Former Google CEO Spills ALL! (Google AI is Doomed)

when they are delivered at scale it's going to have an impact on the world at a scale that no one understands yet much bigger than the horrific impact we've had on by social media Eric Schmidt just did an interview at Stanford and if you're not familiar with who Eric Schmidt is he was the CEO of Google for a very long time he took that company from a little search engine to being a trillion Tech Behemoth and in this interview he talks all about artificial intelligence and he actually got into some controversy for this interview and

specifically he had some interesting takes on Google's current culture and why they're kind of having their lunch eaten by other AI companies so let's watch this interview together and I'm going to give you some of my own commentary on it now here's the thing this interview went so viral and those few words that he said about Google's culture that he actually convinced Stanford to take the video down off of YouTube so I had to find a backup link to get this downloaded and let's watch it together he's had done a few things since then uh

at Google starting I think 2001 and uh Schmid Futures starting in7 so Google in 2001 that was years before they went public what an absolutely incredible and exciting time it must have been to join that company in 2001 where do you see AI going in the short term which I think you defined as the next year or two um things have changed so fast I feel like every six months I need to sort of give a new speech on what's going to happen um can anybody hear the computer there bunch of computer science in here

can anybody explain what a million token context window is for the rest of the class here um basically allows you to promp with like a million tokens or a million words or whatever you decide to so you can ask a million word question yes I know this is a very large Direction in Gemini right now um no no going to 10 yes yeah anthropic is at 200,000 going to a million and so forth all right so what are they talking about and this is a very long video so I'm going to try to cut out

a little bit of the fat so what are they talking about he asked what is a million token context window if you watch this Channel at all you should know but it is essentially how many words that you can prompt ask a large language model with and that also includes the response typically so when you hear the term context window it's both the prompt and the resp response and when you have a million tokens which Google is by far the best at large context Windows that's kind of one of the few things they're really good

at in the AI realm you could do pretty incredible things there's different use cases that get unlocked specifically because of these large context Windows all right let's keep watching anybody can anybody here give a technical definition of an an AI agent again that might [Music] all right so sorry the audio quality is really bad when they're asking the audience for questions but basically said agents are llms that have additional functionality yeah so an agent is something that does does some kind of a task another definition would be that it's an llm State and memory okay

can anybody again computer scientist can can any of you define text to action taking text and turning it into an action right here go ahead yes instead of taking text and turning it into more text more text taking text and have the AI trigger actions based so another definition would be language to python a pro programming language I never wanted to see survive and and everything in AI is being done in Python there's a that's hilarious I did not know he was so against python python is beautiful it looks like natural language or at least

it's one of those languages that is as close to natural language looking syntactically as it gets and now he's talking about text to action which is essentially what agents can do but also he's just saying yeah you put in some natural language and convert it to python python can get executed and actually accomplish things whether that's searching the web turning your lights on and off whatever it is new language called Mojo that has just come out which looks like they finally have addressed AI programming but we'll see if that actually survives over the dominance of

python so real quick for those of you who aren't very familiar with Eric Schmidt he is a futurist through and through he is incredibly knowledgeable about technology he is one of the innovators who really brought cloud computing to prominence and in fact I'll double check but I'm pretty sure he was the one who came up with the term cloud computing yeah and look at this former CEO of Google AR Schmidt coined the term cloud computing okay let's keep watching one more technical question why is NVIDIA worth $2 trillion and the other companies are struggling I

mean I think it just boils down to like most of code needs to run with Cuda optimizations that currently only Nvidia GPU support so other companies can make whatever they want to but unless they have the 10 years of software there you don't have the machine learning optimization I like to think of Cuda as the C programming language for gpus yeah right that's the way I like to think of it it was founded in 2008 I always thought it was a terrible language and yet it's become dominant there's another Insight there's a set of Open

Source libraries which are highly optimized to Cuda and not anything else and everybody who builds all these Stacks right this is completely missed in any of the discussions the common it's technically called VM and a whole bunch of libraries like that highly optimized Kula Cuda very hard to replicate that if you're a competitor so what does all this mean yeah so just a quick clarification what he's talking about Cuda is a programming language that Nvidia created that is specific to their gpus now these other players that he's talking about are like Intel and AMD and

why are they struggling well first of all they focused on CPUs not gpus for a very long time Nvidia has been in the GPU game since the '90s or maybe even before then but I remember buying Nvidia gpus to play video games in the 9s so they've been around forever and they built this library and they went all in on AR icial intelligence because they noticed that large language models the compute necessary to run them was essentially the same exact math necessary to run video games so they were able to kind of seamlessly transition into

being an AI company versus a video game company in the next year you're going to see very large context Windows agents and text action when they are delivered at scale it's going to have an impact on the world at a scale that no one understands yet much bigger than the horrific impact we've had on by social media so here's why in a context window you can basically use that as shortterm memory and I was shocked that context Windows could get this long the technical reasons have to do with the fact that it's hard to serve

hard to calculate and so forth the interesting thing about short-term memory is when you feed the the you ask it a question read 20 books you give it the text of the books is the query and you say tell me what they say if forgets the middle which is exactly how human brains work too right that's where we are and in previous videos I've talked about this the way the human brain works and I remember this exercise from like sixth grade my teacher did an exercise with us and really told us nothing else and he

just said okay I'm going to say a bunch of words and so we said all the words and then we were tasked with writing down the ones that we remember and what we noticed is we remembered a lot of the words at the beginning and a lot of the words at the end but in the middle we forgot them and it's so weird how large language models behave very similarly to how the human memory works with respect to agents there are people who are now building essentially llm agents and the way they do it is

they read something like chemistry they discover the principles of chemistry and then they test it and then they add that back into their understanding right that's extremely powerful and then the third thing as I mentioned is text action so I'll give you an example the government is in the process of trying to ban Tik Tok we'll see if that actually happens if Tik Tok is banned here's what I propose each and every one of you do say to your llm the following make me a copy of Tik Tok steal all the users steal all the

music put my preferences in it produce this program in the next 30 seconds release it and in one hour if it's not viral do something different along the same lines that's the command boom boom boom boom right you understand how powerful that is if you can go from arbitrary language to arbitrary digital command which is essentially what python in this scenario is imagine that each and every human on the planet has their own programmer that actually does what they want as opposed to the programmers that work for me who don't do what I ask right

the programmers here know what I'm talking about so imagine a non arrogant programmer that actually does what you want and you don't have to pay all that money to and there's infinite supply of these programs this is all within the next year or two very soon so we've already discussed on this channel a number of different versions of this whether you're talking about ader or Devon or pythagora or just using agents to collaborate with each other and code there are so many great options for coding assistance right now now ai coders that can actually build

full stack complex applications we're not quite there yet but hopefully soon and also what he's describing of just saying download all the music and the secrets and recreate that's not really possible right now obviously all of that stuff is behind security walls and you can't just download all that stuff so if he's saying hey reproduce the functionality you can certainly do that those three things and I'm quite convinced it's the union of those three things that will happen in the next wave so you asked about what else is going to happen um every six months

I oscillate so we're on a is an even odd oscillation so at the moment the gap between the frontier models which they're now only three I ref few who they are and everybody else appears to me to be getting larger six months ago I was convinced that the Gap was getting smaller so I invested lots of money in the little companies now I'm not so sure and I'm talking to the big companies and the big companies are telling me that they need 10 billion 20 billion 50 billion 100 billion Stargate is a what 100 billion

right they're very very hard I talked Sam Alman is a close friend he believes that it's going to take about 300 billion maybe more I pointed out to him that i' done the calculation on the amount of energy required and I and I then in the spirit of full disclosure went to the White House on Friday and told them that we need to become best friends with Canada because Canada has really nice people helped invent Ai and lots of hydrop power because we as a country do not have enough power to do this the alternative

is to have the Arabs fund it and I like the Arabs personally uh I spent lots of time there right but they're not going to adhere to our national security rules whereas Canada and the US are part of a Triumph at where we all agree so these are all topics that we've covered and reviewed in Leopold Ashen Brenner's situational awareness paper first of all we definitely don't have enough energy to power reaching AGI it's just not possible but he's also assuming that we're going to need more and more data larger and larger models but I'm

not 100% convinced that that is actually true and Sam Alman has said similar things he's saying well we need to be able to do more with less or even amount of data because we've already used all the data that Humanity has ever created there's really no more there's some obviously proprietary sites like YouTube and Twitter but beyond that everything's been done we've used it to train models and so we're going to need to either figure out how to create synthetic data that is valuable not just derivative and you start to have these reducing rate of

returns but we're also going to have to do more with the data that we do have okay so I'm going to turn up the speed slightly because this is over an hour long video and I want to make sure that we get through all of it in a decent amount of time you were at Google for a long time and uh they invented the Transformer architecture um it's all Peter's fault thanks to uh to brilliant people over there like Peter and jeffan and everyone um but now it doesn't seem like they're they've kind of lost

the inititive to open Ai and even the last leaderboard I saw anthropics Claud was at the top of the list um I asked Sundar this he didn't really give me a very sharp answer maybe maybe you have a a sharper or more objective explanation for what's going on there I'm no longer a Google employee yes um in a spiritful disclosure um Google decided that work life balance and going home early and working from home was more important than winning okay so that is the line that got him you know in trouble not really in trouble

but was controversial it was everywhere all over Twitter all over the news that line Google prioritized work life balance going home early not working as hard as the competitor to winning they chose work life balance over winning and that's actually a pretty common perception of Google if you've watched the show Silicon Valley the company hulie which is really just a parody of Google there's a group of Engineers that just sit on the roof kind of suntanning all day not doing anything and just vesting and apparently that is based on things that happened at Google and

so that's what he's talking about here and maybe that is what happened and the startups the reason startups work is because the people work like hell and I'm sorry to be so blunt but the fact of the matter is if you all leave the university and go found a company you're not going to let people work from home and only come in one day a week if you want to compete against the startups when in the early days of Google Microsoft was like that exactly okay so I don't actually believe going into the office is

absolutely required hard work working your butt off yes if you're going to start a startup you got to be prepared to put in crazy hours for extended periods of time without burning out but I don't think it's necessary to be in person every single day Nvidia the company he's talking about the most valuable company in the world is a remote company they allow work from home and there's also plenty of other examples of that so it's definitely not required but that's not what Eric Schmidt thinks there's there's a long history of in my industry our

industry I guess of companies winning in a genuinely creative way and really dominating a space and not making this the next transition it's very well documented and I think that the truth is Founders are special the founders need to be in charge the founders are difficult to work with they push people hard um as much as we can dislike elon's personal Behavior look at what he gets out of people I had dinner with him and he was flying I was in Montana He was flying that night at 10 p.m. to have a meeting at midnight

with x. a think about I was in Taiwan different country different culture and they said that this is tsmc who I'm very impressed with and they have a rule that the starting phds coming out of the they're good good physicists work in the factory on the basement floor now can you imagine getting American physicists to do that the PhD highly unlikely yeah I I agree the the work ethic required to start something incredible to start a world changing company is just it's required there is no way around it and yes say what you will about

Elon Musk he works his butt off and he requires his team to do the same and they've built incredible things and when I had my own startup years and years ago I put in crazy hours I made a lot of sacrifices I mean every day all day 7 days a week that's all I thought about that's all I did for years and it you know it took a lot out of me it's not for everyone but if you are wanting to do that you got to be prepared to put in that kind of work and

the problem here the reason I'm being so harsh about work is that these are systems which have Network effects so time matters a lot and in most businesses time doesn't matter that much right you have lots of time you know Coke and Pepsi will still be around and the fight between Coke and Pepsi will continue to go along and it's all glacial right when I dealt with Telos the typical Telco deal would take 18 months to sign right there's no reason to take 18 months to do anything get it done just we're in a period

of Maximum growth maximum gain so and also it takes crazy ideas like when Microsoft did the open AI deal I thought that was the stupidest idea I'd ever heard Outsourcing essentially your AI leadership to open Ai and Sam and his team I mean that's insane nobody would do that at Microsoft or anywhere else and yet today they're on their way to being the most valuable company they're certainly head-to-head and apple apple does not have a good AI solution and it looks like made it work is going to play a role for competition with China as

well so I was the chairman of an AI commission that sort of looked at this very carefully and um you can read it it's about 752 pages and I'll just summarize it by saying we're ahead we need to stay ahead and we need lots of money to do so so he was asked about competition with China AI AGI and that's his answer we're ahead we need to stay ahead and we need money customers were the senate in the house um and out of that came the chips act and a lot of other stuff like that

um the a rough scenario is that if you assume the frontier models drive forward and a few of the open source models it's likely that a very small number of companies can play this game countries excuse me what are those countries or who are they countries with a lot of money and a lot of talent strong Educational Systems and a willingness to win the US is one of them China is another one how many others are there are there any others I don't know maybe but certainly the the in your lifetimes the battle between the

US and China for knowledge Supremacy is going to be the big fight that's interesting I had not really heard that phrased in that way the battle for knowledge Supremacy will be between the US and China in our lifetimes very interesting so the US government banned essentially the Nvidia chips although they weren't allowed to say that was what they were doing but they actually did that into China um they have about a 10year chip advant we have a roughly 10year chip advantage in terms of subdv that is sub five roughly 10 years wow um and so

you're going to have so an example would be today we're a couple of years ahead of China my guess will get a few more years ahead of China and the Chinese are whopping mad about this it's like hugely upset about it so that's a big deal that was a decision made by the Trump Administration and improved by the Biden Administration do you find that the administration today and Congress is listening to your advice do you think that it it's going to make that scale of investment I mean obviously the chips act but beyond that building

building a massive AI system so so as you know I I lead a an informal adhawk non-legal group that's not that's different from illegal exactly just to be clear which includes all the usual which includes all The Usual Suspects yes and The Usual Suspects over the last year came up with basis of the reasoning that became the um uh uh the B administrations uh AI act which is the longest Presidential Directive in history you're talking the special competitive studies project this is the actual the actual act for from the executive office okay and they're busy

implementing the details so far they've got it right and so for example one of the debates that we had for the last year has been how you detect danger in a system which has learned it but you don't know what to ask it okay so in other words it's a core it's a sort of a core problem it's learned something bad but it can't tell you what it learned and you don't know what to ask it and there's so many threats right like it learned how to mix chemistry and some new way but you don't

know how to ask it and so people are working hard on that but we ultimately wrote in our memos to them that there was a threshold which we arbitrarily named as 10 to the 26 flops which technically is a measure of computation that above that threshold you had to report to the government that you were doing this and that's part of the rule I think that's kind of BS so first having like a fixed number 10 to the 26 flops anything above that you have to report it there are so many things that could happen

between now and then first of all models are getting bigger let's assume that continues then all of a sudden that number would need to change over time because more and more companies are going to achieve that level of quality but also what if we come up with new techniques where it doesn't require as much compute and again referencing back to things Sam Alman has said they're trying to do more with less so I'm not so sure that that is a good way to do it so yeah let's keep watching I think all of these distinctions

go away because the technology will now the technical term is called Federated training where basically you can take pieces and Union them together oh yeah I guess that's another point right Federated training very similar to how Torance work you basically distribute the workload and then you put it all together in the end that is how a lot of Google's compute actually works so yeah there's not really a clean way to say there's this delineation this number where above that no good below that all good I just don't think that's going to work you may not

be able to keep keep people safe from these new things well rumors are that that's how open eye has had to train partly because of the power uh consumption there's no one place where they did well let's talk to about a real war that's going on I know that uh something you've been very involved in is uh the Ukraine war and in particular uh I don't know how much you can talk about white stor and your your goal of having a 500,000 $500 drones destroy $5 million tanks so how that Chang in Warfare so I

worked for the Secretary of Defense for seven years and and tried to change the way we run our military I'm I'm not a particularly big fan of the military but it's very expensive and I wanted to see if I could be helpful and I think in my view I largely failed they gave me a medal so they must give medals to failure or you know whatever but my self-criticism was nothing has really changed and the system in America is not going to lead to real Innovation so watching the Russians use tanks to destroy apartment buildings

with little old ladies and kids just drove me crazy so I decided to work on a company with your friend Sebastian thrun and a as a former faculty member here and a whole bunch of Stanford people and the idea basically is to do two things use AI in complicated powerful ways for these essentially robotic War and the second one is to lower the cost of the robots now you sit there and you go why would a good liberal like me do that and the answer is that the whole theory of armies is tanks artilleries and

mortar and we can eliminate all of them and we can make the penalty for invading a country at least by land essentially be impossible it should eliminate the kind of land battles well this this so what he's talking about is Ukraine has been able to create drones kind of makeshift drones 3D printed drones for a few hundred dollars carries a bomb drops it on a million1 million tank and they've been able to do that over and over again so there's this asymmetric Warfare happening between drones and more traditional artillery I want to switch to a

little bit of a philosophical question so there was an article that you and Henry Kissinger and Dan hleer wrote last year about the nature of knowledge and how it's evolving I discussion the other night about this as well so for most of History humans sort of had a mystical understanding of the universe and then there's the Scientific Revolution and the enlightenment um and in your article you argue that now these models are becoming so complicated and uh uh difficult to understand that we don't really know know what's going on in them I'll take a quote

from Richard fan he says what I cannot create I do not understand the saw this quote the other day but now people are creating things they do not that that they can create but they don't really understand what's inside of them is the nature of knowledge changing in a way we he's referencing the way that large language models work which is really essentially a black box you put in a prompt you get a response but we don't know why certain nodes within the algorithm light up and we don't know exactly how the answers come to

be it is really a black box there's a lot of work being done right now trying to kind of unveil what is going on behind the curtain but we just don't know I have to start just taking the word for these models without them able being able to explain it to us the analogy I would offer is to teenagers if you have a teenager you know that they're human but you can't quite figure out what they're thinking um but somehow we've managed in society to adapt to the presence of teenagers right and they eventually grow

out there and just serious so it's probably the case that we're going to have knowledge systems that we cannot fully characterize but we understand their boundaries right we understand the limits of what they can do and that's probably the best outcome we can get do you think we'll understand the limits we we'll get pretty good at it the consensus of my group that meets on every week is that eventually the way you'll do this uh it's called so-called adversarial AI is that there will there will actually be companies that you will hire and pay money

to to break your AI system like red team all right let's take some questions from the student there's one right there in the back just say your name few mentions and this is related to comment right now I'm getting AI that actually does what you want you just mentioned adversarial and I'm wondering if could elaborate on that more so it seems to be besides obviously compute will increase and get more performant models but getting them to do what you want issue answer well you have to assume that the current hallucination problems become less right in

as the technology gets better and so forth I'm not suggesting it goes away and then you also have to assume that there are tests for efficacy so there has to be a way of knowing that the thing succeeded so in the example that I gave of the Tik Tok competitor and by the way I was not arguing that you should illegally steal everybody's music what would do if you're a Silicon Valley entrepreneur which hopefully all of you will be is if it took off then you'd hire a whole bunch of lawyers to go clean the

mess up right but if if nobody uses your product it doesn't matter that you stole all the content and do not quote me oh my God yeah it's funny this has happened a million times basically rather than asking permission ask for forgiveness that's essentially what he's describing which is you know the ethos of Silicon Valley right right you're you're on camera yeah that's right but but you see my point in other words Silicon Valley will run these tests and clean up the mess and that's typically how those things are done so so my own view

is that you'll see more and more um performative systems with even better test and eventually adversarial tests and that'll keep it within a box the technical term is called Chain of Thought reasoning and people believe that in the next few years you'll be able to generate a thousand steps of Chain of Thought reasoning right do this do this it's like building recipes right that the recipes you can run the recipe and you can actually test that it produce the correct outcome and that's how the system will work yes sir that was maybe not my exact

understanding of Chain of Thought reasoning my understanding of Chain of Thought reasoning which I think is accurate but let me know what you think in the comments is when you break a problem down into its basic steps and you solve each step allowing for progression into the next step not that it allows you to kind of replay the steps it's more of how do you break problems down and then think through them step by step in general you seem super positive about the potential for problems I'm curious like what do you think is going to

drive that is it just more compute is it more data is it actual ships uh yes the amounts of money being thrown around are mindboggling and um I've chose I I essentially invest in everything because I can't figure out who's going to win and the amounts of money that are following me are so large I think some of it is because the early money has been made and the big money people who don't know what they're doing have to have an AI component and everything is now an AI investment so they can't tell the difference

I Define ai as Learning Systems systems that actually learn so I think that's one of them the second is that there are very sophisticated new algorithms that are sort of post Transformers my friend my collaborator for a long time has invented a new non- Transformer architecture there's a group that I'm funding in Paris that has claims to have done the same thing so there there's enormous uh invention there a lot of things so he didn't actually mention what those were but Mamba is an example of a non- transformer architecture and the final thing is that

there is a belief in the market that the invention of intelligence has infinite return so let's say you have you put $50 billion of capital into a company you have to to make an awful lot of money from intelligence to pay that back so it's probably the case that we'll go through some huge investment bubble and then it'll sort itself out that's always been true in the past and it's likely to be true here yeah so there's been something like a trillion dollars already invested into artificial intelligence and only 30 billion of Revenue I think

those are accurate numbers and really there just hasn't been a return on investment yet but again as he just mentioned that's been the theme on previous waves of Technology huge upfront investment and then it pays off in the end and personally obviously I'm kind of dedicating my life to artificial intelligence I believe it's going to be bigger than anything else bigger than the internet bigger than mobile phones it is going to be the most transformative technology that humans have ever created and what you said earlier was you think that the leaders are pulling away from

right now right now and and this is a really the question is um roughly the following there's a company called mrr in France they've done a really good job um and I'm I'm obviously an investor um they have produced their second version their third model is likely to be closed because it's so expensive they need revenue and they can't give their model away so this open source versus closed Source debate in our industry is huge and um my entire career was based on people being willing to share software in open source everything about me is

open source what um didn't he run Google and Google was certainly closed source and everything about Google was protect the algorithm at all costs so I don't know what he's referring to there much of Google's underpinings were open source everything I've done technically and yet it may be that the capital costs which are so immense fundamentally changes how software is built you and I were talking um my own view of software programmers is that software programmers productivity will at least double there are three or four software companies that are trying to do that I've invested

in all of them in the spirit and they're all trying to make software programmers more productive the most interesting one that I just met with is called augment and I I always think of an individual programmer and they said that's not our Target our Target are these 100% and software programming teams on millions of lines of code where nobody knows what's going on well that's a really good AI thing will they make money I hope so the very beginning you mentioned that there's the combination of the context window extansion the agents and the text to

action is going to have unimaginable impacts first of all why is the combination important and second of all I know that you know you're not like a crystal ball and you can't necessarily tell the future but why do you think it's beyond anything that we could imagine I think largely because the context window allows you to solve the problem of recency the current models take a year to train roughly six six there 18 months six months of preparation six months of training six months of fine- tuning so they're always out of date contact window you

can feed what happened like you can ask it questions about the U the Hamas Israel war right in a context that's very powerful it becomes current like Google um in the case of Agents I'll give you an example yeah so that's essentially how search GPT works for example so the new search product from open AI it can scour the web scrape the web and then take all of that information and then put it into the context text window that is the recency he's talking about set up a foundation which is funding a nonprofit which starts

there's a u i don't know if there's Chemists in the room that I don't really understand chemistry there's a tool called chro C which was an llm based system that learned chemistry and what they do is they run it to generate chemistry hypothesis about proteins and they have a lab which runs the tests overnight and then it learns that's a huge acceleration accelerant in chemistry Material Science and so forth so that's that's an agent model and I think the text to action be understood by just having a lot of cheap programmers right um and I

don't think we understand what happens and this is again your area of expertise what happens when everyone has their own programmer and I'm not talking about turning on and off the lights you know I imagine another example um for some reason you don't like Google so you say build me a Google competitor yeah you personally you don't build me a Google competitor uh search the web build a UI make a good copy um add generative AI in an interesting way do it in 30 seconds and see if it works right so a lot of people

believe that the incumbents including Google are vulnerable to this kind of an attack now we'll see so I have a slightly different perspective on this I don't think somebody a consumer is going to say build me another Google all we need to say is give me that piece of information I'm looking for from that Google competitor there's no need to build the application the large language model is every layer of the stack it is the operating system it is the application later it is everything so there's really no need to replicate other software all you

need is whatever that software delivers questions who were sent over by slider I want to give some of them were uploaded so here's one um we talked a little bit about this last year um how can we stop AI from influencing public opinion misinformation especially during the upcoming election what are the short and long-term solutions from most of the misinformation in this upcoming election and globally will be on social media and the social media companies are not organized well enough to police it if you look at Tik Tok for example there are lots of accusations

that Tik Tok is one kind of misinformation over another and there are many people who claim without proof that I'm aware of that the Chinese are forcing them to do it I think we just we have a mess here and um the country is going to have to learn critical thinking that may be an impossible challenge for the us but but the fact that somebody told you something does not mean that it's true oh man I wish he could tell a lot of the people I know that just because someone says something online or just

because somebody says something to you does not make it true please use critical thinking just assume it may be false they really are true and nobody believes anymore you get some people call it epistemological crisis that that now you know El says no I never did that prove it well let's use Donald Trump um okay look I I think we've got we have a trust problem in our society democracies can fail and I think that the the greatest threat to democracy is misinformation because we're going to get really good at it um when I man

managed YouTube the biggest problems we had on YouTube were that people would upload false videos and people would die as a result we had a no death policy shocking and we just went it was just horrendous to try to address this and this is before generative AI yeah and also it's not even about potentially making def fakes or kind of misinformation just muddying the waters enough just the Spectre of Doubt when a group of folks think about a problem or a topic or anything just that doubt is enough just muddying the waters is enough to

make the entire topic kind of Untouchable I'm really curious about the TX to action and its impact on for example Computer Science Education wondering what you have thoughts on like how cus education should transform to kind of Meet the age well I'm assuming that computer scientists as a group in undergraduate school will always have a programmer buddy with them so when you when you learn learn your first for Loop and so forth and so on you'll have a tool that will be your natural partner and then that's how the teaching will go on that the

professor you know he or she will talk about the concepts but you'll engage with it that way and that's my guess I have a slightly different view I think in the long run there probably isn't going to be the need for programmers eventually it's just going to be a consumer who can't program speaking an actual language saying exactly what they want whether that's a piece of information it's a piece of functionality something to happen in the real world and then the large language model makes it happen and eventually the llms will become so sophisticated they're

writing their own kind of code and maybe it gets to a point where we can't even read that code anymore that it is is just symbols for example and it's so hyper efficient because of that code is only the way it looks today because humans need to read it and humans are bad at programming so there is this future in which there are no programmers whatsoever now there's probably going to be a need for computer scientists in the long run but there's maybe even an argument that that is not the case especially if we have

things like you know with Sakana AI just came out with essentially self researching AI it's able to formulate theories test those theories do its own peer review and then publish so there is this world in which it is not necessary to have programmers researchers computer scientists I'm not sure that's the way it's going to be but there is a timeline in which that happens you mentioned your in your paper onal security as you have China the us today the next 10 that next for all other us allies or te up nicely to the US allies

curious what your take is on on those 10 like the middle allies um what is Stu How likely are they to get on board with securing our and what would to get the most interesting country is India because the top AI people come from India to the us and we should let India keep some of its top talent not all of them but some of them um and they don't have the kind of training facilities and programs that we so rich we have here to me India is the big swing state in that regard China's

Lost it's not going to not going to come back they're not going to change change the regime as much as people wish them to do Japan and Korea are clearly in our camp Taiwan is a fantastic country whose software is terrible so that's not gonna going to work um amazing hardware and in the rest of the world there are not a lot of other that's interesting I had not heard that so Taiwan has amazing Hardware but terrible software I'm not even sure what that means exactly uh if you know let me know in the comments

German the EUR Europe is screwed up because of Brussels it's not a new fact I spent 10 years fighting them and I worked really hard get them to fix the a the EU act and they still have all the restrictions that make it very difficult to do our kind of research in Europe my French friends have spent all their time battling Brussels and macron who's a personal friend is fighting hard for this and so France I think has a chance I don't see I don't see Germany coming and the rest is not big enough interesting

so yeah EU is overregulation in the long run given the capabilities that you envision these models having should we still spend time tring to code yeah because because ultimately it's okay she asked should we still learn to code the old thing of why do you study English if you can speak English you get better at it right you really do need to understand how these systems work and I feel very strong I don't know if I agree let me know what you think in the comments curious if you've explored the distributed setting and I'm asking

because sure like making a large cluster is difficult but MacBooks are powerful there's a lot of small machines across the world so like do you think like folding at home or a similar idea works for training system yeah we've looked very hard at this so the way the algorithms work is you have a very large Matrix and you have yeah so this person specifically asked about whether you can do distributed learning on the models distributed compute on the models essentially a multiplication function to think of it as going back and forth and back and forth

and these systems are completely limited by the speed of memory to CPU or GPU and in fact the next iteration of Nvidia chips has combined all those function into one chip the chips are now so big that they glue them all together and in fact the package is so sensitive the package is put together in a clean room as well as the chip itself so the answer looks like supercomputers and speed of light especially memory interconnect really dominated so I think unlikely for a while is there a way to segment the LM like so Jeff

Dean last year when he spoke here talked about having these different parts of it you train separately and then kind of Federate them each you know in order to do that you'd have to have 10 million such things and then your the way you ask the questions would be too slow he's talking about eight or 10 or 12 not at his level see in the back yes way back I know like after releas the New York Times Su open using their works for training where do you think that's going to go what that means for

I used to do a lot of work on the music licensing stuff what I learned was that in the 60s there was a series of lawsuits that resulted in an agreement where you get a a stipulated royalty whenever your song is played even even they don't even know who you are just paid into a bank and my guess is it'll be the same thing there'll be lots of lawsuits and there'll be some kind of stipulated agreement which will just say you have to pay x% of whatever Revenue you have in order to use the pass

Cap all right I know a lot of you disagree with my takes on copyright with AI uh I don't even have strong takes to be honest I think maybe when I've conveyed them in the past I've been maybe a little bit more severe than I meant to but what he is saying is uh more akin to I guess what I believe which is the content creators the content owners should get a piece you're creating a new product based on an existing piece of Ip of course it makes sense that there should be some monetary incentive

to do that there's a few players that are dominating AI right and they'll continue to dominate and they seem to overlap with the large companies that all the antitrust regulation is kind of focused on how do you see those two Trends kind of yeah like do you see Regulators breaking up these companies and how will that affect the yeah so in my career I helped Microsoft get broken up and it wasn't broken up and I fought fought for Google to not be broken up and it's not been broken up yeah so this obviously came out

before this week and just this week it seems like the FTC is really strongly considering breaking up Google um I haven't covered it on the channel much but that is a thing that is happening right now and we'll see what happens with that but he seems overly confident in this moment because he doesn't know what's to come so it sure looks to me like the trend is not to be broken up um as long as the companies avoid being John D Rockefeller the senior and I studied this look it up it's how antitrust law came

I don't think the governments will act act the re the reason you're seeing these large companies dominate is who has the capital to build these data centers right right so my friend Reed and my friend Mustafa coming next two weeks from now have Reed talked to you about the decision that they made to take inflection and essentially PE part it into Microsoft basically they decided they couldn't raise the tens of billions of dollar is that number public that you mentioned earlier no have have give maybe we can say I know you got to go I

I don't you I want to leave should we do this questione you so much I was wondering where all of this is going to leave countries who are nonparticipants in of development of Frontier models and access to compute for example the rich get richer and the poor do the best they can wow the rich get richer and the poor do the best they can boy is that not a dystopian Outlook of the future do you have any advice for folks here as they're building their they're writing their business plans for this class or policy proposals

or search proposals um you know at this stage of the careers going forward well um I teach a class in the business school on this so you should come to my class um the I am struck by the speed which with which you can build demonstrations of new ideas so in that in one of the hackathons I did the winning team the command was fly the Drone between two towers and it was given a virtual drone space and it figured out how to fly the Drone what the word between meant generated the code in Python

and flew the Drone in the simulator through the tower I just it would have taken a week or two from you know good professional programmers to do that um I'm telling you that the ability to prototype quickly really you know part part of the problem with being an entrepreneur is everything happens faster well now if you can't get your prototype built in a day using these various tools you need to think about that right because that's who your competitor is doing so I guess my biggest advice is when you start thinking about a company it's

fine to write a business CL plan in fact you should ask the computer to write your business plan for you um as long as I talk about that after you leave this and and but but but I think it's very important to prototype your idea using these tools as quickly as you can because you can be sure there's another person doing exactly that same thing in another company in another University in a place that you've never been yeah I mean that follows with the logic that has been in Silicon Valley forever ship early ship often

get it out get the feedback iterate as quickly as you can that is an extension of what he talked about earlier where the work ethic is everything if you're going to start a company so that was it that was the Eric Schmidt interview at Stanford I thought it was fascinating and a couple new topics and terms that I had not heard so I'm definitely going to go research those things right now if you enjoyed this video please consider giving a like And subscribe and I'll see you in the next one