OpenAI Explains Why NOBODY Is Ready For Whats Coming?

Now, recently, the person who is the head of mission alignment at OpenAI actually stated something that was quite remarkable. He actually stated that the world isn't grappling enough with the seriousness of AI and how it will upend or negate a lot of the assumptions many seemingly robust equilibria are based on. And I actually agree with this statement quite a lot. On this channel you might see a lot of AI news, but one of the topics that I did cover quite a lot was of course, the post AGI Economics series where actually trying to speak about how the world is changing right before our eyes. And it's just that most people can't see it or won't pay attention to it. And basically this tweet is suggesting the global discourse and preparation regarding AI is pretty insignificant compared to the scale of the implications. And when we look at people, governments and institutions, they're not really fully acknowledging or even addressing the profound impacts that this is going to have on society, economies and, you know, many different technologies. And when we think about it, it's going to be disrupting a lot of the foundational beliefs or norms that we currently take for granted. Things like jobs, education. I mean, there's just so many things that we can get into, social dynamics, global markets. It's pretty crazy stuff that I will dive into in this video. Now, I'm showing you guys this piece of a blog, but the blog is actually from Sam Altman. Sam Altman is Of course the CEO of OpenAI, but I'm showing you this to show you guys that it's not just one person at OpenAI that thinks this is company the case. Sam Altman also agrees that, you know, it's quite important to actually prepare for the future. He basically talks about how, you know, the technological revolution is unstoppable and a recursive loop of innovation as these smart machines themselves help us make smarter machines, will accelerate the revolution's pace, and that there's going to be three crucial consequences that will follow. Now, the reason that this is important is because if we actually take a look at the entire blog, which Sam Altman wrote in 2020 21, it's actually remarkably accurate with as to what he predicts. But he talks broadly about the future and a lot of this stuff has already happened. Now, one of the first things that he talks about is the fact that this revolution will create phenomenal wealth and the price of many kinds of labor, which is essentially, you know, driving down the cost of goods and services will essentially fall towards zero once sufficiently powerful. AI joins the workforce. And I want you guys to pay attention to two things here. Number one being AI joining the workforce. Okay? Keywords are joins the workforce. Because literally, in another blog post a couple of days ago, Sam Altman actually said that we believe in 2025 we may see the first AI agents join the workforce and materially change the output of companies. I covered this entire thing in another video. But it is important that, you know his predictions are coming true. So it's important to look back and, and see what was said. Now, the second point that I want you to focus on from the first point, because there were two things, is the fact that, you know, the price of many kinds of labor, which drives down the cost of goods and services, will essentially fall to zero very soon. Now, a phrase that I've seen echoing around the AI community is one that is called the Intelligence 2 cheap 2 meter. And this is basically the fact that when you have AI systems that are able to replicate human intelligence for a fraction of the cost, things start to get really weird. And you can see that Logan Kilpatrick, the person working on AGI and the AI products at Google, actually said that if you are not planning for the price of intelligence to go to zero, the next three to five years are going to be incredibly disruptive to your business and to life. This is the main idea for the rest of the decade. You might want to buckle up. Now, broadly speaking, what he means by this is that when we actually take a look at how AI systems have progressed every single year, the intelligence price versus the tokens, it actually gets a lot cheaper. So right here you can see that, you know, the previous version of GPT4, which wasn't that good compared to the version that we have now, and if we actually look at the cost per token, it's a dramatic decrease in that cost. And the point here is that, like, in the future, we're going to have really powerful AI systems and they're going to be borderline free to run, which basically means that if we have sufficiently powerful AI that is smarter than most humans, can run agentically on a variety of different computers, then of course that is going to disrupt a variety of different things. And this is something that has continued to happen time and time again. And like I said before, this is what these companies are moving to, okay, towards intelligence too cheap to meter. This is something that Sam Altman has tweeted and run has tweeted, which is of course a member of technical staff at OpenAI. Now, remember how I spoke about you know, the fact that like these models are getting cheaper, but the crazy thing about that is like the quality is staying the same. Recently we had DeepSeek V3, which is a model that literally just, I don't want to say ruined everything, but definitely shook up the AI industry because this was a model that was borderline on par with Claude 3.5 sonnet according to certain people. Some people may argue against that, but the point is, is that this is at a fraction of the cost. One trend that I have currently seen is that the majority of people that are building applications with models, you know, sometimes the cost is a factor. And I've seen many people who are using AI agents actually switch to deep seek version 3 because it is just so cheap for what you are getting in terms of the evaluation. You can see the mml evaluation score versus the input API price per a million tokens. Deepseek v3 just comes out to be the best in terms of the performance to price ratio for the optimum range. And when we take a look at that, I mean you have to understand that because it's capitalism, it's a free market, people can compete and when people compete, prices often get cheaper and that's often better for the customer. But this is going to have a variety of different knock on effects for society that most people just simply aren't prepared for. I mean, when we actually think about what they are saying here, and this is OpenAI the leading company, we should really be paying attention. He says nonetheless, change is coming. It will be reflected in the first, you know, in the prices of goods and labor. And this is essentially referring back to what Sam Altman has said. And I think I can muster up a very simple example for you guys. Let's say you are a software engineer. And I do hate the fact that I'm using Software Engineer as an example because there has been some recent controversy around AI software engineers. But let's just say, for example you have an AI software engineer or let's say pretty much anything that is on a computer. What happens when you have an AI agent that is able to effectively do any task on a computer faster, quicker, more reliably than a human, all for a fraction of the cost. There are entire industries that are built upon jobs that are at desks. And if, let's say for example companies realize that, wait a minute, these people that I'm paying $40,000 a year and of course I have to give them sick pay, holiday pay, if I can just literally spend $2,000 a year, on a few AI agents, that is going to be a much more efficient strategy for them. And if that does occur, then that actually means that, that secondly, we have a company that is going to be operating on high margins. Now think about it like this, guys. This is where prices start to change. If that company goes from paying their employees, let's say 6k a month each, to, you know, digital employees that cost $200 a month each, that means that they have now so much profit. But here's the thing, okay? They aren't the only business in that space. Other businesses in that space will now likely have the same digital employees that are the same cost. Which means that if a business just lost 80% of its overhead and is now 80% more profitable, companies are going to start undercutting each other because of course they're going to be starting to fight on price. And if things start to fight down on price, well, for consumers, that means for us, the prices of things will eventually, of course, come down. It's going to just change everything. What happens when I can say, okay, I need you to go ahead and create this business by the time I wake up, I, I want this website, I want you to reach out to these many businesses and I want you to have a step by step strategy plan and start executing it by the morning. And let's say I can do all of that for a fraction of the cost of hiring a full media team and whatever it may take to actually get that orchestrated in the real world. I think when we actually start to look at that, we can start to say that, okay, this is going to be a drastic change that is coming. And I've heard this time and time and again. And one of the craziest examples that Sam Haltman has actually spoken about is the fact that the coming divergence between changes to cognitive work and changes to physical work probably are going to be quite dramatic. He says that in 2023, it cost around $30,000 to get a simple iPhone app created, but around $300 for a plumbing job. I wonder what those relative prices will look like in 2028. Currently it's 2025, so that's three years away. But everyone knows who's in the AI industry, just how quickly timelines can move. Now, this is crazy because $30,000 to get a simple iPhone app created is already changing. I know people that are building iPhone apps with, you know, low code tools, and if we actually look at software being developed, I've seen people whip up really simple sass from just a Simple text prompt. So, you know, previously when you'd have to hire a software developer or maybe even a small team, whereas now you can do things with a simple prompt. How does that change certain industries? And when we take a look at the fact that humanoid robots are not where they need to be in terms of the speed of development compared to what you would need to actually get robots to complete a plumbing job, it's just a complete divergence between the two, which means that potentially it could be more expensive or more lucrative to be a plumber in 2030 than it is an iPhone app developer. Whilst that might sound like a crazy statement, you have to understand that AI has continued to surprise us. Of course, anything is possible, but with the way things are going, I wouldn't be surprised if that is the case. And this is of course due to the fact that. And this actually refers to Moravec's paradox, where this is the paradox that basically states that, you know, tasks that are really easy for humans are actually quite difficult for machines. So, you know, tasks that machines find really easy, things like, you know, complex calculations, playing chess, that are really easy for AI systems, we actually find these things quite hard. So when we actually manage to duplicate human intelligence, it's going to be really interesting because, of course, whilst yes, the humanoid robots are going to be coming, it's quite likely that they won't come as quick as the software side, because software is really easy. But with the physical side, you are interacting with the physical world, which means, you know, there can be mistakes. There is, of course, feedback that you need to get, it's just a longer feedback loop to iterate. Whereas with software it's just, you know, you're changing code, you're fixing things and eventually you manage to figure out how to optimize it and you eventually get to your solution. But with physical robots, there's just different things. There's supply chains. You actually have to go out and build it to interact with the real world and ensure it is safe. So this is going to be something that is eventually there, but I can't imagine that this is going to be kind of, you know, replacing physical labor jobs at the same rate that jobs on a computer will be replaced by. So there is going to be a really interesting discrepancy between those kind of jobs in terms of the demand for them. But you know, what kind of goes on. It is actually rather interesting because, you know, Elon Musk has actually said what comes next. The advent of unemployment, of joblessness, will not be considered a time of crisis, depression or condemnation, but a glorious blessing. Probably none of us will have a job. This is Elon Musk's statement and he's said this for quite some time now, so I'm not surprised by him, by this statement. But it is something that we can't really fathom at this current state, considering the majority of people are employed. But it's quite likely that in the future when AI is able to do a lot of things better than us, it's quite likely that we will have to focus on human centric work. And this is why the Post AGI Economy series which I created on my channel was basically talking about the fact that we're headed towards a meaning economy where human centric industries and experiences are going to be taking center stage. And in this new paradigm you're going to need to rethink what we value as a society. So maybe things like sports stars, musicians, comedians, communities, those kind of things are going to become more and more important as AI just handles the mind numbing work that we used to do. Now another thing that this individual at OpenAI actually states is that, you know, it will also change how we live in how healthy we are. Our ability to use technology to change our own bodies and minds, every single facet of the human experience is going to be impacted. We can see here that, you know, this is really, really true. I was actually reading a research paper the other day, actually made a video on this. But if we actually look at the performance of differential diagnosis generators and LLMs on clinical pathologic case conferences, we can see the O1 preview actually manages to surpass all these other systems. And I wouldn't be surprised if O3 manages to completely dominate this benchmark and even potentially saturate it. And there was this other benchmark where O1 preview manages to completely outdo physicians and resources when looking at certain diagnoses. So it's pretty crazy how much this is going to affect the healthcare industry. But the point is, is that this is going to impact a vast array of different things. And even the Nvidia CEO actually talks about, he actually talks about, you know, AI's potential to transform biology from a life science into life engineering to improve healthcare. You know, he can't even imagine what that's going to look like in the next 10 years.

Applying this technology and advancing it further for some of the most challenging and pressing issues of our time, whether it's digital biology or healthcare is some of our best futures. I can't imagine what we're going to be like in 10 years, frankly.

Yeah.

And when we apply artificial intelligence to the field of biology and to. For us to not to move beyond calling it life sciences to life engineering, likelihood of dramatic medical advances, even in neurological, one thing, Alzheimer's, I'd say those would be solved.

One of the things that people actually start to discuss is Lev. Now, this is longevity escape velocity. And this is a hypothetical situation where life expectancy increases faster than people age. And this is basically. I know this is going to sound crazy, by the way, is the fact that, you know, people could live indefinitely in the near future by defying aging through money, ambition, and possibly AI. So some people believe that this is possible by 2029. And Ray Kurzweil is a big supporter of this.

Expect by 2032 we'll develop what I call longevity escape velocity. So this means right now, as you go forward a year, using up a year of your longevity, so you're a year younger. However, we have scientific research that's coming up with cures, new treatments for diseases. So right now you're getting back around four months. So you use up a year, but you're getting back four months from scientific research. But the scientific research is expanding exponentially. So by 2032, you'll get back not four months, you'll get back a full year. So you use up a year, but you get back a year and you basically stay in the same place past 2032. You actually get back more than a year. So you go backwards in terms of time.

Now we can actually see what Demis Hasabis said. And this is the guy that is the CEO of Google's AI team. He actually talks about AI being able to cure all diseases, which is pretty remarkable. But like I said before, AI has done some pretty crazy things, especially with AlphaFold.

The reason I'm doing all of this is because I think AI is going to be incredibly positive for the world. I think we're. I think we're within shooting distance of, you know, curing all diseases with AI, helping with climate through material science and new energy sources and other things that I think AI can invent as well as, you know, in our daily lives, just improving productivity and enriching our daily lives and making, you know, mundane admin things sort of be dealt with automatically. I think those are all amazing. And that's all coming very soon.

And if you don't think this is going to happen, you know, we've already had the slight signs that, you know, with ASI or advanced AGI, we're going to be able to, essentially, we've already had the slight signs that, you know, with ASI or advanced AGI, we're going to be able to essentially compress time. That's the main concept that they're trying to drive home. It's not like an AI is going to be able to immediately solve all diseases. It's the fact that we're going to be able to do accelerated research. So the research that maybe we would have got after 70 or 50 years, we're basically just going to be able to duplicate humans digitally and do that research virtually. And essentially from there, that's where we compress down all of that research by running simulations, either having a billion agents scaled out to be able to run those tests for us, and then of course, bring that into the real world. But that is the key point here, is that, like, we're basically compressing time. And that's what they actually talk about with AlphaFold, which is, I think once people understand that once you're able to compress time, that is where the timeline speed comes from.

There's, you know, 20,000 proteins in humans and then there's 200 million known to science in all of nature. So it would just take forever to do it like that. So we want a computational solution to it. And that's what AlphaFold2 was, and sort of essentially solved that problem. And then we spent a year folding all 200 million proteins. So that would have taken, you know, equivalent of like a billion years of, you know, PhD time, which is incredible, done in one year. So that's, I think, the power of computational method. And then alpha fold three, which we just released, as you said, you know, last couple of weeks, takes that one step further now. And biology is, is not a static system, right? It's a dynamic emergent system. And so you, to really understand biology, you want to understand the dynamics. So in Effect, Alpha fold 2 is the static picture of a protein. And Then Alpha Fold 3 now allows you to understand how proteins interact with other biomolecules. So of course other proteins, but also strands of DNA, rna, and also things like ligands, you know, compact chemical compounds, effectively useful for like, drug discovery. So it's sort of a. Another huge leap. And where we're going with this is eventually I would like to model an entire virtual cell, the workings of an entire virtual cell really accurately, so that one could do experiments in silico on the cell, the virtual cell, and that would be informative of what would actually happen in the, in the real lab.

Another thing that this individual also says is that, you know, it feels outside of the Overton Winden right now to suggest that small much change could happen very quickly or to even realistically grapple with what those changes might entail. It is too easy to say that the present is more urgent and more real. And it's really true. You know, right now it doesn't seem real that you can even talk about these issues. You have to understand that a lot of the times when we realize it's just one breakthrough that ushers a huge variety of different changes, like for example, the paper. Attention is all you need is what not started all of this, but definitely ignited the global consciousness in terms of AI. And imagine Google just didn't publish that paper, or imagine, you know, ChatGPT didn't do that demo. Where would we be? I mean, it's just truly fascinating stuff. So overall, I think this is going to be something that is truly fascinating. The people at OpenAI have key insights into the future and they're essentially saying that, you know, right now nobody is really paying attention to exactly what's occurring. The world just isn't grappling enough with the seriousness of AI and how it's going to completely change the world. The economy, different trees, jobs, and potentially the current economic system as we know it. Now I could go on and on about this, but I'll leave a link to a few videos that actually talk about this in a decent bit more detail. But Definitely, guys, by 2030things are going to change a decent amount. With that being said, I'd love to know your thoughts on what you think about this.