Sam Altman FIRES BACK At DEEPSEEK " We Have Much Better Models Coming

So Sam Altman finally responded to the Deep Seek drama. As you may know, Deep Seek versus OpenAI became the number one discussed topic around artificial intelligence, and it is probably still one of the major topics. So most people would be wondering, what on Earth is the CEO of OpenAI saying? Because he's actually usually quite vocal when there are any major AI announcements. And considering the fact that this announcement was probably the biggest one ever with regards to normal individuals who don't pay attention to AI now shifting their focus to Deep Seek, it actually seems like Sam Altman was unusually quiet, but nonetheless, he did actually state some things, and I think most people are going to miss exactly what Sam Altman is saying and a few things that OpenAI have said in a few blog posts. So I'm going to dive into those right now and trust me, you're not going to want to miss this. So he actually states that Deepseek R1 is an impressive model, particularly around what they're able to deliver for the price. But the clear statement is this, okay? And I'm actually not doubting this. Of course you could argue that, yes, of course he does have every incentive to state this statement, but he says that we will obviously deliver much better models. Says here that it's quite exciting to have a new competitor and that they're going to pull up some releases. Now, I think that we will obviously deliver much better models. Saying it's something that most people are overlooking. I think once again, OpenAI are probably a lot further ahead than people think. I will talk about this later on in the video regarding certain statements, But I think OpenAI have shifted their entire focus and most people aren't really paying attention to this. And I think OpenAI scope of what they're doing may have confused the general public. But when analyzing certain cryptic blog posts from the past few weeks, I think this is something that genuinely, on the surface it looks bad, but when you peel back the layers of what's going on in the AI industry, it doesn't look bad. So Sam Altman here, okay, he says we will obviously deliver much better models, and it's quite invigorating to have a new competitor. Now, most people are going to say, he's the CEO. Of course he's going to say that. He's not going to say that they managed to outdo us. But here's the thing that I want to show you guys all, which is why I think the people at OpenAI aren't really panicking, and most people are going to be once again surprised by OpenAI in probably 6 to 12 months. So most people that are coming into the AI space right now, it's actually good, it's getting a lot of publicity. But most people will have missed this statement, okay? And essentially this statement is something that isn't a joke. This was written on OpenAI's blog post by Sam Altman and essentially they talk about asi. Now, ASI is artificial superintelligence, which is a system that is, I guess you could say, I wouldn't say marginally better than AGI, but it's so much better than AGI. And AGI is something that is heralded as big, basically electricity in terms of technology. Like that's how life changing it could be. And the thing is that most people have missed this statement. Now, in this statement, OpenAI have basically said, okay, that whilst they are going to be focusing on certain products, they're actually turning their end beyond that. And I think a lot of people didn't really understood what this meant by now, but I think when we actually take a look at what OpenAI have done in the past, they've actually shown us that, look, they don't really care that much about the GPT series. They're not really focusing on that because short term you're just competing for completely marginal gains. Anytime you've seen these GPT series released, the marginal gains that you do get aren't simply worth it. And I think OpenAI, and this is the point I want you guys to take away from this video, is that OpenAI are playing the game of artificial superintelligence. I know it sounds crazy at this moment in time, but it genuinely makes sense. Think about the last time that we have got a GPT4 update and they says that we're beginning to turn our aim beyond that to super intelligence in the true sense of the word. And I actually need to show you guys the full article because it is a little bit more in detail. And so I want you guys to pay attention to this, okay? So you can see right here, it says we continue to believe that iteratively putting great tools in the hands of people leads to great broadly distributed outcomes. And that's the tools like ChatGPT Canvas or one of those simple tools. But you can see right here clearly that they state that they're beginning to turn their aim beyond that. Okay, beyond AI agents, beyond that standard stuff. And. And super intelligence, guys. So a lot of people are thinking, oh my Gosh, how are OpenAI losing the moat, guys. GPT 4.0 was released ages ago in terms of AI timelines. In fact, let's actually take a look at that date now, guys. GPT4 was released on 13th May 2024. Do you not really believe that if they wanted to, they couldn't have released a model that is superior? The only problem is that it's not worth the actual reward. And we've seen this in articles where they talk about the fact that the gains they get, they do make better models, but it just isn't worth it to serve that model to every user because number one, the inference costs are high and of they actually have to trade off from focusing on providing good customer service now versus actually getting to ASI and then being able to dominate everything. And I think OpenAI is a ridiculously smart company and I wouldn't bet against someone like Sam Altman. He's actually incredibly, incredibly intelligent. So you have to understand that this guy at The Helm of OpenAI, I don't think he's going to lead them off course. And this is why I'm saying super intelligent tools could massively accelerate scientific discovery, innovation beyond what they're capable of doing on their own. And basically think about it like this. Whatever superintelligence is going be able to do, it's going to seem like magic to prior civilizations. If you were to roll back the clock 50 years and show someone the current technologies we do have, a lot of things would definitely surprise them. And we can do the same if we manage to go 50 years into the future on how crazy technology is going to be. And the point I'm trying to make here guys, is that OpenAI are basically saying that, look, obviously we are going to deliver much better models. And not just that, but guys, they're actually focusing on artificial superintelligence. And you do have to remember that there was an entire fiasco where people actually wrote a letter to the board stating that this technology is super dangerous. Someone actually left OpenAI, went off and started their own company. And whatever they showed investors, they managed to raise a billion dollars without any kind of product or service. So whatever secrets are in this stuff, and I've read the research papers, it definitely looks like they're going to do it. And I've looked at past AI systems before and it does look like that is the path that we're headed at. So the point I'm trying to make here is that maybe OpenAI aren't behind now. Sam Altman also said that we are more excited to continue to execute on our research roadmap and believe more compute is more important now than ever before to succeed at our mission. So for those of you guys stating that compute is not important, I do think that compute is still going to be more important than people think, because there are so many different ways to use AI and I think in the future it's going to be probably the most widely used thing ever. And with that, you're going to need so much compute to be able to serve the users. A lot of the reasons why now, a lot of the reasons right now why we don't have access to a lot of the things that we do want is simply because we don't have the compute. I know a lot of people are going to say, well, deep seq R1 is much more efficient. It's faster, guys, over time, these things are always going to get faster and more efficient. It's just that I think these companies aren't focusing on that right now. And of course, you can see right here, he says the world is going to want to use a lot of AI and they're going to be really quite amazed by the next generation of models coming. Of course, he's doing subtle hints referring to the next generation of models. This could be 04, this could be other models that they haven't spoke about. And Sam Altman has spoken about how in 2025 they're actually going to release a tool that nobody has even predicted. So it's quite likely we're going to get maybe another GPT3 moment in some kind of AI product. I really don't know what that is just yet, but I do believe that that is probably going to be something that exists somewhere down the line. So I wouldn't just count OpenAI out of this race just yet. And he says, look forward to bringing you all AGI and beyond. And that's, of course, the promise for this year. Now, of course, we're going to remember OpenAI have been under some criticism because Sam Altman, you know, himself, they're saying that he spent more on his car, which is a Koenigsegg Regera, which is around $5 million or even a bit more than that, but basically saying that, look, deepseek actually spent the same amount and managed to dethrone your entire company, which is of course a pretty hilarious tweet. But of course it does go to show just how popular the deepseek term is, because right now they are trending larger than ChatGPT. Now, other individuals have, I don't want to say fired back at OpenAI, but they've said some interesting statements regarding OpenAI's positions and they said OpenAI is not good. They necessarily won't always be at the forefront. And this is of course quite true considering how quickly markets can shift in the AI related space. And this was related to someone who works at a hedge fund basically trying to predict where the market's going to go. And they're stating that, look, the markets are very dynamic and anything can change at any moment. And another thing that I think people are missing is that recently OpenAI actually did, or they're going to do a meeting in like a closed door meeting with the US government to actually show them the PhD level super agents that they're working on. And these are going to be essentially super agents that are designed to tackle, see, multilayered real world problems that human minds struggle to organize and conquer. So I do think that this is going to be something that once again people are underestimating. And this talks about how they don't just respond to a single command, they actually pursue a goal. And super agents synthesize massive amounts of information, analyze options and deliver products. And this is going to be a game changing thing because as we round out 20, 26, these things are going to be doing a lot of work, particularly in certain parts of the economy. So the point I'm trying to make here guys, is that this is something that I would say that like, yes, Deepseek is a wake up call to the industry showing us that, look, you can train these things efficiently and those other companies that are just based on open source kind of technologies, you definitely need to focus on other areas. But Sam Altman here clearly stating that, look, we are clearly going to deliver some better models, which I do agree with and I personally believe that, look, they're basically saying, look, right now you can create those models, you can have them. And I think it's important to understand that because the scope of everyone's opinion has been that ChatGPT has always been the leader in AI. And I think that because people think that since they're not the leader in GPT 4.0 in terms of the current models and even the O1, that they're still not the leader in AI. But I think OpenAI are playing a much longer term game in the sense that they are allocating more of their compute towards superintelligence and allocating more of their compute towards super agents, because that is truly where the value lies in terms of an actual product that will generate actual revenue for the company. And I think that's why they don't really care about this. Because if you've been someone that you've been paying attention, you'll know that Google is actually a better model than GPT4O, and Claude Anthropic is actually a better model than GPT4O. And there has been some way now for around maybe even three to four months. And OpenAI, they don't seem even in the slightest bit concerned about this. So for me, that goes to show me that OpenAI are thinking on the next level. And I would probably bet that in the next few months, we're going to see exactly what those bets are going to be, or we're going to hear some pretty information coming out of the company. With that being said, if you guys did enjoy the video, I'd love to hear your thoughts about the deep Sea drama, and I'll see you guys in the next one.