Should We Be Fearful of Artificial Intelligence? w/ Emad Mostaque, Alexandr Wang, and Andrew Ng | 39

we're all going to have some version of Jarvis right an AI that is in your ear on your body and so forth frankly even as someone in AI I feel like man it's tiring there's so much going on physical things in the real world are much further from being disrupted by Ai and it's going to create wealth faster than anything we've ever [Music] seen the beauty of AI is that they can analyze information faster than any person we just l starting to explore these all of us are engaged with governments we're trying to help them it's a bit difficult when they're just catching up to the internet right now now it's got to the point where this technology could potentially be dangerous so not everything needs to be open but we do need to be transparent and careful when stuff gets out of our control I have to admit I'm not letting Chad GPT tell me where to invest at this moment in time in school a lot of kids have been cheating and they've been using Chad GPT to write essays and book reviews and do their math cheating implies as a contest School should not be a contest you will never not be without this AI as you grow up my question is about content creation especially in music you will have perfect music models by the end of the year we believe this technology could be existential and we will treat it as such one of the greatest drivers for peace on the planet is making sure every mother has their children with the best health and the best education a more peaceful world for us is a world in which we uplift everybody which is what this technology has the ability to do everybody this year at abundance 360 given the meteoric rise of AI I decided to put together an entire day on AI with three of the most extraordinary thinkers uh you've met one of them Imad mustak the chairman founder CEO of stability AI also brought to the conversation Alexander Wang who's the founder and CEO of scale AI it's a big data machine learning platform that accelerates the development of AI he founded it when he was at MIT at age 19 as a student dropped out after the first year to become the youngest self-made billionaire ever I also brought Andrew Ang the managing partner at AI fund he's co-founder and chairman of cousera uh the founder of Google brain I love the name of that and one of times's 100 most influential people between imod Alexander and Andrew you're get an overview of this field how fast is it moving how disruptive is it is it something you should be excited about or fearful of my massive transformative purpose is to inspire and guide entrepreneurs to create a hopeful compelling and abundant future in humanity and that's what I'm doing with this podcast to open up every relationship every conversation I have with you to inspire you to support you and going big in helping uplift Humanity if that's of interest to you please subscribe to this podcast allow me to share with you the wisdom that I'm learning from the most incredible moonshot entrepreneurs on the planet let's jump into this episode it's a conversation that everyone needs to be having in your company in your family and definitely at the heads of every nation enjoy thanks Peter and thanks to our amazing guests here today um we've been talking about this actually for years um when we look at the rapid acceleration of AI and that this disruption that's kind of happening very very quickly from this point forward and and and when we look at Industries and workers and whether they're getting the the the understanding the knowledge and the reskilling that will that needs to happen to approach this current world and what ways do you see or do you believe um we can do that faster and work with governments and larger corporations and and education systems to help enable this um because we're all talking about positivity and abundance um and I think we're very privileged in this room the question is for the rest of the world what are we doing and in what ways um are you involved as leaders by the way we've got 40 minutes here so lots of great questions be asked who wants to take that on you know so it's a very important point I think the world is changing so fast it's difficult for a lot of people to just keep up with one is going on frankly even as someone in AI I feel like man it's tiring there's so much going on um so I think maybe one one of the things I'm doing years ago I started co-founded corera which To This Day offers all of causes often works with governments to train up different uh uh uh different people I think that one nice thing is with um online education with digital education we can speed up the rate at which we help people figure out you know learn the new technologies and figure out what to do with it as well um and I think media also has huge important role to play and eventually I would love if we can build tools like if you look at Google the UI is really simple the background is insanely complicated just type what you want you get answers so I hope that AI tools can evolve to the point where you know you could just use it without nearly as deep of Education needed and I think a lot of generative AI tools are head in that direction yeah I think it's supernatural I mean that's why super space natural like chat GP was so successful because you just type and it just came stable diffusion just type and then you have images right and so I think making these things more and more usable has led to the explosion because it kind of comes from the natural Lang understanding of humanity itself it's trained on these large corpuses of an entire media set I think on the policy side um I think all of us are engaged with governments we're trying to help them it's a bit difficult when they're just catching up to the internet right now um but by the way the most linear organizations on the planet right if not sublinear they are and you know and again the typical approach to this is Regret Minimax you know like stop it ban it but it's inevitable now and again it's kind of a global phenomenon one of the problems as well is that you know you've had an acceleration but you haven't had a maturation so there's lots of excitement how much implementation is there now there's going to be a bit of a pause because it has to be in the current wave even as research keeps going and you go into engineering and application and so I think we're not sure exactly where things are I think Andrew talked a bit about where the value lands on the stack and again we are all kind of at different areas of this stack so we ourselves haven't figured this out and we're doing our best with the community to try and do that to communicate to policy makers and make education courses and others accessible Michael good to see you you're next pal so this is probably a question more to imod but anybody's perspective is welcome I am uh done lots of programming I usually like to think about abstractions or concepts you have a concept let's say drawable that's fit for for some purpose and so it can do that thing now with AI we're starting to we are using English language we're using prompt so definition what you're asking for a lot more vague so so when it's vague it may have Edge problems right you have to clarify things a lot more and so my question is do you still see a value in having well defined structured uniquely named things which you can compose for well understood result or do you think we'll diverge more to everything's a bit fuzzy everybody just provides a lot more text and it just somehow the system figured out for you well I think SE question I think a lot of the applications right now have been very surface level without the contextual awareness of like you know Game of Thrones in South Korea or something like that um Amazon had this thing where they had the six-page memos right rather than presentations but another thing they used to do was was write the press release beforehand and so if you write the actual overview of the program in the comments now for kind of co-pilot and things like that or gp4 it will write the program based on the comments you know so again I think this is a failure for us to understand prompting and some of these other things is something that Andrew and I have been talking about you know again it's gone so fast we at the surface level we have to explore another way to think about this as like game consoles when the Wii first came out games were a bit crap by the end it was fantastic we're just literally starting to explore these so it's simple words now and then we'll compose and build structures around it to get the most out of these weird thingies everybody I'm want to take a quick break from our episode tell you about a health product that I love and that I use every day in fact I use it twice a day it's called seed Health now your microbiome and gut health are one of the most important and modifiable parts of your health plan your gut microbiome is connected to your brain health your cardiac health your metabolic health so the question is what are you doing to optimize your gut let me take a second to tell you what I'm doing every day I take two capsules of seeds daily symbiotic it's a two-in-one probiotic and Prebiotic formulation that supports Digestive Health gut health skin Health heart health and more it contains 24 clinically studied and scientifically backed probiotic strains are delivered in a patented capsule that actually protects it from the stomach acid and ensures that all of it reaches your colon alive with 100% surv ability now if you want to try seeds daily symbiotic for yourself you can get a 25% off your first month supply by using the code moonshots at checkout just go to c.com back/ moonshots and enter the code moonshots at checkout that's seeds.com back/ moonshots and use the code moonshots to get 25% off your first month of seeds daily symbiotic trust me your gut will thank you all right let's get back to our episode when to go to zoom next uh Ned uh alofi Ned where are you on the planet and what's your question hi guys can you hear me okay yes we can hear you great perfect I'm from Chicago uh I am in healthcare and I love the idea of healthcare and Longevity I may be your only practicing physician here um I had the idea of joining this group because as a practicing urologist we are getting pummeled and as Peter and uh and AI and health span go on most urologists are going to see people who are older I think all you guys are going to be a patient of either me or my colleagues here soon the point I'm making here is that there is an overflow of patients and we as clinicians are getting bombarded and the question is in light of workflow workflow shortages with our our our staff and the incoming number of patients how to use AI most efficiently in terms of point of care for patients getting into our offices as well as having AI be part of the evaluation process because as many people may know you know 90% of the time we see the the same 10 conditions so the the question to the uh panelists is number one how would you advise a physician uh or health care provider to set this up and do you think that this is something that is easily attainable I I like Alex's uh thought about starting small and then the second question is if you were to do it is it the sort of thing that would prevent I mean is there anything in there to prevent others from doing the exact same thing so that if you are spending a lot of time and money into this that nobody else just kind of Scoops it from you thank you for taking my question yeah I'll just say it real quick if you have a concrete idea for what you like to do in neurology let me know you know I think teams could help maybe my team could help execute that and one one example something AI funded we worked with a fantastic founder Allison Darcy and now different CE Michael ever to um support building up wot which is a dig digital mental healthc care chat bot um you can actually install it on your phone that you know from the data published out the Stanford seems to be able to take effect relatively quickly in terms of treating symptoms of um for for for users of symptoms of anxiety and depression um so I think that there actually lots of opportunities to apply AI in interesting ways to healthcare um figure out those concrete use cases that allows us or others to execute on it yeah I'll say two more things which is that I think the the first is that you know getting to Market first is actually an advantage you know certainly other people can copy the idea but I think um you know this is one of the things that's well studied in business but there's going to be a Moe from whoever's able to establish the use case first and then able to um use that pull position to get more and more data into the system so let's say you built you know the best possible system for or um identifying some sort of um some sort of um uh disease or or condition using AI then you're going to just by launching that service get far more data than anyone else and race ahead of of where anyone else could possibly be so the the the sort of currency of the realm of AI really is uh is data and if you have a unique a niche data set or a unique data set that's going to give you an initial advantage and if you're able to build a strategy continue amassing more and more data that'll keep you ahead we're going to go next to uh to John uh here at Mike 2 let me remind you please look at the questions in slido up vote them if you're here in the audience or on Zoom uh we'll be pulling them John please what's your question and keep the question short if you could sure so I'm a medium to small business owner I'm a physician I own a diagnostic laboratory so I hear a lot of buzzwords you need to implement AI you need to create um you know you need to put your dat on a balance sheet that sounds awesome uh but the but my initial question is how do do you get started going down that road of you know we we have a ton of data but how do you put it in a package that you can actually monetize it put on a balance sheet so what are the process of just going about doing something like that and then number two implementing AI into something that I think is a very concrete you um uh us like a case point for our company is digitizing pathology and allowing an AI over engine to help pre-diagnosis or whatnot how do you go about implementing something like that in a company that's my size by either pairing with someone else or do you try to bring a team in to develop it so like what kind of Playbook just to get you off the ground to move your company Al I think you're yeah so we have tools um you go to scale.com we have data set management tools that you can upload all your data and basically build that into a data set um that's step one is to get all of it into one tool build up an entire data set and then you can do a bunch of things from there you can label and annotate the data turn into a data set you can train a new model on top of um you can uh bring in a host of companies we could be one of them or there's a bunch of other vendors out there that can take that data and turn it into a a customized model for you and then um and then launch it and figure out how how much impact you can possibly have but then you find out who that data is valuable to ultimately uh we're going to go to slido there's a great question here I'm going to send it to you imod it says here is there any commercial activity which AI will never be able to disrupt what will be the factors to understand that I'm Cur I mean it's an interesting question right it's like the question asked of Bezos what what won't change right there's nothing that will never be able to disrupted because you'll have autonomous robot agents that basically it's Blade Runner right IND from humans so I can't see any commercial activity cannot disrupt do you guys agree Alex let's go with you you know I think um I think at least in the short term and who know this this stuff could all change uh but in the short term uh physical things in the real world are are much further from being disrupted by AI so you know anything that we have to sort of like interact with or do something physical in the real world um robotics is relatively behind a lot of the sort of pure digital pure online kind of AI systems so you know I don't know if the word is never but at least for a long time um you know we won't have a you know an AI robot that can do construction or that can do um you know mining or some of these things that are very uh very interesting I'm I'm I go into the uh I go into emods category here but that's my opinion you guys are actually Andrew you know my my friends and I some of my ACH friends I we used to challenge each other name an industry that AI will not be able to disrupt we challenge each other to name that and I had a hard time coming with them until one day I thought all right maybe hairdressing industry you know which hairdressing hairing hair um and and I used to say this on stage until one day one of my friends who's a robotics professor she was in the audience when I said that on stage and um afterward she stood up she pointed at my head and she said Andrew for most people's hair styles I don't how to put a robot to cut the hair like that but your hair side Rob to do that oh that's great all right we're going to go to Kieran on Mike four do you think uh with the L large language models GPT will they ever be able to explain why they generated what they generated or made a new breakthrough so right now you can ask the models to explain why it said what it did um it's not clear if those answers are actually correlated at all with the actual reason why they aren't probably correlate with why they act actually said that and so there's sort of you know humans have this bug which is that we will do things and then if somebody asks us to explain why we did the thing we'll come up with some post Hawk rational a that often times is not literally the reason why we did a certain thing usually it's you know there's the analogy of the elephant and the elephant rider which is that our emotional brain is an elephant and our rational brain is this is this Rider of the elephant that just sort of is trying to explain what the elephant's doing um so I think I think the large language mods are similar which are that they're going to go through some process by which they make a decision or they say something and then you can ask it to explain and it'll come up with something on the spot to try to explain it um you know this is something that a mod and I were actually just talking about is like how do you get greater levels of interpretability how do you how do we do research in the direction of like actually truly understanding where it's coming from but that's all active research nice Jazelle you're the next contestant good to see you um so um given what has happened to social media I actually raided the way I feel about AI as scared shitless because you guys are you know you're at the Leading Edge of this this is your space is anybody doing anything to build parameters around this technology so that what you were saying you m about someone calling you in with your mother's voice and selling you stock uh that is we already live in a world where we don't know what truth is how are we going to fight this well I think you see information curation going through social networks I mean this is the Twitter verified it's Apple identity it's kind of a bunch of other things webank content authenticity. org which is a doe thing around visual media but we do need to have centralized repositories of trust because information the cost of information becomes zero and creation becomes zero and unfortunately there is nothing that can really regulate this output like there are laws in various countries that you must identify AI output like with stable diffusion there's an invisible WM Mark and things but it's become so accessible that we need need to have standards quicker than the system is adapting so this is why I think information Distribution Systems need to standardize around this right now which is something we're pushing a laot yeah I mean there's a big question can we even govern against any of this right uh all of this is bit we have we live in a world of poest borders data flows so uh jezelle we're going to have a lot of these conversations around ethics and uh around implications it guys this is happening right now right this is this year next year not 10 years from now 5 years from now what gets played out this is why we need to be paying such attention uh to it jelle thank you I'm sure we'll hear from you a bunch more times this let's go uh to uh to mic number three in the back there and then I'll go to uh we go to zoom and we'll go to uh uh our slido all right um my name is hajim uh running a AI startup in Japan series d uh now getting uh much closer to the Japanese government and they're concerning about the disaster and the resilience because uh there's a huge risk on an earthquake as well as like the F Fuji erosions so uh my question is like what could be the interesting but the practical way to apply AI for those like disaster recovery so we actually worked on um something in this direction for in in the Ukraine conflict which has obviously been um you know uh this horrible thing to happen to to region in a country um and you're able to one of the things that um is really unfortunate in any sort of disaster situation is that uh you know seconds really matter you know every second spent that you aren't you know deploying resources to help resolve one of the you know uh people who are who are in a damaged structure or uh you know tackle a certain area is is going to result in either lives loss or further damages to infrastructure so one of the things that you can do with uh a combination of satellite image plus drone imagery on in major cities or major areas that are being impacted is use AI to automatically identify damage on a building by building level and a change in in that damage um in every area and use that to coordinate humanitarian response so um this could be applicable you know this is applicable in any sort of natural disaster um any sort of scenario where fast action really matters the beauty of AI is that it can analyze information faster than any person yeah we're doing a variant of stable diffusion for satellite imagery and time series data as well and so that I'm sure we'll be used in that toolkit yeah amazing all right uh to our Zoom audience we're going to you next uh Nico um NCO dronic where are you on the planet and what is your question hi Peter I'm calling in from Austria from Europe beautiful and I have a question I have a question um regarding the profile of a chief AI officer so what priv profile should this AI officer have to fulfill this or best um should it be technical should it be another profession should it be a mix because it seems to me that combines a lot of traes that are found in entrepreneurs and CEOs anyway so what else is needed and I asked this specifically coming from a non Tech traditional background as long in my case thank you ni a great question let me preface a second what my advice as a first step for folks is I like the idea of get your team to experiment and so forth but in my what's been working for me is finding some body who knows the field out there understands who to partner with is advising as a strategic advisor to the CEO about the platforms and and such as a chief AI officer in that regard do you guys agree with that idea and what would be uh your you know uh what do you think that person should have as as background yeah so as as far as I know I think I might have been the one that coined the term Chief AI officer I stole it from you then no but but so actually years ago I actually wrote an article in har Business Review I remember Googling for chief AI officer didn't really exist on the internet so I actually wrote a piece in h Business Review with my specific recommendations of the chief AI officer profile um I think person needs to be technical enough to understand the tech and then also business oriented enough to work cross functionally to figure out what are the valuable business use cases for your specific application but I have like a multi-page article on on HP uh talk I would imagine the CEO or the or the head of whatever division says this is what I think is possible this is what I need the chief eye officer is sort of an interface to the world out there to bring in a scale to bring in part of what uh imod what you're building is that sound like a reasonable role in a company yeah I think I have be technical enough to make good judgments and then also to make the you know buy versus build decisions and often for many companies you should buy a lot and build a little bit but to make those decisions requires both deep technical judgment as well as um the ability to figure out understand the business well enough to figure out the the use cases I can't tell if you're agreeing or disagreeing no I I think so I think that there's an implementation side of it like an applied engineer is a good thing I think passion is kind of key cuz you have to throw yourself otherwise there's no way you can keep up actually my suggestion would be that um given this the suggestion that you have there should be an internal a360 repository around the ecosystem the latest transend and others that can go to all the chief a officers here so they can keep on touch through that all right Steve you heard it you're in charge uh wonderful uh thank you I I hope Nico that that answered your question let me go to slide on next and thank you some great questions uh Katie I didn't I I apologize I didn't uh note your your your name earlier um actually what is the first step companies can take to evaluate the data they have uh how can we get started extracting uh from the systems we have in place today yeah so I think the first thing to do right now the the new large models are are um quite capable of just sort of um ingesting in existing dat that you have in basically whatever format you have you know in some cases it can be literally as simple as as uploading a bunch of PDFs uh into into the models and and sort of extracting all the information from there so I think by the way it's something you just said is very important right it's like you can almost upload anything yeah if it's text right your outlook files everything well soon images as well Im yeah mul text images yeah one one C turns out a lot of Enterprises have what's called structured data which means basically giant Excel spreadsheet so structured data you know models like CH gbt and so on are less directly well suited for but text and images is getting really good yeah that's true yes right now General mods Still Still wor at uh at Excel than uh than than many of us but um yeah so the these are the data formats you know I would basically start the initial process just catalog catalog all this data that you would want to upload into an AI system and then um and then reach out to one of us to to help you help you put it into to an AI system yeah I want to go to Christie's question on uh on slido here and I'm heading this towards uod uh uh if you had to put all of your money your family's money your friend money into one of the Technologies here or Industries which one would you put it into Ching I'm all about risk diation no I mean um it's an incredibly hard thing to say um my thing is that like what I advise all the young people is actually I'm advising young people don't go to university anymore do phds don't get an NBA go work the thing can get in gr right already so I think you know put it into Andrew's fund I think it'll do well um but there are very few kind of options here apart from you have to use it to basically upskill yourself this is the thing or upskill your community or up yourself inves in your company this is literally the case because this is such a disruptive massive game Cher um and there are no easy ways to do that other than like actually let's spend money reasonably quickly to get up to speed on this technology because you will have an ed of everyone else as alexand said it's first to Market in the stock market itself there are no real investable things here at the moment are we going to see the first trillionaires in this area uh most likely is this where the greatest wealth creation is going to happen or we're going to see that in Bitcoin first yeah I think we see this uh I think again like Bitcoin had a GPU and then it had an asset K where at the GPU era of this technology but it's real and we're going to get to aexs it's going to be everywhere and it's going to create wealth faster than anything we've ever seen uh I call it the do bubble actually um which is why I'm not joking like you know if you have funds that are focused in it almost a rising tide W lifts all boats you know even if you have the alpha side that will come through like a trillion dollars will go into this like 20 billion went into delivery startups last year a total I think probably of six billion gone into this sector to date 100 billion g into self-driving cars a trillion went into 5G yep just amazing anything here that's investable go to for next I want to add one to what im said about jumping in there is one is I think some people sometimes people think am I too late and and the answer is you're not you're actually very early still AI on you know don't it's exponential some very rapid growth but if you jump in now I think you look back a few years from now and people say wow you know that my buddy over there was really early jumping in because because it's still growing so it's Bitcoin it's literally 6 billion it's going to go to 600 billion in the next few years 100 times increase right this episode is brought to you by levels one of the most important things that I do to try and maintain my Peak vitality and Longevity is to monitor my blood glucose more importantly the foods that I eat and how they Peak the glucose levels in my blood now glucose is the fuel that powers your brain it's really important High prolonged levels of glucose what's called hyperglycemia leads to everything from heart disease to alzheimer's to sexual dysfunction to diabetes and it's not good the challenge is all of us are different uh all of us respond to different foods in different ways like for me if I eat bananas it spikes my blood glucose if I eat grapes it doesn't if I eat bread by itself I get this prolonged spike in my blood glucose levels but if I dip that bread in olive oil it blunts it and these are things that I've learned from wearing a continuous glucose monitor and using the levels app so levels is a company that helps you in analyzing what's going on in your body it's continuous monitoring 24/7 I wear it all the time really helps me to stay on top of the food I eat remain conscious of the food that I eat and to understand which foods affect me based upon my physiology and my genetics you know on this podcast I only recommend products and services that I use that I use not only for myself but my friends and my family that I think are high quality and safe and really impact a person's life so check it out levels. link SL Peter give you two additional months of membership and it's something that I think everyone should be doing eventually this stuff is going to be in your body on your body part of our future of medicine today it's a product that I think uh I'm going to be using for the years ahead and hope you'll consider as well I've been uh sitting here and really blown away with all the great technology Ai and and I'm in a business of U in uh real estate which is I'm not playing in that Arena but I'm here to find out what I need to do but what I'm what I'm actually brought me up to to the uh speak f are in real estate here raise your hand please just get a sense so you're not alone okay thank you but my main question is really comes back to the sort of a uh what do we do with all these great things and how do we hack the world of Peace in around the world and how do we uh understand with the data that it's the need of what Russia China Iran North Korea what is their problem what are their challenges what are the stuff that they want to blow everything up and what is our problem and what how do we look at ourselves as a citizen of the planet Earth and how do we work with the Geniuses we have in front of us to really be able to bring this thing because there's a lot of not cool things we're here around the world all right we got your question how do we save Humanity that's an easy one right thank you seriously I mean and and imod please dive in cuz this is something you care deeply about these are Universal translators go to chat GPT pay something and say write it from the perspective of a tea party conservative it'll do that and then you'll say rewrite it from the perspective of a Libertarian and it'll do that so you're going to have the ability very soon we're all going to have some version of Jarvis right an AI that is in your ear on your body and so forth and that AI will be able to tell you are you being biased in this conversation because it can look at the conversation other points of view right I think one of the greatest drivers for peace on the planet is making sure every mother has their children with the best health and the best education if you are educated if you're healthy if you have opportunity you're not going to want to throw your life away a more peaceful world for us is a world in which we uplift everybody which is what this technology has the ability to do yeah thank you I I think there's one more thing as well it comes to trusted third party when you have a disagreement with someone someone and you have someone trusted between you then a lot of this stuff that can speak both your languages as it were understands both your contexts so again this is the hope now that we have this Universal translator not for language but for stories context and other things we just got to build it right and make sure it's distributed as opposed to centralized and controlled with one specific worldview I'm going to ask a question and jump in here open ai's approach right now uh and you know Elon tweeted the other day that I guess open AI dismissed or Microsoft dismissed their Ethics Committee or some something like that uh is there a general sense in the community of concern uh around that or is that not a valid concern thoughts not tweeting this out and if you don't I well I think one thing I'll just speak for everyone I think we all have to admire what opening ey has done you know the um the work in large models was has really been driven by by open Ai and um kind of as I mentioned before I I shouldn't say that all of AI was on the wrong track but certainly um we were not as focused on large models and the incredible capabilities that we get from these large models if not for open AI um and you know I think so far they've been very responsible in their deployment of the technology um uh I think they've been very thoughtful they've they've invested a lot into research to to make sure that these models are deployed um safely and I think they have a very you know it's hard thing for any organization to do to build such incredible capability that has never existed before and and try to deploy it safely so okay you know I I feel like the easy answer would be to say oh we need to be super concerned about ethics and it's all going self or it's all going poorly um I think we do need to be really concerned about ethics and responsible Ai and to be really pragmatic I think when IMA released his model which fantastic love what he did um I think that IMA had a lot of flak about how could you release thing like that but net net I'm sure some of the models was used for negative use cases but I think imot created massively more value than harmful use cases so I think that we should be very clear eyed about the problems in the harm and do our best to mitigate that and then also be clear eyed about the huge benefits you know that releasing these models and all the wealth and abundance that this also creates and that trade-off is a very tricky one and and I I do see many highly ethical well-meaning AI teams agonizing over that yeah I think I agree with kind of both and you know as aland said amazing technology and it's been really a breakthrough that allows now humans to scale and they've been at the core of it um like I have disagreements with openi uh my core one is just a case of now it's got to the point where this technology could potentially be dangerous like as we scale more we don't know exactly how it works and so I'd love for them to be responsible to to be more open transparent governance transparent kind of procedural stuff and again really adhere to a very strong Charter they don't have to release their models you can have proprietary models like I funded the beta of mid journey and if you look at Mid Journey version 5 it's amazing I said you never have to open source it because it's great for Humanity to be creative you know so not everything needs to be open but we do need to be transparent and careful when stuff gets out of our control in their own AI AGI the artificial general intelligence document they say we believe this technology could be existential and we will treat it as such we believe this technology was surprised when I read that as well so in that case you should be transparent and you should be really transparent about your governance extra more than anyone similar to companies that can affect the environment hey everybody this is Peter a quick break from the episode you know I'm a firm believer that science and technology and how entrepreneurs can change the world is the only real news out there worth consuming I don't watch the crisis News Network I call CNN or Fox and hear every devastating piece of news on the planet I spend my time training my neural net the way I see the World by looking at the incredible breakthroughs in science and technology how entrepreneurs are solving the world's Grand challenges what the breakthroughs are in longevity how exponential Technologies are Transforming Our World so twice a week I put out a Blog one blog is looking at the future of longevity age reversal biotch increasing your health span the other blog looks at exponential Technologies AI 3D printing synthetic biology AR VR blockchain these Technologies are transforming what you as an entrepreneur can do if this is the kind of use you want to learn about and shape your neural Nets with go to demand.com back/ blog and learn more now back to the episode all right we're going to go to zoom next uh Charles uh where are you on the planet and what is your question so I'm in Palm Beach Florida I welcome all of you to come down and visit uh most of us from New York have already left this City and uh we're reestablishing ourselves here this is going to be very specific because Andrew you asked us to be specific so this is actually addressed to you and and perhaps to Peter as well if he wants to chime in but how do you use AI or chat GPT to recommend an investment for your Venture fund you have a lot of inputs a lot of variables in in in my business of investing there are too many variables to do it so if this is a thumbs up thumbs down are we going to invest in this company how do you do it using AI thank you great question so I think CH is amazing I play of it building multiple businesses using it I have to admit I'm not letting chat GPT tell me where to invest at this moment in time maybe it gets much smarter well well I I I won't mind you maybe replace me I'll retire but not yet there um maybe the reality is making an investment decision or decision to build a business is so complex and so multifaceted um I don't know we have the mechanisms to even digitize all of that data right now and even if we had the ability to digitize all that data I think that while very exciting progress is being made to improve the reasoning capabilities of LGE language models I think it'll still take a while to get through the very complex multi-step reasoning that I think investors are using to make decisions by the way I remember one of the founders of Google Ventures who is a friend he's not there anymore told me that Google Ventures actually had built algorithms for making decisions on who they invest based upon where the company was based what your established a whole bunch of other parameters I don't think they're doing that anymore but it's an interesting idea but with that let me go to Mau who's going to have a brilliant question for us M all right uh speaking of positive use cases I'm going to play out a scenario and I want to hear from all three of you if that's possible um I'm a large Health System right I've done the sick care for all of our existence and I'm trying to build a chatbot that supports our nursing staff or our triaging staff and I have all this proprietary data the specific question is what what Foundation models should we start with if my team's working on this project what Foundation models should we start with you do you want to start um yeah you'll be able to use stable LM and stable chat soon right now probably it's going to be um GP Neo jnx on kind of that side of things and that would be the open models that you can use inside your cloud or on Prem uh for the proprietary ones I'm not sure like CLA is Claude available yet privately available privately available Yeah so basically the next generation of language models aren't coming via apis just yet so right now the people that we know in the healthcare industry they use um GPT neox and T plan T5 uh precisely um but there's no real services around that again we're still in the research to engineering phase but in the next couple of months you'll have services that enable you to do this far more seamlessly than before including the uh Microsoft uh gp4 chat Service as well that you can find tune on yeah llama as well uh advention llama although that's problem yeah right um but internal well regardless I think I think one thing that's critical that uh that aad is sort of implicitly mentioning is that um on Health Data you're very likely going to need a model that you can control and own uh because there's a lot of Phi and pii that's going to go into these models um so because of that I think these sort of offerings where uh you know you have to give data to another service are probably not workable so um so I you probably need to take one of these open source models of which Amad named a bunch llama's a new one from meta and and a host of others and and train them up why choosing the right LM language model is important I feel like um a lot of when I think so AI fund we recommend to different companies what language model to start building on top of so these trade-offs with the model size and cost and do you want to fine tune or do you do prompting or con there all these things but I feel like the language model is important but my mind is also going to lot of the decisions you need to make after choosing a language model and the processes execute those as well all right we're going to go to my Deputy son over there hi bear hey Peter what's your question buddy um so I'm a kid so I still go to school and um in school a lot of kids have been cheating or cheating and they've been using Chad GPT to write essays and book reviews and do their math and I guess my question is how do one how do you think like schools should evolve to deal with that or do you think that it's really declared cheating if it's a tool that we're going to be able to use later on brilliant question let's give it up for Bear here you know to me this big question of transparency it's a great question thank you for ask that I feel like um if you use chat GPT and you're willing to openly tell the teacher I'm using it and the teacher says you can use it it that's fully transparent that doesn't feel like cheating to me the thing that is concerning is the students do it but the rules requires at Le them to hide it and that kind of I'm going to do it but I can't tell about it that's that's what doesn't feel very good and kind of the schools are struggling um the debate is people will be to use these tools in the future so we train them to use it or is it a crutch that will limit right a person's growth I have a controversial opinion so I have a four-year-old daughter trying to teach her math and because she's four years old I could stop her from using her calculator I just don't buy her calculator so she's learning to add with her own brain one of the challenges of chat GPT is you know most students using it are in an age where parents and teachers don't have an ability really to easily stop as used and so it does concern me would be a crutch that stops people from delivering certain things um or or because I love to I use calculators but I feel like maybe learning to add with other calculators useful so figuring out what we need to do with chai GV and educational system is something I see a lot of schools struggling with if I wish I had didn't answer yeah I've got so cheating implies it's a contest school should not be a contest you got it so I think most schools are basically Child Care Systems Mi of status games I think schools should embrace this technology and they should really think how can we impart knowledge into individuals how can we impart critical systems cuz again you will never not be without this AI as you grow up and you have to prepare for that awesome uh we have three minutes left I'm going to do the following I'm going to ask you to mention your questions very briefly and we'll do a speed round uh caric what's your question in in 20 seconds or less thanks Peter and gentlemen thank you so much for your brilliant insights quick short what's your question my question is about content creation especially in music and arts okay okay Ellen what's your question very short privacy and AI in uh healthc care information okay Ben what's your question uh should my 15-year-old continue taking Python and shouldn't go to college Okay so we've got uh privacy we've got say again carthic uh creation of music and aration music and arts okay your question answer what you want yes to python yes to college yes to python yes to college okay uh creation of music and art uh music has had it has it stable diffusion moment audio LM and others you will have perfect music models by the end of the year Ellen's question again privacy in terms of healthcare uh uh I think this a really big topic privacy of data when it comes to AI systems I think that this is going to be probably the biggest uh regulatory battle when it comes to AI is what data is allowed to be used to train these models and what are the sort of what consent do you need to get over data and people's data to be able to train models on top of them um I think it's a really big ital topic that we're all going to wrestle with over the next few years and um I uh I I I encourage everyone to get involved if you have an opinion yeah I think um also AI for healthcare has to be auditable and interpretable as a medical device typically so it's going to have to be based on open foundations for that particular use case nice all right Harry and Babs what are your last two questions here real quick chat GPT asks what measures can be taken to counter the misuse of AI by countries like China okay we're going to talk about China in our next session so hold your question for that bab close us out same I was going to ask about the evil empire using the our AI against us we're talking about ethics and all that kind of stuff they're not into ethics okay we're going to be talking about ethics and empathy and US versus China in our very next session I'm going to try and keep us on time let's give it up for these incredible [Music] individuals

Channel: Peter H. Diamandis

Views: 148,516

Keywords: peter diamandis, longevity, xprize, abundance, the future of ai, artificial intelligence, singularity university

Id: _b__L_olCWs

Length: 48min 0sec (2880 seconds)

Published: Thu Apr 20 2023