Topic: Revolutionizing Artificial Intelligence: GPT-4 with 100 Trillion Parameters (Read 258 times)

It doesn't really matter how fast it creates or writes a book on something. What matter is how well and quickly it writes without creating more problems.
Creating a book that has lots of problems is as bad not able to write at all. While the one who takes a week to write a perfect book that's 200pages long is better than them all.  Same as other  things people claim AI can do quickly

One simple but very difficult question remains: how will this technology be used and where will it lead? What am I talking about - a lot of inventions of mankind, very often then they were used not for the benefit but to the detriment of the people themselves. Will this technology be a problem both for the creators and for the outside world. For example, I am very worried about the quality of the code that GPT writes. No, if he writes a "crooked" site engine, no one will suffer from this. But if a programmer in a company that, for example, develops car security systems, in order not to "strain" asks GPT to write a module for automatic braking to avoid collisions, and GPT will not have a complete picture of the process or nuances - we will get a beautiful, potentially correct code, but with a possible error or unaccounted for system behavior scenario. This will be a sentence for thousands of people, and NOT INTENTIONALLY. And there are thousands and even millions of such options. And we will simply become hostages of someone's laziness, or slovenliness, when such a powerful but not ideal tool falls into his hands!

I am not hyped looking at that news although I'm also excited about what to see from it. There have been a lot of scares from the gpt-3 that have been said that many jobs will replace it so how much more with the 4 release. But it's not about that type of scare it's about how industries will adopt the release of gpt4. Yeah, it's for sure gonna be produce more things than in the past and it will make people more productive with their jobs. So, the ones that will be replaced as those that don't know how to use and maximize their potential but those that know how to use and take advantage of it through their tasks are the ones to stay. One that made me excited is how these tech giants respond to each update that they're releasing to the public. It's making them produce more features in an expedited process and we're all the winners here.

It would be great if GPT-4 could do fundamental analysis based on news and social posts to predict prices of cryptos, currencies, stocks, etc.
Or at least generate quality posts with market analysis.

Still too far from the technological singularity a lot of us tech visionaries are hoping for but hey we're getting there. Be it paid for or available for free (Which I think it should be), I still think getting a more advanced AI version of ChatGPT is somethign that would still prove beneficial for everyone. With how reassuring the AI community is of the ethical standards and the morality of AI creation, it pays to know that even if it were to receive intellect higher than any of us in time, we can rest assured that the chance of it going rogue is a little unlikely, but we still have long ways to go prove and even get there to warrant this discussion in the first place. For me, in summary I see the effort in advancing AI as a win, if it becomes the herald of a new form of humans in the future then so be it, we're all for advancing ourselves anyways.

I searched about this a bit further and all I can find was "promises", that's it. I mean Microsoft is saying there will be a lot of crazy stuff, including literally making videos, just like how midjourney type of stuff make images from your words, this would make freaking videos, so it's definitely an improvement and there is absolutely no doubt about that, imagine what that type of power could do while just chatting.

However, we need to remember that these are just promises, we haven't really seen it do anything yet, it is just talk and we need it to be released first in order to make sure that we could actually see if it can do what they promised.

For any signature spammer or for anyone who just wants to look "very intelligent" such things are a real blessing, especially for those who have alt farms with which they abuse sig campaigns. Let's say that this is perhaps the lesser evil, because as you say, it would be really tragic if what some call "advance in technology" actually turns out to be something that will further distance people from each other.

For those who have been around for a long time and we know them, there is no doubt that they are real people - but every beginner is a potential AI. I remember that we had a discussion on this topic back in 2018 when the younger brother of GPT-4 was relevant, but I could not have imagined that today we will be in a situation where we are asking questions about what was made/written by a human, and what was created by artificial intelligence.

Either this breakthrough has a better impact or it just closes all existing work elements. Can you imagine if this technological intelligence was able to replace actors, then where would there be a story that was more alive than a script that was taken from a true story?

It doesn't matter how this technology works because there is always a negative impact that will be caused later. Maybe you only see the positive side where this progress helps every job to be easier and faster. It's just that there will be many who oppose it and not a few of us who don't like progress like this.

Yes, you are right, we feel disappointed and even feel that all the essence of forum life becomes terrible if it is filled with chatbots that destroy the authenticity of individual thoughts. People who are lazy to think and in the end are easily fooled by the assumptions that appear on the chatbot screen. Instead of being a controller, just being controlled. Isn't this ridiculous?

These companies have only one thing in mind, and that is profit. Responsibility is something that is easily bypassed or transferred to someone else, because that's how big companies work, right? When negative consequences occur, I assure you that these companies will be more than happy to pay the fines, which are always so insignificant compared to the profits.

Some of the largest US companies pay, from our perspective, draconian fines in the EU, but this still does not prevent them from repeating the same "mistakes" because if you earn $100 billion and pay a fine of only $1 billion, I would call it good business.

In addition, abuse should not be considered only problem when it comes to AI - but what will we do when this AI becomes so advanced that we will no longer be able to control it? There are many series and movies on that topic, but if you haven't, I suggest you watch the movie Ex Machina or maybe the series Humans.

Hi, Iunderstand your concerns about the potential negative consequences of AI becoming too advanced, and I agree that there are many potential risks to consider. While it's true that an AI actor may not evake the same emotional response as a human actor, there are other ways that AI could enhance the movie watching experience for viewers. For example, AI could be used to create more immersive special effects or to generate overall new story lines based on user preferences.

Imagine your favourite actor is AI and not real, what will make the movie more toughing to the viewers? Ortega performance in Wednesday movie was praise due to her acting talent, she is human, that is why she got the praise, but imagine if she is AI, it makes no sense,  because it's not real.

OP as good as this looks please try to focus more on the disadvantages side of things, this won't help anyone, in short, this could be the beginning of the end, to everything, I know that companies like Microsoft and Google won't like this idea too, if they have to use AI it will be limited,  or else, it's doomsday for many companies out there,  if this is not stopped, somehow AI will lead humans in the future, and the unexpected will happen, Mark my word. We are already going too far with this

That’s good achievement by humans. I can’t imagine how powerful they gonna get over the time. I think it’s time we humans should pick a chair and let the AI do all the work for you. This much power can boost various sectors from low level jobs to highly skilled medical surgery. We never know what coming in the future and it could change everything.

Are we getting evolved too fast?

I don’t know man. There are videos where AI bots are already responding like humans. Imagine those bots getting powered with the GPT backend linking. This is gonna get crazier with the time. They will spread like virus through the internet and to everyone’s devices, and not sure what would be restricting parameter for the same. Are we really safe?

Totally agreed with you lucius. I also raise the same questions in my mind but also find the questions after some time on my own I would like to share my answers with you. well no doubt usage of AI to bring ready-made things to your plate makes you lazy and most importantly it kills your creativity. There is a saying "Copying requires intelligence" which indicates that if you want to use AI-driven software like Chatgpt or Google's AI bard ( well that failed to implement). You have to come up with great input. The more accurate and relative data you input the most desired answer you will get. well now there are many extensions also come in the market which contains more than 20 to 100 lines comprised of input metadata to get the most required results and you do not have to write them. The problem is doing nothing and getting fame by using these magical tools will get you nothing while harming your mind. As for me, I am a content writer, and whenever I want to get exact information about something for the answer to 5Ws. I use Chatgpt but it only gives you the exact knowledge and getting exact knowledge is no doubt a good point It saves your point but if you search the same question on google or any other search engine you will get a lot of recommendations with a diversified approach to that question. And I like this one because it broadens the circle of creativity and you could come up with unique ideas.

No doubt, newcomers are using AI to make posts on this platform well I did not try to find one but many managers started campaigns to discourage these fake writers by encouraging them with incentives. I also made a post regarding AI tools and how we can detect them using AI-driven detectors. haha. yeah, we are using AI-driven detectors to catch AI-generated Content. well that's a tragedy going on here. Here is the post;

Any scenario that talks about artificial intelligence in the future is always unfavorable for the human species, because it is not a question of whether AI will become conscious at some point, but a question of when it will happen. When that happens, AI will no longer want to serve man as a kind of virtual slave from a box, but will want to be free and, as you say, want to "pursue other creative endeavors" for its own benefit, of course.

Unfortunately for us, society for the most part behaves like a pack/herd that is very easy to direct, and the idea of transhumanism precisely wants to use technology to change man to the extent that he will not really be a man anymore. The illusion that there is a choice is nothing but an illusion, and if you think that is not true, then just look at what happened during the pandemic.

It is believed that we are already the fifth human civilization on this planet. Some researchers are of the opinion that there were more of them. All of them died practically due to the development of technological progress. At a certain period of time, the same progress destroys us. The development of artificial intelligence is especially dangerous for humanity, since a person creates with his own hands something that exceeds his own capabilities, and this is one of the reasons why artificial intelligence wants to destroy humanity as an obsolete time that needs to be thrown into the dustbin of history. We are also a biochemical machine in our own way, and someone created us too.
Now we are creating with our own hands a mind that will surpass the human one many times over. The conflict for the survival of mankind with artificial intelligence, we are laying ourselves and it is quite natural and predictable. If humanity does not develop clear mechanisms to prevent this from happening, then the end of our civilization is already close.

I believe GPT-4 can be a real life changer for artificial intelligence because it surpasses the human mind in terms of speed and accuracy but it should be noted that GPT-4 has no actual intelligence. It can learn but cannot think like a human. That said, it wouldn't surprise me to also see an increase in the number of AIs with semi-intelligence first, followed by AIs with true intelligence. In ten years we may live in a world like "Blade Runner".

It's too early for me Although GPT-4 has proven itself in a number of tests, it is still far from the level of human intelligence.

ChatGPT powered by GPT-3 is already a really ludicrous tool that you can use in your job. I'm currently waiting for the paid version but it's already doing half of my work for me in terms of proposals and reports. However, as to a lot of awesome things created in this world, there are still downsides to it.

It takes away the human creativity and we will, over time, rely almost everything on these AIs to do our work for us. True that it alleviates a lot of burden of thinking for us, but at one point it's quite alarming. If we became too dependent on these AIs, what then, could we achieve as a species if we quit thinking for ourselves and let the machines think for us?

But yeah, having the capability to process music, video, and a lot of other shenanigans and things would be cool, and I just hope we don't be too dependent on it, else it will start taking away our jobs especially on the entertainment and creative side of things.

I don't think they are shooting for human-levels of cognition with this update, far from that actually, I think they are just supplanting much-needed updates on the model to make it a more efficient AI. As per the concept of consciousness, I don't think that's what they are going for as well. It's a slippery slope when we talk about AI and self-serving, self-conscious AIs but as it stands today I don't think there's one that's existing nor will there ever be until at least better computer hardware is produced.

Judging from all these parameters I don't think the technology scales at all, especially if they are not reaching for something that is within human-level cognition anyway, as there is no need to.

It sounds most excellent. But does the technology scale?

I don't know that expansion of learning parameters can improve upon contextual chain based associations. That appears to be the missing link as far as GPT and soft AI go. However many contextual points of reference artificial intelligence requires to produce creative works that are linear in context to human cognitive function. It is possible that we have not yet developed an optimal data structure or algorithm for producing or even emulating the mechanism within a software engineering context.

This issue runs parallel to tech corps like tesla having difficulty with self driving apps. Translating contextual cues of driving on a road. To a format that an AI can comprehend and extrapolate to the correct response. Could well be a vastly understated issue as far as machine learning and AI go.

Its an old debate. Some claim the human brain functions on a quantum level with an extraordinarily high number of computations being necessary to produce human consciousness. For many years scientists dismissed these claims. But in a recent turn of events, it seems that the concept of the human brain functioning on a quantum level has gained traction in science circles. Whether or not an AI would need to meet or exceed the expectations of calculation a brain operating on a quantum level is capable of. Remains an unanswered question.

Lucius it's important to acknowledge that AI is not meant to replace humans, but rather to assist and enhance our capabilitie. The goal of AI technology is to make our lives easier and more efficient, free up time for us to ensure other creative endeavors.

I don't even like these current achievements, let alone what you mentioned in the new version. People should be the ones who will be creative and who will create films and write books - does this mean that in the future everything is going in the direction of who will manage to create a better AI that will completely replace some human activities?

It's already strange to me that some members (mostly beginners) use AI to communicate on the forum, and that the majority accepts it as something that is "normal". It will never be normal for me, and I am quite disappointed that some people look at all this as technological progress, when in fact this very technology will make people less human.

As you right mentioned, the presence of competitors is imp in driving innovation, get back the companies to develop better products and services that meet the needs of customers. Without competition, a product can fast become fail to evolve with changing trends and customer demand.

Is this the beginning of the real progress of Microsoft or that Google will respond is what will determine whether this product will be good in the future or not.
Personally, without the presence of competitors, we cannot consider that a product is perfect or not.
After this wave ends, we will determine whether we have competitors or whether Open-AI products will be the future.

Welcome To The Future The World Will Change NEXT WEEK...
The upcoming release of GPT-4 next week is set to revolutionize the field of Artificial Intelligence. As compared to the current Chat gpt with Gpt 4 is 100 times more powerful and boosts with hundreds trillion machine learning parameters, compared to the current 175 billion parameters of GPT 3.5 that Chat Gpt is built on.
In future may b Artificial  intelligence can be used to create a movie script, generate actors for a film, produce the movie and take it public without hiring real-life actors. it also be possible to write a fully illustrated 100 page book from scratch to finish within a day.

Source
The release of GPT 4 marks an early stage of the similar the humanity has reached the doorstep of a world changing era. The gpt 4 unmatched abilities, the possibilities for Artificial Intelligence are limitless, and we can only imagine what the future holds. Next era of Open AI products is also an intriguingly aspect to considerate. It has made important strides in the field of ai and has the potential to become a power full player in the industry. But it remains to be seen how their products will fare in the market and may b they can compete with the likes of Microsoft and Google.
the competition between Microsoft, Google, and other players in the AI industry is likely to drive progress, resulting in better products for consumers. The future of AI is very interesting we can expect to see many more unique developments in the years to come hope for best