Ex-Tesla Insider Tells ALL About Robotaxis & Robots

Ex-Tesla Insider (AI LEAD) Tells ALL About Robotaxis & Robots: Andrej Karpathy On Optimus, FSD & Autonomy you guys remember Andrea Kathy right spent half a decade at Tesla as a director of AI leading their computer vision team on autopilot I can tell you a couple of things about Andre one his brain is definitely a lot bigger than yours and mine combined he's a brilliant mind one of the AI goats and he sat down with the folks on the no PRI podcast to share some thoughts so I thought this could be relevant now keep in mind Andre no longer works a Tesla so you're going to hear the unvarnished truth from an independent mind Andre wrapped up his time at Tesla 2022 so it's been a couple of years had a brief at open Ai and now I believe he's doing his own thing with AI and education over to you Andre from a demo to an actual globalization of it I think there's a big gap there that's how it's related I would say to AGI because I suspect similar it will look in a similar way for AGI when we sort of get it and then staying for a minute in the Sal driving space uh I think people think that wayo is ahead of Tesla I think personally Tesla is ahead of wayo and I know it doesn't look like that but I'm still very uh bullish on Tesla and self-driving program trigger alert Andre just said something that's true actually two things that are true Tesla is way ahead of wh and it doesn't look like that which is why I thinking is important and a bit of an understanding by the way please don't send this clip to any of the money managers who just say oh Andre Andre doesn't know anything about Ai and self-driving he he wasn't the head of self-driving it for 5 years ignore ignore but this is actually important because he is completely correct I said this a trillion times maybe a quadrillion by now companies like Cruz Ando have the false perception appearing to be aead of Tesla because they are operating their impressive party tricks in small pockets in cforia maybe Phoenix Arizona Austin that's about it but the way they're achieving this result with the amount of handh holding and guard rails necessary is completely different to what Tesla is doing with autonomy developing generalized solution Andre appears to get this if I didn't know any better I might think that he perhaps knows what the is going on I think that Tesla has a software problem and I think wayo has a hardware problem is the way I put it and I think software problems are much easier Tesla has deployment of all these cars on earth uh like at scale and I think uh way more needs to get there and so uh the moment Tesla sort of like gets to the point where they can actually deploy this and it actually works I think it's going to be you know really incredible uh the latest builds I just drove yesterday I mean it's just driving me all over the place now they've made like really good improvements uh I would say very recently yeah I've been using it a lot recently and it actually works quite well it was it did some mirculous uh driving for me yesterday so I'm very impressed what the team is doing and so I still think that Tesla mostly has a software problem wayo mostly horor problem and so I think Tesla wayo looks like it's winning kind of right now but I think when we look in 10 years and who's actually at scale and where most of the revenue is coming from I still think they're they're ahead in that sense I couldn't agree more as as I've discussed in terms of scale Andre importantly points out Tesla's already got many millions I mean it won't be too long before they've got 10 million vehicles on roads capable of autonomy with an over thee software update 10 years from now that Fleet could be 100 plus million all capable of autonomy does anyone have any ideas please please explain in the comments below how any company other than Tesla would be able to match a fleet of that scale capable of autonomy and also how much would it cost to do that by the way the actual answer many trillions of dollars of hardware on roads importantly Andre mentioning that he believes not just scale but he's talking about Tesla capturing a massive slice of all of the revenue oh and spoiler alert the larger a company's share of autonomous Revenue in the future the larger their profits and I don't just mean that they're going to be in proportion the company with a greater scale is going to benefit the most and I personally believe take the Lion's Share of the profits on the Lion's Share of the revenue more profit per mile on average than second third and 29th Place of course we do have to remember that Andre is really dumb as a small brain plus he has no understanding of autonomy and AI so he's probably going to be wrong and I'm definitely going to be wrong so ignore what we have to say oh and then unrelated matters just for the record still buying Tesla stock with every spare scent still hav soldy just saying how far away do you think we are from the software problem turning the corner in terms of getting to some equivalency because obviously to your point if you look at a way car it has a lot of very expensive lar and other sort of sensors built into the car so it can do what it does it sort of helps support the software system and so if you can just use cameras which is the Tesla approach then you effectively get rid of enormous cost SL complexity and you can do it in in many different types of cars when do you think that transition happens I mean mean in the next few years I'm me I'm hoping like something like that but actually what's really interesting about that is I'm not sure that people are appreciating that Tesla actually does use a lot of expensive sensors they just do it at training time so there are a bunch of cars that drive around with Lars they do a bunch of stuff that like doesn't scale and they have extra sensors Etc and they do mapping and all all the stuff you're doing it at training time and then you're distilling that into a test time package that has deployed to the cars and is Vision only and it's like an Arbitrage on like sensors uh and like expense yeah and so I think it's actually kind of a brilliant strategy that I don't think is fully appreciated and I think it's going to work out well because the pixels have the information and I think uh the network will be capable of doing that and yes at training time I think these sensors are really useful but I don't think they're as useful at test time and I think you don't it seems like the one other thing or transition that's happened is basically a move from a lot of um sort of uh Edge case designed turistic associated with it versus end to end deep learning and that's what other shift that's happened recently do you want to talk a little bit about that and sort of what that yeah I think that was always like the plan from the start I would say at Tesla as I was talking about how the neural nut can like eat through the stack because when I joined there was a ton of C++ code and now there's much much less C++ code uh in the test time package that runs in the car because uh there's still a ton of stuff in the in the back end uh that we're not talking about the neural net kind of like takes uh takes through the system so first it just does like a detection on the Image level then it does multiple images gives you prediction then multiple images over time give you prediction and you're discarding C++ code and eventually you're just giving steering commands and so I think Tesla is kind of eating through the stack my understanding is that current Whos are actually like not that but that they've tried but they ended up like not doing that is my current understanding but I'm not sure because they don't talk about it but I do fundamentally believe in in this approach um and I think um that's the last piece to fall if if you want to think about it that way and I do suspect that the entn systems for Tesla in like say 10 years it is just a neural nut I mean the the videos stream into a neural nut and commands come out you have to sort of build to build build up to it incrementally and uh do it piece by piece and even all the intermediate predictions and all the things that we've done I don't think they've actually like misled the development I think they're part of it uh because um there's a lot of solid reasons for this so so actually like Inn driving when you're just ating humans and so on you have very few bits of supervision to train a massive neural nut and it's too too few bits of signal to train so many billions of parameters and so these intermediate representations and so on help you develop the features and the detectors for everything and then it makes a much easier problem for the um end to end part of it and so I suspect although I don't know because I'm not part of the team but there's a ton of pre-training happening so that you can do the fine chaining for end to end and so basically I feel like it was necessary to eat through it incrementally and that's what Tesla has done I think is the right approach it looks like it's working so I'm really looking for if you had started end to end you wouldn't have had the data anyway that makes sense yeah so uh you worked on the um Tesla humanoid robot before you left uh I have so many questions but one is like starting here what transfers basically everything transfers and I don't oh he just said the thing that's completely true that I also was saying that people didn't quite seem to understand maybe they do now this is really important I'm going to let Andre speak just a sec remember I said you can essentially transplant the brain of the car the vehicle into the robot the vision seeing perceiving planning and acting in the real world locomoting like moving it's the same slightly different format this is why I personally believe that Tesla will be untouchable when it comes to useful generalized humanoid robots as well because in order for another company to have competency comparable to Tesla's humanoid robot they're going to need to Sol Vision at scale and how in the are you going to do that unless you have literally millions of sensor encrusted robots in the world whether they're robots or cars it's the same the cars are robots how would you do this without the millions of robots in the world collecting data and training and the answer is I don't think you can so many companies out there as I've mentioned recently and repeatedly may be developing humanoid robots they may do fine with the hardware the hands actuation the movement the Lo that stuff no problem whatever but the most important thing is the brain and particularly Vision how the are they going to solve that without data at scale like Tesla has and Tesla's ability to literally transplant the brain of the vehicle into the robot I mean I just don't see how this happens and of course the haters will say I just sound like a deluded Tesla fan boy oh Tesla's going to win at everything you're a please explain to me how somebody develops a comparable humanoid robot in terms of its capabilities around Vision Without developing the brain let's listen again to what Andre said he's completely right so uh you worked on the um Tesla humanoid robot before you left uh I have so many questions but one is like starting here what transfers basically everything transf first and I don't think people appreciate it okay um that's a big claim are basically robots when you actually look at it um cars are robots and Tesla I don't think is a car company I think this is misleading this is a robotics company robotics ET Scale Company because I would say ET scale is also like a whole separate variable they're not building a single thing they're building the machine that builds the thing which is a whole separate thing um and so I think robotics at scale um company uh is what Tesla is and I think uh in terms of the transfer from Cars to to humanoids it was not that much work at all and in fact like the early versions of Optimus um the the robot it thought it was a car like because it had the exact same computer it had the exact same cameras it was really funny because we were running the car networks on the robot but it's walking around the office and so on and like it's trying to like recognize drivable space but it's all just walking space now I suppose but it actually kind of like generalized a little bit and there's some fine tuning necessary and so on but it thought it was driving but it's actually like moving through an environment is a reasonable way to think of this is like actually it's robot many things transfer but you're just missing for example actuation and action data yeah you definitely miss some components but and the other part I would say is like like so much transfers like the speed with which Optimus was started I think to me was very impressive because the moment Elon said we're doing this uh just people just showed up with all the right tools and all the stuff just showed up so quickly and all these CAD models and all the supply chain stuff and I just felt like wow there's so much in-house expertise for building robotics at Tesla and like it's all the same tools and they're just like okay they're being reconfigured from a car like a Transformer the movie they're just being reconfigured and reshuffled but it's like the same thing and you need all the same components you need to think about all the same kinds of stuff both on the hardware side on the scale stuff and also on the brains and so for the brains there was also a ton of transfer not just of the specific networks but also all of the approach and the labeling team and how it all coordinates and the approaches people are taking I just think there's a ton of transfer what do you think of the First Once Again monre out here spitting facts and again this is really important to understand I've been Haring on about the autonomous opportunity since the started of this Channel and then we hear about the humanoid robot a couple of years later and I immediately bust 12 nuts consecutively because I'm like oh they can put the brain from the car holy Andre's point about the transfer it's not just the brain from the car but also as he points out you got the computer right the chip the batteries electric motors there's so much transfer Tesla has expertise at manufacturing products like this at scale much like it's obvious to me the Tesla's one autonomy generalized autonomy at scale check back like Andre mentions in a decade and let's just see what's happen happened my prediction Tesla is going to be taking the Lion Share of the profits the same fundamental Advantage Tesla has with autonomy huge scale huge data an unable data lead miraculously that can be transferred to provide an equivalent lead with a humanoid robot Tesla has all of the pieces of the puzzle to nail useful increasingly intelligent humanoid robots at scale and there are no other companies that I can think of that have all of the necessary pieces of the puzzle and how many companies can you guys think of please let me know in the comments below that have have expertise in manufacturing millions of pieces of Hardware per year today that happen to have electric motors and lithium ion batteries and an fsds computer and cameras for vision oh wait there's there's none H I'm sure it's nothing first application areas for humanoid robotics or human form stuff I think a lot of people have this vision of it like doing your laundry Etc I think that will come late I don't think b2c should be the right start point because I don't think we can have a robot like Crush grandma is how I it sort of I think it's like too much legal liability it's just like I don't think hug I mean it's just going to fall over or something like that you know like these things are not perfect yet and they require some amount of work so I think the best customer is yourself first and I think probably Tesla's going to do this uh I'm very bullish on Tesla if people can tell the first customer is yourself and you incubate it in the factory and so on doing maybe a lot of material handling Etc this way you don't have to create contracts working with third parties it's all really heavy there's lawyers involved like Etc you incubate it then you go I think B2B second uh and you go to other companies that have massive warehouses we can do Material Handling we're going to do all this stuff contracts get drafted up fences get put around all this kind of stuff and then once you incubate in multiple companies I think that's when you start to go into the b2c applications I do think we'll see b2c uh robots also um like unit tree and so on are starting to come out with robots that I really want I got one you did yeah okay yeah the G1 yeah Andre who just confessed he's still extremely bullish on Tesla as you can tell makes another very important point one I've also made turns out that obvious things are obvious Tesla will be their own first customer with the Optimus humanoid robot and this is important he talks about the grandma problem and the robot falling over I've also made similar statements myself because Tesla literally makes millions of products per year in factories and parts of those processes are extremely simple boring and repetitive Tesla can begin deploying Optimus humanoid robots in their own factories and at first they will be a drag on productivity they will be annoying they will make mistakes they'll fall over they'll drop they'll put things in the wrong place they'll be less than useless however because Tesla will be their own first customer Tesla isn't going to get B and you this is I want to refund instead Tesla's going to go great let's make it better and eventually because Tesla can deploy their robots in their own factories scratching their own itch serving their own needs once they become useful enough in Factory like situations that will transfer to other companies who have comparable needs I the first commercial customers for Tesla's humanoid robot will most likely be other Automotive manufacturers however it's possible that it'll be more generalized Factory stuff but then soon after that they'll be deployed in warehouses as well big commercial orders for warehouse It's not that different from a factory then suddenly they're Off to the Races now again I just want to ask the question CU it's important to understand are there other companies that could potentially do what Tesla's doing Let's ignore the lack of the brain and all that stuff I mean Tesla's already won for like six ways till Sunday but on this particular example how would another company developing humanoid robots generalized humanoid robots go through this process of training testing validating and increasing capabilities at scale which is necessary before they start selling to customers and the answer to that question is uh I mean I guess they would just have to build a bunch of fake factories doing fake jobs just to test them out and train them I mean what what's the alternative the only other way I see this happening is if a humanoid robot company gets a massive investment let's just say that Amazon starts swinging around billions of dollars and investing buying large stakes in said companies and then deploying the robots in their own factories SL warehouses understanding that they'll be extremely useless and a drag on productivity initially that's the only way I see that happening it's just so obvious that Tesla has won this humanoid robot opportunity they have all of the pieces of the puzzle no other company does and to acquire those pieces is near impossible they a't getting the brain they don't have the data they don't have the expertise manufacturing at scale and they can't be their own first customer so how the are they going to win so I will probably buy one of those and there's probably going to be an ecosystem of people building on those platforms too but I think in terms of like what like wins at scale I would expect that kind of a approach I but in the beginning it's a lot of material handling and then going towards more and more HKC things that are more specific one one that I'm really um excited about is the net freedan challenge of the leaf blower yeah uh like I would love for an optimist to walk down the street like tiptoe down the street and like pick up individual leaves so that we don't need like leaf blowers and I think this will work and it's an amazing task and so I would hope that that's one of the first applications just even raking yeah that's should work too just very quietly yeah quiet rcking that's cute they uh I mean they do actually have like a machine that's working it's just not a humanoid can we talk about the humanoid thesis for a second because the the simplest version of this is like the world is built for humans and you build one set of Hardware the right thing to do is build a model that can do an increasing set of tasks in this set of Hardware I think there's a like another camp that believes like well like humans are not optimal for any given task right you can make them stronger or bigger or smaller or whatever and why shouldn't we superum things like so I mean obvious things are obvious here the entire world today we've designed it for wait for it ready ourselves humans so a humanoid form factor will slot most easily into the current world the reason a company would develop a humanoid form robot is if they specifically want it to be quick easy convenient to slot into the real world to do tasks that today humans do at massive scale you can certainly make the argument having super specialized robots will be more efficient and effective for very particular narrow tasks but if your goal is massive massive scale the only option is a humanoid form factor it be a lot easier to build a ton of humanoid robots than it will be to re-engineer everything on Earth specifically to be more suitable for slightly different or extremely differently designed form factors for robots but I mean you already knew that right how do you think about this I think people are Maybe sing Le platform uh I think there's a large fixed cost you're paying for any single platform and so I think it makes a lot of sense to centralize that and have a single platform that can do all the things I would say the human aspect is also very appealing because people can teleoperate it very easily and so it's a data collection thing that is extremely helpful uh because people will be able to obviously very easily to operate it I think that's usually overlooked there's of course the aspect you mentioned which is like the world design for humans Etc so I think that's also important I mean I think we'll have some variations on the humanoid uh platform but I think there is a large fixed cost to any platform and then I would say also one last dimension of it is you benefit a ton from like the transfer learning between the different tasks and in AI you really want the single neuron not that is multitasking doing lots of things that's where you're getting all the intelligence and the capability from and that's also H that's also why language models are so interesting is because you have a single uh regime like a text um domain multitasking all these different problems and they're all sharing knowledge between each other and it's all coupled in a single neural nut and I think you want that kind of a platform and uh you know you want all the data you collect for leak picking to benefit all the other tasks um if you're building a special purpose thing for any one thing you're not going to benefit from a lot of the transferring between all the other tasks if that makes once again the due is right and by the way this also completely directly relates to generalized autonomous vehicles as opposed to the extremely brittle solution that cruise and weo and other companies are attempting to come up with I give an example in the past Tesla's approach with autonomy is to solve vision and solve autonomy such that you could in essence pick up a Tesla vehicle and drop it on any road on Earth and without having previously driven on the road without you having to previously map the area in high definition without you needing to use liar to bounce senses off to make sure that everything's in line with the pre 3D map the vehicle because it has generalized ability to drive will look out into the world and go oh yeah that's a road that's a line I can drive there I can't drive there obviously this is the side of the road I need to be on okay cool I can figure this out no problem and it can drive safely anywhere on Earth because that General capability transfers companies like Cruz and wh on the other hand you try that with one of their vehicles Take It Outside of it little safe sand pit and it becomes a death machine on Wheels this is all about scale and I suspect that we're going to see the same narrative and the same misconceptions about what's happening with humanoid robots in the next couple of years as we currently see with autonomy where as Andre points out people think Cruz wh in particular are ahead of Tesla with autonomy it's a false perception the same is likely to be true for humanoid robots we'll see demonstrations from some of these robot startups where their robot will be doing one very specific narrow task exceptionally well but it will be completely miserable and or incapable of doing anything else take it out of its sand pit that one special task and suddenly useless I mean this Andre guy seems to know what he's talking about fancy that Mak sense yeah I think there's one um argument of like uh it seems I mean the G1 is like 30 grand right but it seems hard to build a very capable humanoid robot under a certain bomb and like if you wanted to you know put an arm on Wheels I can do things like it's a like maybe there are cheaper approaches to a general platform at the beginning does that make sense to you uh cheaper approaches to a general platform from a hardware perspective yeah I think that makes sense yeah you put a wheel on it instead of a feet Etc I do feel like I wonder if it's taking it down like a local minimum a little bit I just feel like pick a platform make it perfect is like the long-term uh pretty good bet and then the other thing of course is like I just think it will be kind of like familiar to people and I think people will understand that maybe you want to talk to it and I feel like the co the psychological aspect also of it I think favors possibly the human platform unless people are like scared of it and would actually prefer a platform that is more abstract of like some but then I don't know if if just some some like monster doing stuff then I don't know if that's like more kind like it's interesting that I think that the other um form factor for the unitri is a dog right and it's almost a more friendlier familiar yeah but then people watch Black Mirror and suddenly the dog flips to like a scary thing like so it's hard to think through uh I just think psychologically it will be easy for people to understand what's happening and what do you think is missing in terms of um technological milestones for Progress relative to substantiating this future uh for robotics robotic yeah or the the humanoid robot or anything else human form yeah I don't know that I have like a really good window into it I do think that it is kind of interesting that like in a human for factor for for example for the lower body uh I don't know that you want to do imitation learning from like demonstration because for lower body it's all a lot of like inverted pendulum control and stuff like that it's for the upper body that you need a lot of like to operation and uh data collection and end to end and Etc and so I think like everything becomes like very hybrid in that sense and I don't know how those systems interact when I when I talk to people working the a lot of what they focus on is like actuation and you know manipulation and sort of digital manipulation and things yeah I do expect in the beginning it's a lot of like teleoperation for uh getting stuff off the ground and imitating it and getting something that works 95% of the time and then talking about human to robot ratios and gradually having people who are supervisors of robots instead of doing the task directly and all this kind of stuff is going to happen over time and pretty gradually I don't know that there's like any individual impediments that I'm like really familiar with I just think it's a lot of gr grun work a lot of like the tools are available Transformers are this beautiful like blob of tissue you can just get just just arbitrary tasks and you just need the data you need to put it in the right form you need to train it you need to experiment with it you need to deploy it atate on it just a lot of grunt work I don't know that I have a single individual thing that is like holding us back technically some interesting final thoughts there how to solve this problem with the humanoids in terms of capabilities deploy train collect data iterate improve so be really convenient if you happen to be trying to solve this problem the humanoid robots if you could deploy them across your factories on increasingly more complicated tasks over time to collect the data to train and so on being your own first customer a massive Advantage here so in conclusion Tesla has one autonomy their lead is unassailable doesn't appear like that if you look at the surface level wa way more operating robotex is already they're ahead that's why I got to dig a little bit deeper and the reason they want autonomy is because they've been able to develop the brain of the vehicle due to the fleet scale data which which transfers directly into the humanoid robot meaning guess what yes they've also won the humanoid robot opportunity but it's not likely to appear that way for a few years well at least that's my personal opinion and it appears that hre Kathy who's a very low IQ guy who doesn't know anything about AI or robotics seems to agree in general but of course he's dumb and uninformed so ignore him instead I would probably listen to some of the suggestions the wisdom the reports from managers of money the same kind of folks who previous thought Tesla was just a car company they're probably going to be right but on the odd chance it turns out that my predictions come to pass 10 years from now I'll be happy to look back and say you know what tried to warn these said the same every single day for half a decade and gting I did my best so uh in conclusion don't say I didn't warn you want more Want More Content? Early Access? content Early Access bunch of perks click the links in the pin comment ag1 AG1 is awesome I've been taking it daily now for more than 3 years it's a great way to fill a nutrition gaps it's packed full of vitamins minerals and Whole Food Source nutrients plus has prebiotics probiotics and adaptogens to improve gut health regularity and help your body handle stress I'm always looking for an edge to help me feel and perform my best which is why I haven't missed a day of ag1 for more than 3 years and I haven't missed a daily video in more than 3 years must be a coincidence right just try it and see how you feel click your link in the pin comment or head to drink a1.com smmr and get yourself a free 1year supply of vitamin D3 plus K2 two and five travel packs but don't take my word for it this is what viewers of the channel had to say after trying ag1 I feel like I have a lot more energy since I started on ag1 by the way viewer that makes two of us on to the next just got my ag1 in the mail legit feeling the effects after day three this F has been taking ag1 for 8 months and says what an investment another 3 months ago I started ag1 and have been enjoying the evenness of alertness and energy that lasts the day just started the wife on it too you convinced yet I mean hey it's worth trying right click the link at the bin comment or head to drink a1.com smmr or I can keep going plenty more to come this viewer after about a month on ag1 definitely a lack of fatigue in the afternoon Pleasant side effect is that my coffee intake has imploded and is almost down to zero one more yeah why not I honestly feel younger and will be continuing to use ag1 this stuff really is crazy good I didn't think it would be but this stuff is awesome it really is what everyone is saying one more don't mind if I do I've just received my third month supply I've drank it every day I have so much energy throughout most of the day I'm productive started a new business started socializing refurbished the boat it's no coincidence thank you for your persistence your integrity and your insights now look these are not my words these are not my testimonials this is what you guys and girls are saying maybe it's 100% placebo effect but even if that's the case I think it's money well spent click your link at the pin comment or head to drink a1.com ssmr and get yourself a free 1year supply of vitamin D3 plus K2 and five travel packs if you're still skeptical hey I don't blame you everyone on the planet seems to be promoting ag1 now but guess what they weren't nearly 3 years ago when I had this to say privately to my patreon audience before there was a relationship when I was asked about what I was doing for my health energy and so on just sharing my genuine honest thoughts about a product I'd recently discovered that was at the time called athletic greens now ag1 if I could only recommend one supplement to take athletic greens and I'm not getting paid to say this athletic greens is a game changer I just I cannot believe how itive this is no longer having a lack of energy in the afternoons it's amazing there's only one thing to recommend seriously try athletic greens you won't go back so obviously just like Elon Musk is a liar fraud a con man a scammer a fake engineer and T is going bankrupt you shouldn't trust that guy from about 3 years ago who without any Financial incentive was promoting this product to his audience on patreon when they're asking about health and what he's doing for supplements cuz obviously there was some other reason he recommended that obviously not sure what it was but don't trust that guy and all the testimonials like my mental game has improved with a one I feel better than ever I'm so impressed I've bought it for both my parents I feel more focused and have better digestion incredible difference no more afternoon fatigue it's relieved gut issues these are all just obviously fake testimonials from fake people right wrong just try it unless you hate yourself if you hate yourself and you don't even want to risk possibly feeling better this is not for you but for everyone else what's the worst Could Happen try for a month see how you feel it's a no-brainer just click the link in the pin comment or head to drink a1.com smmr you'll get a free 1e Supply vitamin D3 and K2 plus 5 travel packs and you'll take the Colossal risk that maybe you might have a similar experience to some of the people whose testimonials we've read in this video