AI Panel Discussion W/ Emad Mostaque, Ray Kurzweil, Mo Gawdat & Tristan Harris | EP #96

Wouldn't it be great to manage your portfolio on one platform? Well, now you can trade all in one place on Robinhood. That means you can trade individual stocks and ETFs and also buy and sell crypto using seriously powerful and intuitive tools at one of the lowest costs on average without needing to manage multiple apps. Robinhood makes withdrawing and depositing crypto seamless. Send crypto to your Robinhood account or send crypto from your Robinhood account to other wallets without deposit or withdrawal fees from Robinhood. Trade all in one place. Get started now on Robinhood Trading. Crypto involves significant risk. Crypto trading is offered through an account with Robinhood Crypto llc. Robinhood Crypto is licensed to engage in virtual currency business activity by the New York State Department of Financial Services. Crypto held through Robinhood Crypto is not FDIC insured or SIPC Protected network fees may apply to crypto transfers. Crypto transfers may not be available to all customers. Investing involves risk including loss of principal securities. Trading is offered through an account with Robinhood Financial LLC member sipc, a registered broker dealer.

You're about to make a trade.

Which u do you listen to? Is it get optioning those options or let's do a little research. Learn more@finra.org TradeSmart how do you control.

A technology that runs on laptops and smartphones and is moving in a ubiquitous, decentralized training direction?

These technologies can help us appreciate the world better, appreciate each other's stories, and tell the stories of each other.

This is not a question of technology, this is a question of human fear. The fear here is that the other party is going to beat me.

There's so many ways that we can make AI a part of the solution, but we have to look at all the ways it's been misincentivized so far.

This is a moment where we have to remove the shackles of technology. To be very open and honest, I think we need to use technology in our benefit but not let it use us.

Let me kick it off here Imad. How do you feel about sentience? Do you feel that these models are becoming sentient or conscious?

I don't think so yet. I think they're still like weights and ASCII files that are like a sieve. You put something in and something else comes out. However, as you combine them with humans working on them and in agent based things, I think you'll see new forms of emergent intelligence and potentially sentience from that.

Mo. How about you? Do you imagine we're going to See sentience.

Define sentience. I mean, if you, if you think of us as a sentient, then that's a question. If you think a tree is sentient, that's a different question. If you think the universe is sentient, that's a different question. I mean, how much difference are we, you know, to. To ASCII code? And you know, we are DNA code in a very interesting way.

I define it as. I define it as human level sentience. And then I'm gonna go to Andy and I'll go to Salim next.

Yes, what, what makes us human is that we are. We have free will. They have free will. What makes us human is that we can procreate. They can procreate. What makes us human is that we have emotions. They have emotions. If you define emotions as fear is an understanding that a moment in the future is a little less safe than a moment right now, then they have the same logic. It's a very complex question, but I will openly say it's an irrelevant question. Because if they simulate sentience, then that's the way we should treat them.

Interesting, Annie. And I'm also looking forward to this top question in slido. Annie, what you got? Hey, everyone. Ray, it's great to see you back here in person. The question that I have. We just came back from the Beneficial AGI Conference in Panama and we were talking about humanity and where the fear is coming from. Do you believe. How can more of humanity get involved in the creation of superintelligence? Super AGI 1 and 2. Do you believe this super intelligent super AGI truly belongs to the world? And if so, should companies and governments own it or should it really belong to humanity?

Do you want to.

That's a huge question. A mentor of mine said, with AI, it's like, which entity would you trust to be a trillion times more powerful than it is? Would you trust a corporation to be a trillion times more powerful? Would you trust a government to be a trillion times more powerful? And then what does it mean for something to be a trillion times more powerful and then have some kind of democratic accountability to the will of the people? You know, in our presentation we tried to focus on. We all, we all want those super intelligent benefits. And one of the challenges in the way that AIs are currently trained is the benefits are directly connected to, to the risks. Right now we want the donuts, we don't want the diabetes. The thing that gets you the perfect AI biology tutor for every middle schooler is inseparable from the thing that knows how to tell you about how to make biological weapons. If you want the first, you can't get the other one. The thing that makes cool AI art is inseparable from also knowing how to make child sexual abuse material and removing out the controls. And so I think that's, that's one of the key challenges. How do we get clear about what we don't want so that we can steer towards the world that we do and then have the controls in place as best as we can? And the problem is, how do you control a technology that runs on laptops and smartphones and is moving in a ubiquitous, decentralized training direction?

I'm going to go to Slido next, then Salim and then Zoom. Is there any discussion happening around mixing spirituality with AI? Can AI connect with God? Anybody want to take that on? Imad?

So as a former theology student.

Yes, that's right.

So I think it's interesting because what this AI does is we take hundreds of thousands of gigabytes of images or trillions of words and compress it down to a few gigabytes. And that can't be compression of data, it's compression of context. So when looking at Islamic theology, for example, the 99 names of Alice an Tala, like fasting, is a representation of the divine aspect of samadhiyat, or freedom from want. When you compress this technology down, you can say in Dall? E 3, make it happier. And it understands a concept of happiness from being trained on the human corpus of happiness. So when I see this technology, what I see is certain aspects of the divine through the context window shifting of that. Because what we're trying to do with a lot of spirituality is get beyond where we are right now and understand reflections of something a bit beyond. And that's what a latent space is for me. It is the concept of happiness, the concept of creativity, the concept of fulfillment. And these technologies can help us appreciate the world better, appreciate each other's stories and tell the stories of each other. Just like it says, we made you diverse so you could better understand each other.

Over the years, I've experimented with many intermittent fasting programs. The truth is I've given up on intermittent fasting as I've seen no real benefit when it comes to longevity. But this changed when I discovered something called Prolon's five day fasting nutrition program. It harnesses the process of autophagy. This is a cellular recycling process that revitalizes your body at a molecular level. And just one cycle of the five day Prolon fasting nutrition program can support healthy aging, fat focused weight loss, improved energy levels and more. It's a painless process and I've been doing it twice a year for the last year. You can get a 15% off on your order when you go to my special URL. Go to prolonlife.com p r o L-O-N l I f e.com moonshot get started on your longevity journey with Prolon today. Now back to the episode. All right, we go to Saleem and then we go to go Stacey Hale on Zoom Saleem yeah, A clarification question. A member of our faculty from for.

Each of you, which is a beef.

That I've had for a while.

We worry about AI getting smarter than human beings and the question I have is what do we mean by smarter? Because we have the IQ test, which measures the speed of thought processing and the ability to match concepts across frameworks. We don't measure emotional intelligence, spatial intelligence, the eastern concept of presence and awareness. So what do you mean by smarter? Is my big question that I've not.

Had at some point. Good answer to what's smarter for you, Ray.

Well, I'm holding this because I believe almost everybody here has one and it definitely makes us smarter. We know all kinds of things that we didn't know even five years ago. I generally ask people when I speak to them how many people have their smartphone. Almost Everybody says yes. Five years ago it was maybe 50, 50, 10 years ago. Almost nobody had it. And so in a very small period of time we've enhanced our intelligence by carrying the best we have of machine intelligence and it's really merging with us. And also machine intelligence is learning from us. So it's really part of who we are.

Outsourced. Stacy, I just want to make a quick follow up comment.

The best comment I think we've ever heard about this comes from you, Ray. When we were talking about consciousness and you said language is a very thin pipe to discuss topics as rich as this. So thank you for that comment.

Thank you, Saleem. Stacy, so when we were looking at the four quadrants earlier and down at.

The bottom, the two areas where we have the villains and all the different things that could happen there, is there a way we can leverage AI to mitigate those things from happening? Obviously we talked about child abuse materials, but all of the different things that were happening here. Can we leverage AI to help us with that?

Totally.

And that's the invitation that we want to make to the whole world is this is our urgent challenge. We have, you know, when the framers of the Constitution, we're thinking about, how do we prevent untrustworthy centralized power and governance? They were trying to do that with old mechanisms as just law and institutions. And now Larry Lessig wrote in his book, you know, code is law. And we now need to bring new technological answers in addition to sociological answers to how do we check untrustworthy power. We have things like zero knowledge proofs. We can also use AI, for example, to find consensus opinions in democracies rather than social media, which is sorting for what is a cultural fault line of inflammation and then incentivizing people to tweet, to drive up outrage and anger, to divide society. On every fault line, you could have AI that tries to find the synthesis of different perspectives and rank for the unlikely synthesis. When do clusters of users who typically tweet different things and are angry about different things, when is there actually something that they tend to agree on that's a shared value? And what if we were sorting by unlikely consensus and sorting by division and outrageous? There's so many ways that we can make AI a part of the solution, but we have to look at all the ways it's been misincentivized so far.

I love this question on slido that will go to Anusha. It says here, what are the key skills that us teenagers, I'm guessing it's a teen table here, can learn to excel in the age of AI? I think one of the most important questions, so teens, listen up to the answer. Who wants to IMAD or mo? Do you want to? So we've got teenagers here. What should they be doing?

I think that, you know, you've seen the code models come out, you've seen the art models come out. But I believe that they can build content, they can build boilerplates, they can't do art yet they can't do actual proper programming. A large part of what you have is the leadership, organization and systems thinking is the most important thing you can have. Because again, you must think, what if I had that new continent with 100 billion talented people? What will I do to make my processes better, to make the world better, to make my life and other elements better? I think that's the best mental model to think about what you should have. And those skills are the things you should have, problem solving skills, and you should also use it every single day. I think one of the final questions on our thing was, what should we do? Everybody in all your organizations should use one of these technologies at least half an hour a week and Report back what they find, and you'll find some incredible.

How many folks here are using Gemini or ChatGPT at least half an hour a week, Right? I mean, my comment to my team is I want it open during every board meeting, during every staff meeting. I want it as a tab open on my machine all the time. There are so many times where I stop myself from doing something, go, oh, my God, that's just going to be so much better for me to do on Gemini. And it literally is like, holy shit. Moments where I would have wasted half an hour, an hour of my time. It's boom, it's done.

What teenagers should do is actually answer these questions. How can we relieve ourselves of these risks? And they'll be just as good as adults. It's not like you need to be of a certain age to be able to answer these questions.

Great point, Ray. Everybody, I want to take a short break from our episode to talk about a company that's very important to me could actually save your life or the life of someone that you love. The company is called Fountain Life, and it's a company I started years ago with Tony Robbins and a group of very talented physicians. You know, most of us don't actually know what's going on inside our body. We're all optimists until that day when you have a pain in your side. You go to the physician in the emergency room and they say, listen, I'm sorry to tell you this, but you have this stage three or four going on and you know, it didn't start that morning. It probably was a problem that's been going on for some time. But because we never look, we don't find out. So what we built at Fountain Life was the world's most advanced diagnostic centers. We have four across the US today, and we're building 20 around the world. These centers give you a full body mri, a brain, a brain vasculature, an AI enabled coronary CT looking for soft plaque DEXA scan, a grail blood cancer test, a full executive blood workup. It's the most advanced workup you'll ever receive. 150 gigabytes of data that then go to our AIs and our physicians to find any disease at the very beginning when it's solvable. You're going to find out eventually. You might as well find out when you can take action. Fountain Life also has an entire side of therapeutics. We look around the world for the most advanced therapeutics that can add 10, 20 healthy years to your life. And we provide them to you at our centers. So if this is of interest to you, please go and check it out. Go to fountainlife.com Peter when Tony and I wrote our New York Times bestseller LifeForce, we had 30,000 people reached out to us from for Fountain Life memberships. If you go to fountainlife.com Peter will put you to the top of the list. Really, it's something that is for me, one of the most important things I offer my Entire family, the CEOs of my companies, my friends. It's a chance to really add decades onto our healthy lifespans. Go to fountainlife.com Peter it's one of the most important things I can offer to you as one of my listeners. All right, let's go back to our episode. Anousha Very proud to have Anush Ansari, the CEO of the xprize, here with us today.

Thank you. I agree that AI is inevitable and it's moving very fast. Also like what Mo said about changes in society and a shift where as parents we become responsible parents. However, this is not something that's going to happen in anytime soon. As we can see the state of our world. I would even question human intelligence as being intelligent. So with that given and the fact that these models are using resources of our planet, energy and water resources, even more than human beings, and if we give them more power, more control, there will be a resource contention which causes the wars that we fight as human beings. And now we're adding very powerful machines to fight this wars with us or against us. How can we create a pathway where we slow down, we can't stop it, but we can slow it down and give ourselves time to figure out the answers to this question. Because one thing I heard today is that all of us were saying we don't know and we don't know how to solve it. But is there a way is taxation of all the companies and things that go into powering these models on energy and water a way to slow it down so we can actually find time to find answers to our questions?

Mo, can this be slowed down? Is there a velocity knob, an on off switch?

No, no, no is my answer. Unfortunately, once again, as I mentioned in the in the first initiative, first inevitable, this is not a question of technology. This is a question human fear. The most active motivator humans can ever engage with is fear. And the fear here is that the other party is going to beat me. And the second biggest motivator in our world is greed. And there is a trillion dollar pie to be gained from those who can create the next big thing. And so I would probably say it's not a wise thing to hope for a slowdown. I think the right thing to do is to embrace that intelligence in itself is a good thing and just direct it in the right direction.

We saw this with Google, right? I mean, Google had the tech. The first ethos of this was don't put it out on the World Wide Web and don't give it permission to code itself if you don't want this to go, you know, to dispatch to the right. And unfortunately, their hand got forced. And you heard Sergey recently say, put it out too early. There's a lot of pressures there.

I just say, I think one of the solutions to this is actually open because our releasing of the image model meant that dozens of people didn't need to train their own. And we optimized it for the edge so it wouldn't require these massive compute requirements because you have to send them to university once and then you can put them everywhere. So build open models so other people don't have to, and then optimize them for the edge so it doesn't use up all the resources rather than gigantic GPUs. And I think that can mitigate it somehow.

Nice. We're going to go to Karim on Zoom. Karim, where are you and what's your question? Yes, hi, I'm Karim from Casablanca, Morocco. Well, my question is to each and any one of you first, thank you for all your. Your presentation. They were awesome. And I like the fact that some of you are not pretty close or agree with each other. My question is, as human being, how should we handle our identity knowing that we have AI today and quantum computing coming at a very fast pace, which will also completely cope with the curve that Ray showed us today. How should we, how should we position ourselves in front of this tremendous intelligence? We're going to hear about that a little bit later and tomorrow. Yes. If you thought things were moving fast, hold on, we're about to hit warp speed. Who wants to take that one?

Questions of human identity in the face of all this are super hard. A friend of ours says that AI is like the 24th century crashing down on the 21st century. And if you think about that much change happening that quickly, like imagine if 20th century or 21st century technology was crashing down on 16th century governance, right? The king assembles all of his advisors, send in the knights to do something about wi fi and video games. And like, what are you going to do? And you know, I think the thing that is universal. To bridge on the question that was asked before about slowing it down is a lot of the people that, you know, we talk to, people in the AI safety and AI risk community quite a bit. And what everyone seems to be able to agree on is that a lot more comfortable if this change was happening instead over two years, but over 20 years. And Jeff Bezos said that society does adapt to new technology, but it needs time for its immune system to come. And I think one thing to think about as a principle is how can the immune system of a society have greater compute processing power than the rate of evolution of the mutation of threats? And right now, the mutation of threats has greater compute behind it than the immune system of our society. So we need to, I think, correct for those, those asymmetries.

The question of identity is, is a very dangerous question. And I'm, I'm being philosophical here, but, you know, we've, we've just ended millennia of gender identity, you know, discrimination, if you want. The idea of defining an identity necessarily leads us to believe that one identity is either superior or inferior to the other. And that by definition might mean in the long term that we try to treat AI differently, even if it has rights. I think the idea here is to try and welcome AI, as I always say, as our artificially intelligent infants. I think this is a very big stretch at the moment, but more and more, I mean, your work in the morning today, Peter, while you're talking to all of those bots, is just the DOS level, the very entry level of how they will fit in our society. So we might as well welcome them rather than identify them and discriminate against them.

It's going to be merged with us. We already carry around a lot of digital intelligence today, and that's actually how this will be manifest, merging it with ourselves.

Did you know that your microbiome is composed of trillions of bacteria, virus, viruses and microbes, and that they play a critical role in your health? Research has increasingly shown that microbiomes impact not just digestion, but a wide range of health conditions, including digestive disorders, from IBS to Crohn's disease, metabolic disorders, from obesity to type 2 diabetes, autoimmune disease like rheumatoid arthritis and multiple sclerosis, mental health conditions like depression and anxiety, and cardiovascular disease. VIOME has a product I've been using for years called Full Body Intelligence, which collects just a few drops of your blood, saliva and stool and can tell you so much about your health. They've tested over 700,000 individuals and used their AI models to deliver key critical guidelines and insights about their members health like what foods you should eat, what foods you shouldn't eat, what supplements or probiotics to take, as well as your biological age and other deep health insights. And as a result of the recommendations that VIOME has made to their members, the results have been stellar. As reported in the American Journal of Lifestyle medicine, after just six months, members reported the a 36% reduction in depression, a 40% reduction in anxiety, a 30% reduction in diabetes and a 48% reduction in IBS. Listen, I've been using Viome for three years. I know that my oral and gut health is absolutely critical to me. It's one of my personal top areas of focus. Best of all, VIOME is affordable, which is part of my mission to democratize healthcare. If you want to join me on this journey and get 20% off the full body intelligence test, go to viome.com Peter when it comes to your health, knowledge is power. Again, that's viome.com Peter Paul and then we're going to go to the slido question.

Hello, my name is Paul Abreu, I'm a programmer. I've been a member since July and I want to first thank Peter for organizing. This is a very, very unique, unique community.

Very.

Thought provoking and talking to like minded people. My two questions are for one question for MO and one question for imad. The question for IMAD is do you agree with Yan yakun's idea that LLMs can't take us to AGI because of the reason that he said? Well, According to him, LLMs lack planning and lack imagination and memory. What are your thoughts on that?

I think LLMs are just one part of the memory, like type 1 versus type 2 thinking. And we have systems that of planning and coordination and others. And again, just like organizations, we have people of different specialist types from the strategic people to the tactical people. So any AGI will be a composite system as opposed to an individual one.

So LLMs themselves can't take us to AGI.

Okay.

Be one part of it.

Okay. And this is a question for.

I mean we shouldn't call them LLMs because they already deal with far more than language, pictures, cures for disease. Lots of things that we don't consider language are already being dealt with. So we should call them large object models.

If you don't mind, I want to try and sneak in a couple more questions.

One last question.

Okay, make it quick.

Okay, I'll make it quick. So this question for movies. I know that you emphasize that with all of this the most important thing is human connection and the Internet is great for making human connections. But when all of the things, we're all flooded with bots and fake things, how can we maintain the advantage of the Internet yet still have actual, genuine, genuine connections?

I think the answer is very straightforward. Leave the Internet and go out. Honestly, it seems to me very, very clear that yeah, this is a moment where we have to remove the shackles of technology. To be very open and honest, I think we need to use technology in our benefit but not let it use us.

I'm going to combine these two questions up here. So the first one is how do I invest in AI? Given that IMOD is right with three year future visibility, it's how do you invest when it's moving so damn fast? What companies are building the infrastructure? And then another question, Imad, that you might speak to is the effect of AI is having on our education system because you're extremely passionate about reinventing our education system, as am I. I mean I think the bot museum that Steve Brown is building, as you've seen those incredible bots of Aristotle and it's amazing what he's done, right? So if I want to learn about ancient Greece, I can go have a conversation with Aristotle or Socrates or Plato and not have to read it. It's amazing in that regard. Imad, what do you think on those two questions?

Yes, I think you use the mental model of where would 100,000 graduates benefit this business or other things later on this week I'm launching a democratized AI fund as an angel list rolling fund. That's zero management fees, zero performance fees to invest in the best entrepreneurs in this space that anyone can kind of participate in. I've given 20 million supercomputer hours over the last few years to people like that. But outside of something like that, just focus on where do graduates make the difference? And then when it comes to education, again, if your student had, if your, your child had so many tutors, education is basically a petri dish mixed with social status game, you know, mixed with just childcare at the moment. The whole purpose of education should be to allow them to achieve their potential. And so you need to think about how to use this AI to make them always believe they have agency and interconnection. I think those are the two most important things when you're looking at this AI.

Amazing Mo. Do you want to comment on education?

Yeah, I think education is a technology, you know, putting a lot of people in one place and filling their heads with stuff was the old technology AI, basically. And Ray will always speak about this as it integrates more and more with us. It brings a lot of the elements of what we used to do in school, and accordingly, we want to embrace that change very, very strongly and basically ask the world to move into problem solving, human connection, you know, other skills like telling what's fake and what's true and so on. These are the skills that are going to be needed in the future. The rest is going to be provided to us directly through AI.

Awesome. Ladies and gentlemen, please give it up for Imod, Tristan, Mo and Ray.