Leveraging AI

123 | The “Holy Shit” moment of AI, Open AI introduces o1, lessons learned from interviews with Sam Altman, Bill Gates, Reid Hoffman, Mustafa Suleyman, and more AI news from the week ending on September 13

• Isar Meitis • Season 1 • Episode 123

Share episode

Are we on the verge of AI outsmarting us all?

In a bombshell week for AI, OpenAI launched its groundbreaking O1 model, leaving experts and executives alike wondering just how far AI can go. 

In this episode of Leveraging AI, I also share my key takeaways from the MAICON AI Conference, where industry leaders debated the ethical and practical future of AI in business. This episode dives into why C-suite leaders need to be paying close attention to the new capabilities of reasoning models and what this all means for decision-making, scaling, and innovation in your company.

In this session, you’ll discover:

  • How OpenAI’s O1 model could revolutionize reasoning and decision-making in business.
  • Why O1 performs 5x better on complex mathematical tasks and what that means for enterprise AI applications.
  • The implications of AI models that can self-improve with minimal human input — and how to stay ahead of the curve.
  • My personal takeaways from the MAICON AI Conference, including the future of AI-powered strategy and the ethical considerations all leaders must weigh.

About Leveraging AI

If you’ve enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!

Isar Meitis:

Hello and welcome to a weekend news episode of the Leveraging AI podcast, a podcast that shares practical, ethical ways to leverage AI to improve efficiency, grow your business and advance your career. This is Issar Mehti, your host, and we have a special weekend news episode today because of two big things that happened this past week. First of all, I attended the MAICON AI conference in Cleveland and I want to share with you some of my thoughts and learning from that conference. But also OpenAI dropped a bomb, a nuclear bomb, if you want, this week with the release of Strawberry or Q Star or what ended up being called, 01, which is a completely new kind of AI model. And we're going to dive deep into that topic with the background and examples and what exactly it might mean for us in the near future and the longer future. So most of the episode is going to focus on that. And in the end, if we'll have time, I will go very quickly, like a list of all the topics news that happened this week, but most of the episode we're going to dive deep into the release of O1 and my learning from the MAICON conference. on Thursday, September 12th, openAI released a new model called O1 and actually what they released is O1 Preview and O1 Mini and we're going to talk more exactly what that means later on in this episode, but it's a new kind of model called O1. Before we dive into what is O1, what does it do, what's the big deal, and why am I doing a special entire episode about it, I want to go back to some background information. In Q4 of last year, Ilya Satskovar, who was the chief scientist of OpenAI and one of its co founders, have seen the development of a new project within OpenAI that was called QSTAR. QSTAR was rumored to be able to have reasoning beyond anything that any model that we have seen before. And we're going to dive into what that means in a minute, but that led him to believe that open AI is making progress on things that might be dangerous and that is not reported to the board, which led him to go to the board and get Sam Altman fired from the company in November of that year. And the rumor meal talked a lot about what did Ilya see? What was the big deal of Q Star that made him think that this was a leap forward that might be too dangerous to release without additional oversight? As Ilya was in charge of As you probably know, Ilya was in charge of alignment, meaning safety of the models that they're releasing, and he felt that OpenAI is developing it in a faster speed and not with enough oversight for the capabilities of that model. later, this year, like around mid 2024, we started hearing rumors about project strawberry, which became the code name of Q star. And what we've learned is we've learned that strawberry has very powerful reasoning capabilities and that open AI is even using strawberry for reinforcement learning for a new model called Orion. So what does that mean? When these companies are creating these models, all the big models are created in the same way. They build an initial model and then they use what's called an RLHF, reinforcement learning from human feedback. They basically allow the model to provide several different answers to a question and let people rate those answers, which one is a better answer. And this human feedback is what allows them to fine tune the models before they release it to us to make it as relevant and as powerful to be able to serve our needs as humans. So they were able to now use strawberry to do part of that process for a new model called Orion. What does this mean? It means that you need a lot less people and you can build maybe not now, but this is definitely sets up the foundation for a self learning AI platform, which means you can iterate significantly faster, assuming this produces good enough results. And from what we're hearing and from the rumors from open AI, it does so strawberry. was now being talked about something that may work itself into NextChat GPT. There were, again, multiple names for this thing during the process. So we mentioned QSTAR, we mentioned Strawberry. There was a big presentation by the head of OpenAI Japan that mentioned GPT NEXT. So nobody knew exactly when it's coming out or exactly what it's going to look like. But they landed up on the name of O1. Oh, I must admit that for a company that has some of the brightest minds, when it comes to developing AI platforms and with maybe the most powerful AI platform there is out there today, they could potentially think of using AI in order to give better names to their platform. So strawberry or Orion, I think would have been better than O1, but it is what it is might be Orion one. Maybe that's what it stands for, but why one, because we had GPT four before. And so the reason that they're claiming is that this is completely new type of model with reasoning capabilities that are very different from everything we have seen before. And that's why they decided to restart the count with O1 versus continue to GPT 5. So what's the big deal? What does it do? So now I'm quoting from the release of the model notes by OpenAI and they say Performs similarly to PhD students on challenging benchmark tasks in physics, chemistry, and biology. They also mentioned some additional statistics, like on the Mathematic Olympics, which is a mathematic competition that happens every year, GPT 4. 0, so the previous model correctly solved only 13 percent of problems, while the new reasoning model was able to score 83. on that competition. So a huge jump. This is a 5x improvement on solving math problems, high sophisticated math problem at a global math level, scored 83 on that test. And they're also saying that the coding abilities evaluated under coding contest scored in the 89th percentile. On code forces competitions. So on another thing that requires more deeper thinking and planning, it scored in the 89th percentile compared to the best competitors in the world. So if you compare it to the average developer, it probably exceeds their capabilities. Now I want to connect back for a second to Ilya's concerns. And as Ilya ended up leaving the company eventually after Sam got back. He stayed in the background. Nobody knew exactly what was happening with him. And then he left a few months ago to start his own company that is called SSI which stands for safe super intelligence. He raised a billion dollars as his first round of raise, which tells you that a lot of people with a lot of money respect him as an important mind in this field. And we're going to get to Ilya as well, but he had obviously big safety concerns. So one of the things that open AI. Emphasize during the release is that they were testing this model for jailbreaking as well. So on one of the hardest jailbreaking tests, GPT-4 O, which again is the latest model we had before this one scored 22 on a scale from a zero to a hundred, meaning it's relatively easy to jail, break it while the new model oh one preview, the one they released, scored 84. So four times better in protecting itself from jailbreaking. Now, in addition, they shared, as I mentioned to you last week, that they are collaborating with both U. S. and U. K. AI safety institutes and giving them early access to these models to be able to test them and provide feedback before and after the models are being released. as I mentioned, they released two models. One is called O1 Preview, which is a less powerful version of O1 that they will probably release sometime in the future after they get feedback from users. And they also released O1 Mini, which is a faster, cheaper reasoning model that is particularly effective at coding. It is 80 percent cheaper to use O1 mini on the API, and it will provide a cost effective access to developers who wants to use this model, but do not need some of the other capabilities of this model, such as, and I'm quoting, broad world knowledge. So if you're looking to use it just for coding, O1 mini is probably a better way to go. As of right now, users through the chat GPT interface, so if you're just using it through the chat, you'll be limited for 30 messages for 01 preview and 50 messages for 01 mini. If you're using it on the other hand on the API, you just pay the API token fees and it doesn't really matter, you're not limited. So before we dive into what is this whole reasoning thing, I want to share some things that were said by Noam Brown. So those of you who don't know Noam Brown moved from Meta to OpenAI in June of 2023 to do exactly this. So what Noam is known for, he was the one that developed the models that were able to beat the top human players in diplomacy and then in Open Unlimited Poker to be the first machines that can reason through very sophisticated open ended games and still beat the top humans in the world. So Noam shared a long tweet on X, which is weird to say tweet on X. It should have been either tweet on Twitter or X thing on X. I'm not exactly what's the right term right now, but putting terminology aside, he shared his thoughts on the day of the release on this model, and he shared several very interesting statistics and his thoughts about the whole thing. He shared a bunch of graphs, which I will share with you in words. So math competition, basically competing on the most sophisticated math problems that exist in the math Olympics. GPT 4. 0, the previous model scored 13. 4 points. 01 preview, the one that they released to us right now, scored 56. 7, and 01, the full model that is not released yet, scored 83. 3. Huge spread on all the different aspects. On code competitions, called code forces, GPT 4. 0 scored 11 out of 100. 01 preview, the one we have access to right now, scored 62, so six times better, and 01, the full model we still don't have access to, scored 89. That is an incredible score because that means it surpasses the vast majority of developers in the world on complex coding problems. And now, maybe the most amazing statistics is on PhD level science questions. GPT 4. 0, the model that was released before, scored 56. 1. The preview and the full O1 model scored 78 and 78. 3. And then human experts, so actual PhDs on these topics, scored 69. 7. So these models scored 10 percent better than the experts on those specific topics. That is On one hand, amazing and mind blowing and represents a path for a future where these tools can generate on their own new scientific discoveries. On the other hand, it's very scary because it has a lot of negative implications as well. Now the other interesting thing that they shared that this model is not necessarily better than GPT 4. 0 on all tasks. What they shared is a graph that is showing the win rate of the new model, GPT 01 preview versus GPT 40, which is the model that we were using before the release of this, and on personal writing, our GPT 4. 0, the previous model, is actually winning more times. On editing text, it's exactly 50 50, but then on computer programming, data analysis, and mathematical calculations, GPT 01 preview, which is the new model that we just released, is doing significantly better, winning 60 percent of the time on programming and data analysis, and 70 plus percent of the time on mathematical calculations, which If you tried it on the previous models, it was crap. Like you cannot do math if you I'm going to give it anything more sophisticated than one plus one. The other thing he shared that is very interesting is a direct correlation between the model's time of thinking to the accuracy of the results. So the biggest difference, if you're going to start using this model that you're going to see is when you finish typing your prompt, it doesn't start answering. It literally says thinking, and then it thinks about the problem for a while, and only then it summarizes stuff in its head and it shares the results with you. And the graph that they were showing is that there is the time that it took to compute on the x axis and the accuracy of the results on the y axis. And it starts with a very short amount of time being correct 20 percent and it goes all the way up to 80 percent accuracy when you give it more time to think. So the ability to give the model time to think and teaching it the process on how to think, which is I'm going to explain in a minute, is the biggest difference between this model and the previous model. So what does it mean time to think and why the hell does it need time to think? And how is that different than All the GPT models and actually all the large language models we had before. All the large language models were trained on a huge amount of data. Basically every digital piece of data that these companies could get their hands on, whether legally or illegally, I'm not going to dive into this right now, but a huge amount of data and what they know how to do is to pull the right data and summarize it is as a coherent answer based on the prompt that we give it. So it doesn't need to think. It needs to summarize existing information. Why is that? It's because it was always given the answers, right? We give it a huge amounts of answers to a lot of questions and then you ask it a question, it looks for an answer across all the data that it has and it provides the outcome. It does not know how to work the steps through the process in order to get there. And this is exactly the difference to This new model called O1 with its ability to think, reason, and correct itself through the process. So it can actually work step by step like humans do, evaluating every step, whether that step takes it into the right direction. It can go back and re fix and try something else, just like we solve problems. So why does it matter? It matters because so far, when we use these models, we were the reasoning engine. And I teach that in my courses and I work with that on my clients. If you want an LLM, all of them until a couple of days ago, you had to be an LLM. the step by step process on your own, meaning you had to tell it, let's focus on this first, and this would be step one, and I'll take the outcome of this, and let's do step two, and let's re evaluate step two because it wasn't great, and go back and do it again, and try to fix this, and focus on that, and don't do this, and then go to step three, and four, and so on, and you can get amazing results way faster than you could have done on your own, so the capability of these models is still great, But you had to walk them through a step by step process. You want the tangible example. Let's say you want to write a proposal. If just ask it to write a proposal, you will get a proposal of one page. But if you focus with it, step by step on let's evaluate this first step of the proposal and focus on just this initial problem, you will get a page and a half on just that problem, and then you go to step two and write that component and this component, and you're going to end up with a proposal of 20 pages. That's going to be. Very detailed and a lot better than just trying to do it all in one step because you are forcing it to work the step by step process and to fix the different steps. So now I want to give you a couple of examples that people already did and shared about how this new model works. 01 preview is solving problems differently than existing previous models. Ethan Malik, which if you don't know, you should, he is one of the most amazing thinkers about AI right now. he's a professor and he shares lots and lots of content across every platform that you can think of on AI and giving feedback and his thoughts about it. I follow him on LinkedIn as well as I regularly read his amazing newsletter. And his recent newsletter is called Something New on OpenAI's Strawberry and Reasoning. And I'm quoting one segment out of it, and I will quote a few others, but this is from Ethan Malik's newsletter. So if a Malik is saying the following, let's, the AI think through a problem before solving it, this lets it address a very hard problem that require planning and iteration, like novel math or science questions. In fact, it can now beat human PhD experts. in solving extremely hard physics problems. But the test that he passed it through, which I found fascinating because it's, I think, because I believe it's a great representation of real life, is just solving a complex crossword puzzle. Now he, first of all, gave the problem to existing advanced models and they all failed miserably, or like he's saying, they didn't even come close. Why? Because they cannot do the iterative process of understanding the dependency between the steps. In order to solve a crossword puzzle, you need to solve one step and only then evaluate if that actually fits the following steps. And if it doesn't, you need to go back and change it and try again until it works for the next words that cross with it in order to be able to complete the step. Now, The previous models, as I mentioned, failed miserably, but when we gave it to O1 Preview, the model took 108 seconds to think. That's a very long time. Usually it thinks about 10 to 20 seconds before it gives you an answer. And you can actually see the thinking process when you do this. So if you're using this model, you can click the dropdown menu button, like the little carrot and see what the model is actually thinking, what it's doing. And Ethan Malik shared the process that his model was doing, but you could see that it's actually creating and rejecting new ideas that could help it solve this problem. Now, in the beginning, it failed because one of the definitions was very hard and he actually took it too literally and was not able to solve it. but when Ethan gave it the answer to that first definition, the model was able to solve the rest of the puzzle on its own. And then in the summary of this, Ethan Malek said something that I find profound. He said, AI does so much thinking and heavy lifting, churning out complete results that my role as a human partner feels diminished. So I want to go back to why this is a big deal. It's a big deal because Ethan is probably one of the most knowledgeable people on AI progress and capabilities in the world today. And if he feels that Our role as humans in the process working with AIs is being diminished, not eliminated yet, but diminished. It's a very big deal. It means these models can really reason at a very high level, achieving amazing results in stuff that wasn't possible before. I Want to give you another example. Last month at the 2024 Association of Computational Linguistics Conference, The keynote speaker, which I know just as RAO2Z from Twitter, that's his handle, was titled, Can LLMs Reason and Plan? And in it, he showed problems that tripped up all of the existing large language models. But When it was given to O1 preview two days ago, it solved it on the first attempt. So a problem that was built to trip large language models, this new model handled with no problem without any assistant. You want another example of very simplistic, but very interesting problem. Example, I do my AI Friday hangouts every single Friday at 1 p. m. Eastern, by the way, it's open to the public. If you want to join us, it's just a bunch of people who come in and geek about AI and current progress and things we learn and solutions and tools and so on. And it's a lot of fun. One of our regular participants is Sean Bailey, and he shared his experiment with, and he shared his experiment with O1. So something that he likes to show them some of the limitations of the previous models are, he asks the models to tell him how many R's are in the word strawberry, which is obviously a cool pun about the whole strawberry model. And these models cannot do that. Even if you work them step by step and say, okay, let's break it into syllables. How many are in the first syllable, second syllable, third syllable. And it can do that once you break it into syllables and then you ask it, okay, so how many R's are in the word strawberry, and he will get it wrong. Most of the time. solve this every single time correctly. So again, this could be very simplistic, but it shows you the ability of this model to analyze, understand more complex problems and solve them across things that the previous models were not able to do. Why is that important? Where does that put us on the bigger picture? A few months ago, a person from OpenAI leaked a document that was open AI's roadmap to a GI that has five steps. Level one is called chatbots AI with natural conversation language abilities. That's what we had until Thursday. Level two is reasoners aIs with human levels of problem solving across a broad range of topics. That is the step we jumped into. Again, we did not have before. Level three is agents, each aI systems that can take actions independently from human instructions. Basically, they can act on their own and take action on our behalf without our need to give them exact and specific instructions. Level four is innovators, AI that can aid in the innovation of new ideas and contribute to human knowledge and make new scientific discoveries for the future. on its own and level five is organizations AI that is capable of doing all of the work of an organization independently. That could be a committee, that could be a company, that could be potentially governments. So these are the levels that they have defined on the road ahead. We just went from level one to level two. Now I must say something about the, my personal thoughts between going from level one, To two versus going from level two to three. So going from chatbot to reasoners, which is what just happened versus going from reasoners to agents. I have a feeling that from a technological perspective, going from one to two is a huge step forward going from two to three from a technological perspective should not be a big deal because once they can reason, giving them the ability to take action is actually a big deal. Relatively simple. There's several companies are already doing that across multiple, very specific use cases, but once they can reason, you can dramatically broaden the use cases. And what's going to be the limitation of going from step two to three is us, our ability to trust these models to actually take the right actions and not put whatever thing at risk. That thing could be our personal livelihood. It could be personal information. It could be the success of the business. It could be. a million other different things, but our ability to trust that these models will actually take the right action is what's going to limit the deployment of agents. But I feel that while we had to wait almost two years from the release of ChachiPT to the release of the second model, I think it's going to be months before we start seeing very sophisticated advanced agents that can take actions on our behalf on specific tasks with specific guardrails to reduce the safety risk. talking a little more about the speed things are moving. Noam Brown, which I mentioned before, was the guy that was leading the development of this capability in OpenAI from July of 2023. has said himself that he was shocked on how quickly they were able to achieve these results and that he was thinking that it's going to take them significantly longer. So even the people who are driving this train and that are supposed to know more than all of us on how long it's supposed to take, was surprised with how quickly they were able to achieve these results. And that's And now, because they are potentially using these models to train the next models, as I told you, they're using Strawberry to train or help in training the next models, we might get very quickly to a self improving loop where AI needs very little human engagement in order to make better and better versions of itself, which has on one hand, very promising outcomes, on the other hand, very scary outcomes. Now this May have very profound implications on everything we know about scaling laws. And I'm going to address it with quotes from two different people. So Sam Altman mentioned that strawberry allows them to scale in ways that were not possible before. The other person that said something completely independent of this, but when I thought about it and connected the dots, I found it to be highly intriguing is that Ilya Saskover, again, the guy that led to the firing of Sam Altman because of the beginning of this progress, which as I mentioned, just founded a new company called SSI that is supposed to focus on a safe super intelligence. He said that SSI is going to take a completely different approach to scaling than everybody else is doing, meaning not going necessarily down the path of adding more compute and more data in order to create better models. And he's saying, and I'm quoting, everyone neglects to ask, what are we scaling? He said, some people can work really long hours and they'll just go down the same path faster. It's not so much our style. But if you do something different, then it becomes possible for you to do something special. Meaning, Ilya has a way to scale the models in a different way, with a different approach, not by throwing more hours, more compute, and more data into them. Is that aligned with what he learned in his early steps in that process? Maybe. Maybe. Now that might be just my personal interpretation, but I have a feeling that this new step unlocks significantly faster development and scale of new models without throwing billions of dollars on them. So the scaling laws so far basically said, Throw more money, more compute and more data on these models and you get better and better results. And all the people at the realm of the different companies agree that there's no upper limit to that as far as they can see right now. Meaning if. GPT 4 was trained with a hundred million dollars and now the current model is about a billion dollars. And then in a couple of years, they're talking about a ten billion dollar model. And then three years after that, a hundred billion dollar model, where the ten billion dollar model is supposed to be better than, All human PhDs in the world on every topic of PhD, the hundred billion is supposed to be better. It's supposed to be better than noble prize winners on everything. But if there's a new way to scale these models, maybe the 10 billion model, which is the next model than the one we're seeing right now that everybody's working right now could be better than noble prize winners on. Everything. And that could happen two to three years. So will this new kind of models and its ability to train itself will dramatically accelerate the timelines that are already extremely fast? Maybe, I don't know for sure, but I have a feeling that it might. Now to connect the dots between this new model and everything that I shared with you right now about it with my learning from MAICON. So MAICON is an AI conference that's been going on from before ChatGPT. So for this is now the fifth time it's been run. It's run by the amazing people of the Marketing AI Institute, led by Paul Reutzer and Mike Caput. But they have an amazing team of people that has been doing an amazing work. And I, it's my second year of attending this conference. And I actually want to start with the end of the conference. At the end of the conference, Paul Reutzer interviewed two fascinating individual, Adam Brotman and Andy Sack. So Adam Brotman the executive VP for global retail operations at Starbucks. And before that, he was the chief digital officer for Starbucks, the guy who invented and brought to us the Starbucks digital app that allows you to drive baristas crazy with all your needs and specific requirements. But he knows one or two things about digital transformations. And Andy Sack was the managing director of Starbucks. of Techstars, which is one of the biggest incubators in the world. And also after that, an advisor to Satya Nadella at Microsoft, another person that knows a lot about technology. They were requested by Harvard Business Review to write a book about AI. By interviewing the top minds in the field. And they said, yes, but what they did is they decided to release their learnings step by step versus just waiting for the book to be completed with the very logical argument that saying that if they're going to wait for the end of the process, the beginning might not be relevant anymore. So these two fascinating individuals knew very little about AI when they started the process and the initial name of the book was My AI Journey. The reality is they wanted to change the name of the book to the holy shit moment, which Harvard business review didn't agree with, which I understand, but I actually think after hearing what they had to say about their interviews, that would have been a better name. So let's talk about the holy shit moment of the people that they interviewed. And when I say that they interviewed people like Sam Altman, Bill Gates, Reid Hoffman, the CEO of Moderna, Mustafa Suleyman, and so on. So maybe the top minds in the world when it comes to AI development as well as AI implementation. Now, maybe the most known thing out of this book, which was the first chapter that they released, came out of their first interview, which was their interview with Altman. And Brotman and Sack said basically they walked into the interview wanting to learn how they can write better personalized emails. That was their level of understanding of AI when they started this journey. And all Sam wanted to talk about is AGI. Basically he told them he goes to sleep thinking about how to achieve AGI, and he wakes up early in the morning thinking how he can achieve AGI. And they followed with that and said, Okay. What does that mean? And he said, and I'm quoting, when AI will be able to achieve novel scientific breakthroughs on its own. So they followed that and said, okay, but how does that connect to our day to day? How does that impact our world? Let's say our world of marketing. And the way they framed the question was. And I'm quoting, what do you think AGI will mean for us and for consumer brand marketers trying to create ad campaigns and the like to build their companies? And Sam replied the following. Oh, for that, it will mean that 95 percent of what marketers use, agencies, strategists, and creative professionals for today will easily nearly instantly and at almost no cost be handled by the AI and the AI will likely be able to test the creative against real or synthetic customer focus groups for predicting results and optimizing again, all free instant and nearly perfect images, videos, campaign ideas. No problem. is maybe one of the most profound quotes about what AGI means I ever heard, and potentially anybody ever heard. Now they obviously went ahead and asked you, okay, so when is this thing coming? And the response was five years, give or take, maybe slightly longer. But that was Almost a year ago in this year that we're developing strawberry and now they're using strawberry to train Orion. And that's why the guy that's developing strawberry, Noam Brown, as I mentioned before, mentioned that he was that they were able to develop it much faster than he thought. It is going to take now, you may not be a marketer. So you may think to yourself, Oh, this is just for marketing. It's obvious. Marketers are doing this quicker and it's just creating content and so forth. But remember the question was about marketing. I am pretty sure they would have gotten the same answer. If they would have asked the question about. Any job that we do that doesn't require moving stuff around. So the same answer would have been given to creating business strategy, running teams, making business decisions, finance, sales, operations, et cetera. I'm pretty sure the answer would have been exactly the same for all of the above. So that was just their first interview. The second interview was with Reed Hoffman. And Reed Hoffman, those of you who don't know, is one of the most successful serial entrepreneur from Silicon Valley, was involved in some very big startups. He's also one of the original co founders of OpenAI. And the co founder of LinkedIn and a lot of other ventures. And he's been involved with AI for many years. And in the interview with him, what they were shocked about the most is that he was talking about that in two to three years. All of our interaction with knowledge would be done through AI agents, meaning we will not visit websites, we will not search the web, we will not go to communicate with different channels. We will actually have our AI agents do all that work for us and give us synthesized results outcome. Now, what does that mean? It means that the world we know today is going to be very different, even on simple things. if you have a company website, if in two to three years, no human is going to go to that website, you probably need to rebuild the website. It needs to be structured, not in a way that will satisfy humans, but in a way that will make it the easiest way for agents to get the relevant data. That's just one example, but there are many other profound aspects to what I just said. The third person they interviewed was Bill Gates, and Bill Gates shared two very interesting aspects with them, going back to the holy shit moment. And so Bill Gates shared with them that the thing that was his first holy shit moment in his career was the Xerox Alto computer, which was built by him. More or less if you want the godfather machine of the PCs and he's saying seeing the altocomputer is what planted the seed in his head the PCs will take over the world that led him to start DOS and then Microsoft and then making it the One of the most successful companies in history. So that first moment in his career that created Microsoft and more or less everything we know around us today started with that moment. He's claiming the moment that he saw Chachi PT was bigger than life. and more significant than his auto computer moment. Now, he's saying that while the auto computer started a revolution that he was leading and on the forefront of took decades. Right now, Chachapiti will happen significantly faster because all the pipes that All the infrastructure that we've been laying for the past decades with cloud computing, fast communication, internet, and so on, will just allow it to spread significantly faster and evolve significantly faster. The other interesting point is that he said that while this was cool, he was not completely impressed with the capabilities. And he basically said, I will be impressed when he passes a biology AP exam while really understanding the material. And then Chachupiti 4 came out and he was able to do it with no problem. So again, one of the most advanced minds in understanding technology in our world was shocked when the ability to solve an AP exam happened within almost immediately after he made that statement, when he was thinking it's going to take three to five years. As I mentioned, they also interviewed people like Mustafa Suleyman, who is the, one of the co founders of DeepMind, and he's also the founder, who then left DeepMind and founded Inflection, and now his team and him was brought into DeepMind. Microsoft and is currently running consumer AI at Microsoft. Again, one of the top AI minds in the world. He shares similar things of how much he was shocked when this tool came out. And they also interviewed the CEO of Moderna and even Ethan Malik that we mentioned earlier. The interesting thing, and if you want the silver lining with all of those people, is they all had their holy shit moment when they saw ChatGPT and what it can do. The other thing that is very interesting and very scary at the same time is that they, none of them really understands how exactly these models work or why they work. They don't exactly understand how or why they behave the way they behave. And most importantly, they do not know. And they show that very clearly in the interviews. What will the impact of this will be on the individuals, businesses, the economy, the society, and so on. So what does that teach us? It teaches us that this entire industry is moving very fast, led by a huge amount of investment by some of the smartest people on the planet. And while they're making these amazing new discoveries and capabilities, they do not know where this is taking us as a society, which to me is really scary. That being said, it's not going to stop. If anything, as I mentioned before, it's going to continue accelerating. So what can we do? The only thing we can do is to stay as educated as possible, learn as much as we can, sound our voices as loud as possible, and try to help navigate this so we can enjoy the benefits of this as much as possible. As I mentioned, new novel scientific discoveries, stopping and reversing global warming, stopping. Terminal diseases and so on. there's multiple ways where this could be very helpful. The flip side is this could lead to some catastrophic outcomes that maybe nobody can anticipate. So I'm somewhat excited about the new collaboration between companies and collaborations between companies and governments to put some additional guardrails on these developments. In addition to all of this, the MAICON conference obviously shared multiple people's experience with AI, which showed amazing, incredible use cases of how they're using AI today across multiple aspects of the business, from data analysis to strategy, through data analysis, business strategy. Ideation and yes, also content creation combined with some amazing automation capabilities and also shared thought processes around legal aspects of this and ethical aspects of this. So great conference overall, and I highly recommend you attend it next year. And now, as I mentioned, a very quick recap of the news. So obviously the biggest news we mentioned, OpenAI released these new models that now should be available to everyone. It took them a while to roll it out, but now you should have it both on the API and on your chat platform. still in OpenAI, Another OpenAI leading researcher, Alexis Konu, who I'm not sure I'm pronouncing his last name correctly, just left OpenAI to start a new and his own company. Going from OpenAI to bigger news, but related to everything we talked about before, there was a bunch of. Leading minds in the AI world lead meeting with the White House this week. So NVIDIA CEO, Jensen Hong, Sam Altman, Dario Amadeo, the CEO of Anthropic, Google's president, Ruth Porat, Microsoft president, Brad Smith, Amazon Web Services CEO, Matt Garman. And then the Commerce Secretary and Energy Secretary were all in the discussions, and they were discussing the impacts of AI on energy usage, on data center capacity, semiconductor manufacturing, and grid capacity. Basically, what are the needs of U. S. infrastructure in order to support future AI development? I'm very excited to hear that these things are happening because it will potentially allow, again, collaboration between all these different individuals as well as the government in order to, on one hand, make sure that we stay ahead of some adversaries with AI development while hopefully keeping it as safe as possible. On the topic of safety and regulation, we spoke several times before about SB 1047, the new regulation that has passed Senate in California But has not signed by it, but has not signed into law by the governor. So multiple employees of leading AI companies, over a hundred people from Google and Anthropic and DeepMind and Meta and OpenAI has signed a letter calling the governor to actually sign this into a bill saying that it's required, that it will be government oversight into the ways That these models are being developed and deployed. Since I mentioned Google several times, Google unveiled Data Gamma, which is a new variation of their Gamma open source models. And it's the world's first open source model that is built to address hallucinations. The way they're doing it is they're grounding the large language models results in real world statistical data from Google's data commons. Google's data commons is a publicly available knowledge graph with over 240 billion data points. So basically it allows the large language model to go and check its facts against this huge data set before it provides answers. I'm personally very excited about this because I think hallucinations is one of the biggest problems of these models right now when it comes to to day business usage in a consistent way. A cool thing that Google also introduced this week, they introduced AI generated podcasts From notes in notebook LM. So notebook LM is your note taking platform that allows you to bring in links off different articles and so on. And you can buy a click of a button. Now turn this into a podcast where two AI quote unquote hosts summarize and discuss the material in the article or the research paper that you are going to provide it very, very cool. I didn't test it out myself, but I saw several different examples online and I was shocked as a podcaster, how good it is. Another really interesting, big piece of news this week comes from Mistral. So we talked about Mistral several times before. They are a large French company that is developing and releasing open source models, open source AI models, and they just released Pixtral 12b, which is the first open source model that is multi modal, meaning it knows how to process both text and code images. So a big step forward for the open source world. The first, but I'm sure not the last, there were many other pieces of news this week that were very interesting, which you can find in our newsletter. So if you want to learn more, just sign up for that, there's going to be a link for that in the show notes. So you can just open it. And if you're already opening your screen in order to look at the show notes and find the link, I would really appreciate it. If you give us a review, if you're either on Apple podcast or Spotify, that helps other people find the podcast and helps us spread the word. AI literacy for additional people, which as I mentioned before, is the only thing we can do right now to be better prepared for the future ahead. And also while you're there, click the share button and share it with a few people that you think can benefit from this podcast. I would really appreciate it. I will share one more thing from the conference. it was really fun to meet many listeners. So I obviously know very few people that listen to the podcast out of the. Tens of thousands that listened to it. And it was really fun for me to meet you in person for all of you who came and said hello and say, Hey, we'll listen to your podcast. We'll have your show. And so on. That was really heartwarming as a podcaster. It's a very lonely thing, right? I'm sitting here in a studio recording in front of a microphone and a bunch of lights, but I don't know that people will listen. And so I really appreciate the people who came back to me and said, hello, and said that they liked the podcast That puts a lot of fuel in my systems to continue doing what I'm doing for you. And same thing with the people who write to me on LinkedIn. I really appreciate that as well. We'll be back on Tuesday with another how to episode with another expert. And this coming Thursday, we'll be out with another fascinating live episode. This time talking about how to use Gemini in your day to day work. If you are in the Google universe and on Friday, as always Friday, AI hangouts that is open to the public, that you can join us and jump in the conversation with a lot of people who are figuring out the path along the process of AI and business, that's it. Have an amazing rest of your weekend.

People on this episode

Head Shot

Isar Meitis

Host