MCA7Ā·43:46
Designers Code, Robots Connect MCA7
Comments
Loading polls...
āWe're seeing designers ship production code faster than our senior engineers can review itā
āThe distinction between technical and non-technical roles is becoming meaningless - it's like the early computing era when individual developers could build complete products end-to-endā
āI don't need to pay Salesforce tens of thousands and hire coders - I just tell Claude what I want and boom, it builds my custom CRMā
Listen to this episode
Show Notes
Bill, Gus, Jim discuss AI Turns Designers Into Developers, Nvidia Buys Groq for $20B, Cisco Zero-Day Enterprise Attacks, Tesla Optimus Neuralink Integration. This episode of MorpheusCyber.com covers the latest in tech, AI, cybersecurity, and emerging technologies. Episode MCA7.
šÆ Interactive Slides + Polls: https://ahaslides.com/MCA7
Topics Covered:
⢠AI Turns Designers Into Developers
- Cursor AI editor hits 100,000+ active developers in 6 months
- Replit Agent launches autonomous app development
⢠Nvidia Buys Groq for $20B
- Groq raises $640M Series D at $2.8B valuation
- Groq achieves 500+ tokens per second inference speeds
⢠Cisco Zero-Day Enterprise Attacks
- Cisco releases emergency patches for actively exploited zero-days
- WatchGuard confirms related vulnerability exploitation
⢠Tesla Optimus Neuralink Integration
- First Neuralink patient successfully controls computer cursor with 90%+ accuracy
- Tesla reveals Optimus can now perform 15+ household tasks autonomously
Key Statistics:
⢠AI-assisted developers complete coding tasks 60% faster than traditional methods (Source: GitHub Copilot productivity study, 2024)
⢠Non-technical professionals using AI tools can build functional apps 10x faster than learning to code traditionally (Source: Replit user study analysis)
⢠75% of software vulnerabilities in AI-generated code go undetected by non-technical users (Source: Synk security research, December 2024)
Listen:
Apple Podcasts: https://podcasts.apple.com/us/podcast/morpheuscyber-com/id1837524283
Spotify: https://open.spotify.com/show/58pyOj6GHI9ipF6Ou46vi6
Website: https://morpheuscyber.com
RSS: https://feeds.captivate.fm/morpheuscyber/
@:00 Introduction
04:32 AI Turns Designers Into Developers
18:47 Nvidia Buys Groq for $20B
31:15 Cisco Zero-Day Enterprise Attacks
42:38 Tesla Optimus Neuralink Integration
53:21 Wrap-up
TECH FUTURES INDEX
TechFuturesIndex.com tracks public and private tech companies alongside sovereign wealth funds compared to US debt per capita. Our Baby Franchise program creates investment accounts for newborns - recent additions include the Dell family contributing $6.25B. Building generational wealth through technology.
https://techfuturesindex.com
Fact Slides (5)
View Full Transcript
Welcome to the Morpheus Cyber Podcast. Today, designers code, robots connect. When graphic designers become overnight developers and $20 billion buys you the future of AI chips, you know the world's changing fast. Grab your phone and vote in our live polls. a-haw-slides.com slash MCA7. AI code editors are turning your Figma Obsessed designer into the next full stack developer. No computer science degree required. Nvidia just dropped 20 billion to buy their biggest rival growth because apparently crushing competition is cheaper than competing. hackers are using putty, yes boring old putty, to exploit Cisco zero days and steal enterprise secrets in plain sight. Tesla's robots might soon read your thoughts via a neuralink because apparently walking and talking wasn't creepy enough. Vote along with us at a-haw-slides.com slash MCA7. Let's dive in. Hey everybody it's recording day. Here we are. How's everybody doing? Anything new and exciting Jim? Gus? No. Got a cold. I got that. It survives all the days and Dylan California. I'm deeply in debt. Spent a lot of money on Christmas. Yeah that that Instapod got you didn't it? Yeah, with tip of the iceberg. Yeah. Hey we've got some neat topics today and I'm excited every day I wake up and I go find our new topics and it's just like oh my god it's just super cool. There's all sorts of new things about designers into developers and just a whole bunch of stuff. What do you guys want to talk about first? I think we do stick with that first whole topic about AI's now turning designers into developers. You know that's that's cool. I mean it's like in the past it's always been that developers for maybe over the past 15 years were the software creation had to be sliced into all these narrow specialties. There was a front end guy. There was a back end guy. There was a dev ops and each one had their own silo and then UX was somewhere out there in the ether. Everybody was doing their own thing and all that did is fragmented everything and it created an entire economy around complexity and that's man. Bill and I both been founders and worked with developers and we may have ideas but we had to run that through the developers before anything got turned in any kind of code. It was like pulling teeth and ripping your hair out because it's like I can't do anything about that. I got to wait on these guys to do this but reality now that AI is just completely changing that and turning people with design ideas into real developers and Bill you've been doing a lot. You've taken the lead in our group of learning how to use some of these AI development tools and gotten me inspired. Now I really want to do it because you've shown us how you've created something really cool and people don't know but our own podcast with Peter own dog food. We are creating an application that helps us do our podcast and Bill has been the front lead doing everything on this, spending hundreds and hundreds of hours but man it's just amazing things are turning out. So just go ahead and tell them a little bit about the tools that you've been using like cursor and the rest. Yeah I do use cursor but that's not my primary tool anymore. I now just use mainly cloud code and opus 4.5 which when I first started using it about six, eight, eight months ago I would have to go through and find all sorts of errors. I do some vibe coding. I tell it what I wanted to do. I'd get it all organized and it would write some code and then I would have to go out and look at the code and click on everything to make sure there were no errors. Well guess what? Don't have to do that anymore because into my VM that generates all of this stuff I just simply say hey install puppeteer or playwright and after I make a change go out and use the chrome browser headless go look at all the pages that we just touched or configured and let me know if there's any errors on them. Boom, Shaka Laka. I don't even have my own AI agent that actually goes and does that and finds the problems before I have to go. I used to I'd have to say hey coders let's go do this let's go do that and then they do it and then they'd say well but there might be some bugs in there and of course there's dozens or even hundreds of clickable locations and you'd have to either hire somebody to go click on every eventuality of every click and every function and every entry but now you just use playwright or puppet here those sort of tools and it goes out and it analyzes I don't have to go click around the user interface to find errors it does it as part of the AI. You've got to arrive now having used these tools for a while but might help be helpful to the people that are listening to us to create just tell us how you went through a progression you did go through cursor first. Yeah. Before you've got to where you are now you used a lot of tools over time and you've now let those drop off because you don't need them anymore you've become more of a at a I would say you're becoming an expert level AI developer that's you're not your average guy anymore you've been doing this for about a year. But used to be Jim that I'd go out and find a software as a service and then I would go in and have to learn for hundreds about how to like Salesforce or you name it whatever software as a service I built our own CRM and I built it to have all the functions that we want and I gave Claude code the English instructions on what I wanted to do and what I wanted it to do and then it would come back and ask me questions and then I would fill in the holes and it would develop instead of getting Salesforce which I've done I've paid through the nose tens of thousands of dollars for Salesforce and then you have to have near a full-time coder to be able to customize that monster for your particular company or application or purpose and it's like it's dawned on me it's like I don't need to pay Salesforce for the development environment and then hire people to go do it I can just tell the code what I want to accomplish how I want it to look where I want it to be on the screen and boom do you remember the movie office space you remember yes of course do you remember the guy who when they were the consultants Bob and somebody else asked is it that you do around here when he says he says I'm liaison to the engineers and it's like I have people skills doing it remember that routine that's how I've used to feel because I've had groups of coders that I worked with and I'd say look we need I was a subject matter expert and I said we need to have this and we need to look at that we need to do this capture and when we see the server response time go up we need to click and create an event that captures a packet trace and then tells us hey you need to go look at this area of this packet trace because at 3 30 in the morning the thing went catatonic yeah I used to have to explain that to the coders and they would like first vet it because maybe I didn't know what I was talking about despite 30 plus years of packet analysis I didn't know what I was talking about they had to verify I have to convince them of what I believe is relevant of course we're not even talking about the product manager at this point because then you have to go through them as well because the coders are the technical vetters of the solution the the project manager or the product manager make sure that the customers that he knows about or she knows about want those capabilities and what the market will do how many more people are going to buy the product if it has this feature or that feature that's what a product manager let me feed that put this into some real life here real quick that we can get to the point on a couple things here that still have to be made one my Uber driver the other day when I went to the airport told me that he lost his job as a QA tester I said hey look you AIs the future buddy you're going to have to learn some AI apps you're going to have to learn how to do your job in AI if you do know how to do QA work you need to find out what apps being used to do that make sure it's AI enabled be good at it then walk in with your potential saying look I know this app this is one of the QA the AI apps that I know I'm ready to take that job back again that's what it's going to take he said really I didn't know that I said yeah buddy just get in there start doing it it's cheap if you got to pay $25 a month to buy the AI platform to to code in a little bit just do it learn it get out there you'll get your job back because he was just living off of his Uber driving I said dude there's a future for you don't don't lose hope that was a good thing and I want that's what we're saying to everybody else out there look I know you probably lost your job for some reason in AI but you can get it back just go back and that's going to happen dramatically in the next 12 to 14 months it's just going to be say 14% of the workforce is going to be displaced by AI before the end of the year here's a different angle though that I'm a little concerned about is that now we're going to see all these these applications flooding the internet by people who now can program who didn't know how to program before but the only problem is I put on my CIS S P hat and I say are these going to be secure apps heck no they are not going to be secure apps they are still going to require somebody that does know how to do this to be able to test the app to make sure that it's hardened against hackers and others that would go in and completely take control over their apps and ruin their entire endeavor you can't build a business around an unsecure app they've done a half to get that done because this is just going to create mistakes at scale huge mistakes yes so it was I agree with you Jim but the cool thing is that Bill touched on this but the language is no longer C++ Python whatever it's English or whatever language you speak and that is that builds a bridge and it has Bill alluded to what this allows is for the subject matter expert to be able to put it is that they want to create there's no middleman and I I hear what you're saying about the security and that's no question going to be and hopefully things will evolve to where that's somehow built in but I think it's pretty exciting the AI will do the pen testing for you I'm not kidding you know you just have to code it up and now that doesn't preclude needing security experts to take a look at it and analyze but if you've got a whole bunch of code the code is pretty good at analyzing itself you do need that human everything that I do in my code like for this podcast whether it's a thumbnail generation from all of our content and automatically generate these things or whatever there is a human spot in there that I I create in part of the process it's human QA I can take a look at it before I just boom click a button and it's all up on YouTube and it's all working there's a lot of additional steps and you just put a human in there to do the QA to just make sure that it passes the smell test before it goes out now with our situation we're doing a podcast well we do research vast amounts of research on X on news on podcasts every podcast that's relevant to our topics we go out and get the RSS feeds and on our systems we listen to the audio fully transcribe it and then we tell us here are all the podcasts who had that particular topic and then we choose from news X trends that sort of thing and podcasts and from that we gather all of the topics and then we say which topics are interesting and then we nominate those topics that's what we do every week we nominate them from given all the way and bill this is the number one sub billionaires podcast we could be the last one to keep giving away all our secrets yeah I'd say bill I thought that was proprietary power profile and bill is built a very powerful application let's leave it at that and someday one of you guys are going to want to use this application all right we better move on to another subject here what I wanted to talk about guys was some pretty big news came out Christmas Eve and that is that Nvidia made their largest acquisition ever by far of a company called GROC not to be confused with GROC GROC and what they do for everybody yeah GRO Q Q GROC not to be confused with grass grow grow the reason why this is interesting is that they spent 20 billion dollars buying this they're announcing and if it gets approved they're going to be acquiring GROC and the reason why this is important is that GROC is a new kid on the block and they're doing some really special things we all know that with large language models the training is a big part of it and Nvidia is the market leader in that area but what GROC does is they make these what they're calling LP use learning processing language processing unit and what they're able to do is once the programs trained these LP use will increase the speed in which the AI responds dramatically I mean like they do 500 tokens plus versus 10 to 50 which is what the GPUs do GPUs are very good at training but they're not this inference and this is what brings to the table and it really puts Nvidia in a situation where they control everything you know the ideas that the inference accelerators are dramatic and they in it's responsible for the just the quickness in which AI responds to questions and comments I don't know if it seems like it might be problematic because it gives them an monopoly yet but what do you guys think well you know what what the name of that thing is I don't even remember that yeah yeah there you go that yeah if they control that that's a proprietary component that is their software that runs the stuff over to the GPU and manages GPU tasks yeah yeah I need more experience with that than any of us researching on Nvidia's platforms now they own that which is the primary processing unit that actually does all this processing GPUs and all that but now also they're going to control the LP use if it gets approved if it gets approved well now let me tell you that they are still subject to other competitors and Google and has these new chips these inference chips and they're pretty powerful and guess who's using those bad boys AWS there's a whole bunch of people who are on the bandwagon and then Broadcom is out there who's been one of the chip makers who do all network interface card components and switch components Broadcom is doing all these different things and they are entering into these because they were doing network processors and now these are essentially the same type of component they're just GPU processors for inference there's a healthy amount of competition out there now I use GROC a GROQ and what did I use it for it was when I wanted and I'll just tell you what the application was I wanted to be able to listen to our show in real time and have a screen up that would suggest things that maybe Jim didn't cover or Gus didn't cover before we close out a segment and come up and say hey you forgot to tell everybody about this how do you do that in real time GROC is the only thing that can process tokens quickly and that is there for Tay you use GROC when you need real time information whether you're sitting and listening to a sermon and it's recording on your iPhone when it's going directly to GROC and GROC is listening to it and then it's basically saying oh he made the wrong scripture reference for that problem or whatever or hey here's another one because what do you do when you're sitting and listening in an audience or you're in an interview you're in most interviews these days are yes they do an in-person toward the end but a lot of your interviews are what online you take the feed of all your audio send it to GROC and it comes back and that by the time a an interviewer is interviewing a candidate the interviewer articulates a question and GROC is listening to it and gives the candidate the answer before I'm telling you I'm sorry that is slick this is and I wanted to create this capability that I could listen anywhere just on my iPhone I could just listen to whatever I wanted and then it would basically say Bill you might want to know about this and this and this because when I sit and listen to a lecture I'm always thinking about other applications of it my brain is going in the background and doing the very questions that come up into your head and maybe even before they come up into your head the AI has listened to him and said hey if you're if he's talking about this you should also see the relation to that I'm telling you that's where the world is we're not only looking at systems that for instance there's a lot of talk about systems that just watch our screen we're using our computer all day long doing emails doing whatever and AI can look at what you're doing and suggest shortcuts to be more efficient and do a better job with less effort if it's watching your screen 24 7 it could be watching you watching this podcast it could be listening to the podcast it can be then suggesting and then it knows what it knows your job it knows your title at the company it knows what you're working on it can start suggesting how you could apply things that you learn on the podcast and then put that to start putting that information in and then you can add that to your business things or your personal life and that's why I was using Groke I got real-time analysis of things and that's the only one that can do that real-time analysis is that cool or what yeah you don't see any real issue with they went broke owning Groke what do you think Bill do you think there's any conflict there of interest or does that not gonna stop that at this point I mean take a look at open AI is there anybody going over and pulling out the rug from under them no they were the ones who they were the first mover let's face it and IBM like I said another podcast has been talking AI for years 10 20 years for business this is a merger though this is a merger but yeah but what I'm talking about here is that there has been AI for the last 20 plus years or even longer but the thing is what catalyst was it that allowed AI to start taking over when they put it in the average person's hands and made the average person have the the not the IQ but the ability to be accessing doctoral level information and then acting on it and employing that they didn't have to read 20 books to get that information the AI said hey I've read the book I need you to move over here and do this instead of that because it has a doctorate in about 100 different study areas yeah oh yeah I've used it already with my doctor I don't know and I haven't used it like I wasn't talking about medical I was talking about doctor it I know what you're talking philosophy or whatever yeah I'm sure we're now experts in everything exactly there's another interesting point on this merger or this acquisition and that is that GROKE uses Samsung for their manufacturing and idea has a big problem with the independent on Taiwan it's a huge risk for them and this instantly gives them manufacturing in a friendly environment and and that's the strategic move that I think's gonna benefit that's a good point good point I didn't even thought about that good one I'm sure that's part of their logic and keep talking about security and that's one thing we always want to have everything made in the United States and can you have more control over that more security but a lot of it's just gonna be offshore for now but anyhow the future says that we need to bring as much as we can back on shore now talking about security bill you probably you and I both follow security you heard about that Cisco zero day attack they hit their enterprise applications yeah tell me about it you are the one who knows a little bit more about all that but anyhow apparently the goods I'll go into the blind they are apparently within their their hardware meaning their firewall hardware and other security devices the the bad guys let's just put it that way the bad actors have found a way to get into it and have been in it for some time they've been breached not only them but what was the other one we said watch card firewalls were also breached as well that now to immediately now look at how much of the market that Cisco holds in the firewall market let me do 50 50 percent 60 percent of the market out there in firewalls and networking equipment in general and where the breach is is on their licensing smart and we don't know smart licensing once they get into that they get the keys of the kingdom I mean that's a very vulnerable thing for 50 percent of the network networking components in the world it's Cisco now yeah it will mainly in their a sa and the firewall products but watch card two I don't even know how what it is that watch card is not a subsidiary of Cisco what's the odds of that how do the two of those get breached and not a whole bunch of other what's a related thing it's not direct but the issue is is that once you figure out how to break into a Cisco it kind of gives you the how to break into other firewalls I think that's that's what we're talking about but they're using putty to an SSH to attack your network and basically it's kind of like a burglar wearing a police uniform they blend in perfectly until you look very closely and then like this particular zero day that we're talking about it was like 45 to 60 days after they knew that it was happening it took them for an enterprise network patch deployment it took 45 to 60 days leaving all of that open during that window all those devices were open which is probably way too long you know it on day one and you can't get a patch until day 45 what do you do that's a good case for some technology that that you and I've worked with Jim is that yeah if you we can take in the systems that we've developed you can still run Windows 95 because what we do is we limit how far the packets can go you can still use old technology it's just that bad guys cannot gain access to it because it's too many hops away and that capability hasn't really caught on yet but that's how you can insulate it's not just running old windows 95 it's also running new systems why would for instance now out there in yonder every Oracle database has a default packet lifetime that can go around the world it can go anywhere to anywhere if you shorten that lifetime and say this is only good this is the packet lifetime on packets to and from this Oracle database server is only enough for people within the enterprise to be able to access it and or it restricted but it can't go further than the firewall even if your firewall is wide open for something you can use this technology to limit how far packets travel and essentially you can solve some of these problems that it takes 45 days 60 days to get a patch in if you can figure out a way to allow people who are supposed to get access to that but not allow the entire globe to get to it while it's open and I've thought about this for a long time you've got firewalls out there right now that have a hop count of 255 that means it can go through 255 routers well why would you leave a firewall set that could be broken into on your outside perimeter that can have an SSH session to it and of course your router jockey will say oh well we can do the we keep that no you don't we've been in the thought leadership side of this whole idea of containment and it's not really anymore I think they're starting to find out that's very difficult to stop the bad guy if you're going to give them free access through the network and we've created a system bill and I are a part of and we are getting our word out getting the message out more and more about this needs to be applied in every switch every layer three switch that's out there it should be done now but containment segmentation you're starting to hear those words more and more because they're finding out now that they can't do it logically anymore it has to well they're doing it with VPNs they're making a VPN for every component and every system the problem is that that's pretty far reaching and you have to have that for every user this is a very simple thing that only deals with the device itself and just says this router that's on the internet cannot be logged into more than like three hops away which is would be from the enterprise into that router from their network and it won't accept anything from the Russian Federation or China or Russia whatever and the problem is that this hasn't really caught on very much but it is a very powerful capability it's it's finally starting to and it's really the simple concept of reducing the blast radius I mean you're going to get it you are going to get hit at some point yeah but you must contain it you must bring that blast radius as small as you can that when you do get hit by an attacker that he does not have free access on a trusted device that that you've allowed to go across your network they've taken over this trusted device and now they just move laterally through your network that it cannot be allowed if it takes four to six weeks to get a patch and you know that it's vulnerable yeah why wouldn't you figure out another way to lesson the vulnerability anyway yeah I hope sis goes listening and Bill I think you showed it a little bit short when you say it's simplistic it's basic to the frame the actual packet and the way that you guys do that I think it's ingenious thank you very much Gus the simplicity is sometimes hard for people to understand it's like what you just limit how far packet control people say this technology has never been used before it's not right I'll just tell you that every every bgp router in the world uses this technique to limit who it peers with to the adjacent nodes by limiting the hop count to one don't tell me that it's not used this has been used to keep the entire internet safe the internet routers themselves the bgp routers from peering with and telling other neighbors anything because it has a hop count of one the packet will not traverse routers it stops because the two routers that need to talk to each other can but if that packet got loose it could peer with somebody else and then you're really in trouble it has been used and it is tried and true and and we've been working on that for a long time be aware out there people that there are solutions out simple they may seem are very often the very best ones and those are the this simple solution to control how far packets can travel is critical to security because every other method has failed to this point this is a surefire way to stop a majority I can't stop everything because insider threats are that's a problem you got people somebody who's inside your system that knows how the system works share they're going to be able to break it but in general from an outside attack from anywhere or a lateral movement this is the answer speaking of the next segment tesla optimus is looking at having a neural link integration I can't imagine how how a person who has had a brain injury or some other malfunction in their brain who gets a neural link in and then they keep uping the number of bits per second that you can communicate from a brain out through the neural link but imagine somebody who's you quadriplegic or just can't do anything for themselves they can now speak a lot of times with the neural link implant but think about this instead of speaking to your nurse or somebody who's there to assist you speaking to an optimus robot and saying hey I'd like a glass of orange juice I think that's cool well you can do that you can do that without the neural link the neural link is going to take your thoughts and automatically control and and when I I think that's an interesting subject but who cares this way I look at it it's just like you said you can speak to the robot art you can go tell it go do the dishes verbally that's available right now I think the optimus is huge I think that the neural link is not relevant but what the robot that is a big marketplace and it's going to if you compare it if you look at tesla obviously they're known for cars I mean we all know they're much much more much more than that but that's how they got started there's not a lot of profitability in cars you take a $40,000 car the average net profit is somewhere between one and $2,000 okay that's it with these robots they could probably make a subscription model reoccurring revenue even if the sales even come close to what they're saying it's going to be the profitability is ginormous yes and the application is pretty huge too because I don't know anyone who like last week we talked about the iRobot and having those little vacuum cleaners that make your carpet look really cool and the groom your carpet those are pretty cool of course that company went bankrupt who knows what's going to come out of that but this is an area where they're dealing with the dexterious the dexterity of the robots hands and it's only a matter of time you start out simple and you keep moving and you get better and better and once that code starts working to operate the hands more dexterously once that code is written then they can go learn how to grab this how to throw balls how to do how to catch balls they can sew they can thread a needle that they've arrived to that point already it's it's cool but I don't want us to go far away from the knurling thing I don't dismiss that for the most of you just are listening to us we're not dismissing the value of knurling I mean people paraplegics those that cannot no longer can use their own body to move and do things can now just think of thought and then cause a robot to go do it for you is tremendously great it's amazing that could even do that it might be able to also pick you up pick a person up put them in a car and then the car drives them somewhere they now have full mobility when they may be a paraplegic and they can go and do everything it's just it is astounding what that is doing now think even further beyond that and I had an idea knurling implants what if we have look at we have a crummanal element in our world here some they say they're from they're crazy from birth almost I mean they're they have this what do they say this tendency to want to hurt people all the time anyhow what if what if and I'm just I'm way off here I know but if we had an implant and then anytime they had those urges now the AI said no you're not going to do that that's not what you want to retrain their brain while there are those thoughts and impulses that they have that those people have could be corrected possibly down the road and unfortunately we have to put an implant to do it but I think not even I think Elon Musk said that he in the future would even consider having an implant that would connect him directly to the internet or to information that he didn't have to ever lack for knowledge or information at any time well just think about this as sub billionaire podcasters that are over 50 years old by a little bit we're all we all know that our lifetime is has limits and even though that we are finding ways to live longer to be more healthy that we still know that there is like 120 years maybe or whatever is pretty much the max but or at least to date but imagine if we can have a neural link put in that helps us in you know being informed as we age and to help us in those areas I see that as being a powerful capability not just people who've had brain injuries or whatever I mean look at cochlear implants you see those cochlear implants and it's like it's a revolutionary people could not hear at all and then they can hear which is just absolutely is down yeah you watch them I don't I don't just I don't I don't argue that that neural link is significant for all the reasons you just said but guess what all the reasons you just had had nothing to do with its tie-in to the robot I think we have a perfectly good way to deal with optimists and that's that you can talk to it sure there's probably 0.001 said that people can no longer talk because of whatever there maybe there's an application there but you know keep in mind that when they put these implants in due to scarring they lose their their ability to communicate by 15 percent per year per month what that means is that yeah yeah because of the scar tissue they're going to have to replace these things twice a year is what I figured roughly seven eight months just doing the math and you know I think just for very I just I don't think it's worth to bridge the gap to be able to telepathically tell a robot to go do the dishes when you can say hey go do the dishes I know a little bit of security risks along with this because neural link interfaces create an unprecedented attack surface hackers could literally control human operated robots and medical device regulations weren't designed for wireless brain implants controlling autonomous robots these are you know new security problems to deal with and that's that's just intrinsic with as we develop AI as we do anything it can be used in properly we're just going to have to look at managing those security risks as we go I didn't know that there was a loss at over time of that connector that would make sense that would even be that wouldn't even be good for like down the road where Elon Musk said that he would consider having one unless there was a way that that was not the case that you did well I don't know but they were not persons brain but it impacts the quality of the neural communication it doesn't do damage to the brain it just it hurts the communication between it loses its ability to broadcast it's kind of like a pacemaker which stimulates electrically the heart to beat properly it has to be placed at the right spot to be able to provide that when you go into the brain you have to put the sensors which are very sensitive while they're learning and mapping the brain they're mapping the brain for the ability to accomplish that and they're getting better and better at it we've only been at this for a few years I mean seriously yeah but I do see the optimist is all positive upside having a robot my my own sister has a disability and has difficulty walking she can no longer drive she has problems seeing now yeah I would think if I could ever afford to get her a robot down the road it would be great for her life someone to take her with her to grocery shopping things like that yeah yeah absolutely and Tesla you know they're they're predicting $25 to $30 billion in industry and there's a huge market I mean I can't think of a segment that wouldn't want a robot yeah for sure now the other side of it is the robots you're seeing that you know as they said when do we expect Egypt to hear Elon Musk say they asked him when do you expect slaughter bots to show up he said five to ten years yeah yeah five to motionally slaughter bots yeah ones that go out and kill people yeah five to ten years so it's like oh my gosh yeah I wonder I thought about that too the implications for war that are we looking at the new plus they probably can fly it'd be a half drone half uh uh fighter and then if they have the neuro link then they're all insane yeah it's gonna be like a new episode of Tron did you see a new episode of Tron no oh my gosh that thing can kill anything and rebuild itself again yeah in the story wow it's compelling let's wrap let's put a bow on this thing listen we are talking about all of our topics AI cyber quantum crypto robotics and take a look all of those are starting to converge tesla has three items right now that they've converged the robot the AI and autonomous and we are convert in automobiles we are converging all these technologies at light speed and that is what we are covering in this podcast the number one rated sub billionaire podcast of the world and we are focusing on the tech futures and making certain that we all understand it we keep abreast of it and we bring people up to speed very quickly and every week you know the i tell people often the way to become an expert in anything is to just expose yourself to it on a weekly or a regular basis and within a few months you have expertise in an area that you never had before that's what we're doing here thanks guys for jumping in there i know Gus you're in california gym you're in california and i'm over here in austin texas all by my lonesome and we really appreciate everybody i hope everybody's been able to utilize our system to gain access to the slide materials that we have we have a lot of other information we have poles we have participation that you can get from this podcast by entering the qr code that you'll see on the thumbnail thanks everybody for joining us your