for some potential dark times if they get a hold of our smart light bulbs as well.
[00:52:35] APRIL: Yeah and I don’t want to scare anyone, I mean really like I said these fixes particularly with drones are relatively easy to do. And it’s just something that drone makers need to be aware of is safety and security. So I encourage and DJI has taken this very seriously. I’m sure you know other drone makers have. I want to say Parrot has but then obviously they’re one of the drones that the FDC was able to hack and make fall from the sky. So it’s just something that as more people come into the market and more drone makers emerge that I hope that they’re thinking about.
[00:53:14] IAN: Cool well, April Glaser, thank you so much for joining us. That was a marathon session, we covered so much in such a short amount of time. Thank you so much for joining us. Once again. April Glaser covers robots, drones, artificial intelligence, and other smart machines for Redcode. You can go ahead and check out some of April stories at Recode.net and follow her on Twitter at @aprilaser.
[00:53:45] This is her beat ladies and gentlemen she covers drones. So send her some tips as well if you got them. She’s going to be continually reporting it behind the scenes figuring out really interesting things for us. So April we really appreciate it. Hope we can have you back soon. Talk more about this drone hacking stuff that we’re going to be seeing inevitably. And a pleasure to have you.
[00:54:12] APREIL: Thanks so much. Yeah I look for a day hearing more of these podcasts I’ve been enjoying them so far.
[00:54:17] IAN: Thank you for listening. So if you want to you can follow the podcast @dronespodcast on Twitter, on Facebook.com/drones podcast.
[00:54:26] And if you like what you heard subscribe on iTunes, Google Play whatever you listen to podcasts on and check out the website, commercialdrones.fm, we’ve got a Patreon account (patreon.com/dronespodcast) if you’re interested in supporting the podcast as well.
[00:54:39] So we’re going to now cut off the mic. Everybody take care and watch out for your smart devices. Cheers.
[00:54:57] All right so that’s a wrap. I really hope you enjoyed this episode. One thing you can do to support the show is to visit our new Patreon page and consider donating some change at patreon.com/dronespodcast. Go ahead and donate a buck if you’re feeling it. I will be forever grateful. Over and out.