Open AI Team Surprises Everyone With Statement On Artificial Superintelligence

So Sam Altman, the CEO of OpenAI, Greg Brockman, the chairman of OpenAI, and Ilya Sutskova, the chief scientist of OpenAI, all published a blog post today on a very important threat regarding artificial intelligence that many people haven't really covered. And this is something that goes beyond our current understanding and current levels of artificial intelligence, which is essentially artificial super intelligence. And in this video, we'll discuss exactly why this is a very important post that they made and the very important key points that they talked about. But before we do dive into the exact video, I need to explain to you one of the key terms that is mentioned in this blog post. And the key term is super intelligence. So super intelligence is very different from artificial general intelligence, which you have likely heard over the coming weeks, because essentially, AGI is what we are somewhat close to, and it's a form of artificial general intelligence that can perform at human level against every single task. Now, what this blog post talks about is it talks about superintelligence, and that refers to a hypothetical form of artificial intelligence that surpasses human intelligence in virtually every single aspect. So it represents an AI system that exhibits intellectual capabilities far superior to those of human beings across a wide range of tasks and domains. And the concept of superintelligence is often associated with the potential for future development of AI systems that possess a level of general intelligence comparable to or beyond human intelligence. So the development of this kind of intelligence would have profound implications for society, where it may possess the ability to rapidly improve itself or replicate itself and leading to greater intelligence capabilities. Now, of course, this is something that is hypothetical, but it is a real possibility in the future as. As we start to train these large language models and they get increasingly smarter. And of course, there are emergent capabilities right here you can see that it says governance of super intelligence. And now is a good time to start thinking about the governance of super intelligence, the future AI systems that are dramatically more capable ever than AGI. Now, I think it's interesting that they are actually thinking about this, because AGI isn't even here yet, but it definitely means that we are really close. AGI, if they are thinking about governance of ASI now, essentially, it goes on to say, given the picture as we see it now, it's conceivable that within the next 10 years, AI systems will exceed expert level in most domains and carry out as much productivity as one of today's largest corporations. So that essentially means that these AI systems are going to be able to do what one company could potentially do. Now, and that is insane because of course, as you know, we, with exponential growth, we do have unlimited upside. Now, it does continue to say that in terms of both potential upsides and downsides, superintelligence will be more powerful than other technologies humanity has had to contend with in the past. And we can have a dramatically more prosperous future. But we have to manage the risk to get there. Given the possibility of existential risk, we can't just be reactive. So essentially what this means, this existential risk, they essentially mean that this could pose a threat to the entirety of humanity. Which means that you can't essentially wait for AI to do something wrong before you input the regulations. I mean, something that is essentially a common use example is flight travel. You know that there's this thing called the tsa. The TSA wasn't there before. We had essentially severe events that forced aviation industry to get much more stricter with its rules. But now, as you know, it's very, very strict when it comes to flying. The thing is with AI though, is that if you wait too long and, and you wait for something to go wrong before you decide to implement some kind of rule or policy, it possibly might be too late because the risk is existential, meaning that it could result in an extinction level event. So it goes ahead to talk about how nuclear energy is the most commonly used example of a technology with this property. And synthetic biology is another example. So we must mitigate the risks of today's AI technology too. Superintelligence will require special treatment and coordination. For those of you who think that this might be some kind of exaggeration, take a look at this clip of Sam Altman testifying in front of Congress about AI safety and regulations and informing them on what the future of AI is going to be.

Look, we have tried to be very clear about the magnitude of the risks here. I think jobs and employment and what we're all going to do with our time really matters. I agree that when we get to a very powerful system, the landscape will change. I think I'm just more optimistic that we are incredibly creative and we find new things to do with better tools and that will keep happening. My worst fears are that we cause significant, we, the field, the technology, the industry, cause significant harm to the world. I think that could happen in a lot of different ways. It's why we started the company. It's a big part of why I'm here today and why we've been here in the past and why and been able to spend some time with you. I think if this technology goes wrong, it can go quite wrong. And we want to be vocal about that. We want to work with the government to prevent that from happening. But we try to be very clear eyed about what the downside case is and the work that we have to do to mitigate that.

So it's clear from that statement from Sam Altman that the time to regulate AI is now. We simply can't waste any time debating whether or not certain countries should start to regulate AI. We need to do it as soon as possible before many different repercussions that we may not even think of could happen. And one small example of bad actors taking advantage of the rise in this very efficient technology is recently the AI image generation softwares have been getting remarkably good. Now essentially what someone did, we aren't sure where the source has come from, but they were able to entirely crash the stock market briefly with an AI generated image. So this AI generated image that you're seeing on screen actually managed to spread with quick succession to the point that the markets actually decided to plunge a significant amount of points. Now, once it was revealed that these images were actually entirely AI generated and there was no explosion near the Pentagon, the S and P and Bitcoin did rally back to their usual levels. But this just goes to show what could potentially be the scenarios where many different bad actors could do have access to these tools and they're able to distribute these images or maybe use AI software to essentially potentially attack a platform or just potentially absolutely anything that you can do with the increasing capabilities of AI. Now, something that was recently brought up for one of the capabilities that AI is potentially going to be used for when it comes to bad actors is of course biological agents. And this just essentially means biological warfare or creating harmful pathogens that can essentially cause viruses or engineered pandemics. This isn't something that is just some kind of hypothetical scenario. This is something that has been ran on before. You see weaponizing AI machine learning model creates 40,000 new chemical weapons in six hours. You see the sheer capabilities of machine learning allows it to create tons and millions of different combinations that essentially allows it to create workable pathogens that can affect humans in a negative way. And of course we know that humans aren't the very best when it comes to time efficient strategies such as making these kind of pathogens. This is where machine learning and AI comes in and they can literally perform many of these computations while we do other things. And this article goes on to further explain how this was done. It said collaboration, Pharmaceuticals retooled an AI drug discovery system to successfully identify 40,000 new potential chemical weapons in just six hours. Now, this is pretty crazy because other people are also saying bioweapons designed by AI, a very near term concern, Schmidt says. So artificial intelligence could bring about biological conflict, said former Google chief executive Eric Schmidt, who co chaired the national who co chaired the National Security Commission on Artificial Intelligence. Then of course, we had OpenAI's statement, which continues to say, second, we believe it would be unintuitively risky and difficult to stop the creation of superintelligence. So essentially what they're saying in the statement is that superintelligence is going to come. We can't just completely stop it because now that the genie is out of the bottle, the transformers are out, and people understand how to build these transformers, and of course people understand how to train these large language models. Machine learning is going to advance year over year to the point where eventually after AGI, the only thing next is artificial superintelligence. And they're saying that currently we just need to govern it because stopping it is pretty much impossible. I mean, like we saw from the letter, even if OpenAI does stop training GPT5, because of course it is the most powerful model, other companies and other smaller organizations aren't going to stop training it. So right now we do need some regulatory oversight. If we don't, this is where things do get out of hand.