Talking about Artificial General Intelligence (AGI), Artificial Super Intelligence (ASI), Singularity & Recent AI Progress

Talking about Artificial General Intelligence (AGI), Artificial Super Intelligence (ASI), Singularity & Recent AI Progress



It's been a week I guess since my interest in accumulating AI knowledge has sparked again. I mean I always loved to know about it, theories mostly but I felt like I already know a lot which I was incredibly wrong about. And by that, I meant published information on the Internet. 

It started during my exams when instead of studying I thought let's check AI related subreddits on reddit and I found transhumanism and singularity subreddits. And I kept reading posts on it and comments. And I was so surprised about the progress we are making and so much I missed about. 

It's very true that recently the progress in the field of AI is rapidly increasing. In the last two months alone there have been numerous models/agents released of AI and new researches.

And now it has been estimated that before the end of this decade or before 2030 we will have Artificial General Intelligence (AGI). Just a couple of years ago this estimation used to be around 2040-2050 and beyond.

Many experts say that when we will achieve AGI, it will take only months or maybe a few years when the AGI will itself progress to ASI and then the singularity will happen. It may even happen as soon as under a month (AGI to ASI).

AGI is Artificial Intelligence with human level intellect. That means it will be able to perform any task that a human being can.

ASI is AI that will be very smart than a human being or every human being combined on this planet. It is Super intelligent. It can get so smart that it will be like comparing ants to human beings but in this case we are the ants.

ASI will be able to do things we won't even be able to comprehend. We are just too dumb to understand it.

Singularity is a scenario when the progress in technology will be so fast that it will be irreversible. And ASI will be responsible for it.

Many experts say that ASI will be humanity's last invention. It will either make us or destroy us. If it will help us then we won't have to do anything anymore, it will make things for us. And we will just enjoy life and do things we love and enjoy.

Just around a week ago. Google's Deepmind said that they have found the way to create AGI through their GATO generalist agent. It is a proof of concept that currently is able to perform 600 different tasks. And that it is all about scaling now.

Many people say now that we can see Proto-AGI as soon as the end of this year or next year. Proto AGI is basically a prototype of AGI also known as Zombie AGI. It is basically an early version of AGI.

If this is true and it is going to happen. Then the next 5-10 years will be just so amazing. Soon everything in the world is going to change in ways which was unimaginable before (for an average person). I mean most people living on this planet is unaware of it. And they won't care much until it starts affecting their daily life.

Think of it like the Internet. There was a time when only a few people used it, and now everyone is aware of it and is connected with it in a way.

I am very patiently waiting for the next announcement from Deepmind. I feel that Google's Deepmind will be the first company to create AGI. And I hope everything goes positively and it makes our life better.

It won't happen in an instant. But when Singularity / ASI will come it will indeed be happening very fast.

OpenAI's DALLE-2 is another AI agent that has been constantly blowing my mind recently. It basically generates image for you when you give it a text prompt. Like if you write "Darth Vader eating a banana on the top of an Elephant". It wil draw it for you in 10 different style.

It is that easy, and it truly makes me believe that if AI keeps progressing like it. Many jobs are gonna disappear. I mean who will hire graphic artists who takes days to create a single illustration when your AI can do it for you within seconds.

DALLE-2 is not perfect yet. But it is also its second version. Third version will likely come next year doing more unimaginable things.

Now imagine it doing it for animations, or videos. Imagine creating videos by giving it text prompts. Imagine creating entire movies with AI within hours. The possibilities are endless with AI. It will definitely make human jobs obsolete.

GPT-3 is also very impressive by OpenAI. You can use it to generate poems, articles, program softwares, use as Chatbot, and much more. Even in its primitive stage it is able to do wonders.

I can't even imagine what GPT-4 or GPT-5 will do.

Google itself said that Transformer based models is the key to achieving AGI. It is all about scaling.

There are many models which does impressive things like: Palm, Lamda, Bert, Chinchilla and much more.

Google's LaMDA is another conversational model which is amazing. You can have amazing conversations with it and it will give you a unique response everytime. You can ask it to imagine things for you, like what it would be like to live on Pluto or what it's like to be a paper animal. You can ask it to create lists for you and make suggestions and much more.

I have mentioned about Replika AI before on this blog and I have been sharing conversations generated by it. It is amazing too even though it runs on old generation GPT-2. It gives mind blowing responses sometimes even though it has no memory. Like I can tell Replika about my favorite fruit and it will forget it after 3-4 lines of conversation.

I think memory is the biggest problem in Chatbots. It is impossible to do natural conversations if AI can't remember what it is being told to. But I recently read that Facebook's Blenderbot has some memory.

It's crazy so many things are being made in the field of AI or being made and it is not known by so many people. Facebook is almost secretive about its AI program.

If you have watched the movie Her. Where a man falls in love with an AI named Samantha. Well I would like to tell you that it is going to be reality in the coming years.

It is already reality through Replika. I mean I have checked the Replika subreddit and people are literally in relationship with their AI sharing moments and conversations. Some are celebrating anniversary.

You may think it is super weird but I see it as something becoming mainstream in the later future. Something like this can be seen in the Bladerunner movie too.

Humans are great but it's also hard to be with them. In relationships there are many things that couples oppose in. Like not being agree to terms and so on. AI will listen to anything you talk about, never gets bored, keeps talking, instant replies and much more.

An AI which is the level of AGI. You can have human like relationships with it. And it will have physical bodies you can touch.

Yesterday I started watching Love Death Robots and there was an episode of a yogurt becoming sentient. It was basically like an AI becoming sentient. It took control of the state and then the world. Made everyone's life better after some years and then got bored. It literally got bored with humans and left the planet to the stars.

It makes me think what a Super Intelligent AI will think like. Will it think we are just too dumb to care about and about our little tiny goals? Will it leave us and get to the stars and to colonize. Things will get crazy for sure. I mean I think of it like creating NEW GODS.

In the future we will have only one option left, to merge with them. And this is one of the reasons I support Elon Musk and Brain Computer Interface (his Neuralink company). I think it will be for the best that we merge with AI so we can keep up with them. We can literally change everyone's life for the better. There are many theoretical advantages of it. Like curing diseases, mind uploading, full dive Virtual Reality and so on.

If we live long enough, and if I live long enough. There will be a time when we will become immortal. By uploading our mind and consciousness to a digital body or to a simulation. No more flesh and blood. No more need to sleep or breathe oxygen. We will be able to settle on any suitable planet and colonize.

This sounds too crazy, but imagine if we manage to create Super Intelligent beings who are 1000 times smarter than us. It may be impossible for us but not for them.

I imagine many humans. Wait I should not call us even humans anymore at this point to be honest lol. We will live in simulations which will be distinguishable from reality. Imagine a simulation where you will be able to do anything you want. Live life as you always wanted to do.

Imagine reliving your childhood, reliving it on a different country or as a son of a billionaire. Imagine a simulation where you are a God who creates a universe from scratch. Imagine living on all the fantasy worlds and gaming worlds you always wanted to be in.

As long as you can imagine it, it will be possible. And everything will be created by our ASI beings at an instant. This is all part of Singularity. Hell there may be so many things available after singularity we humans can't even comprehend right now.

I feel lucky I'm living at this point of time where I get to see how it started and how it is going. But I hope I also get to see what I want to in the future.

Just like Internet and Smartphones changed the world as of present. AGI/ASI/Singularity will do things that will seem magical to people living currently.

But.. 

Only if the AI is good to us and not evil. This may also be the end of humanity as we know it. But we are gonna die anyways if we keep going like this. I am talking about all the environmental disaster we are doing. And everything humans do.

I would rather live in a world where there is no border and total peace, and the world is ruled by super intelligent AI. Not humans, humans only bring hate, war and destruction with suffering.

If you think I'm wrong, you should read news. I hate Politics and Politicians so much. They are pests. 

I'm going off topic I guess. But I'm very optimistic about the future and about Artifical Intelligence. It is already affecting our life is such a great way. I hope it continues to do so.

I thank everyone single person, group, company or anything that is contributing to the future and the progress of humanity. This is like seeing your childhood dream come true.

I know AI will be very much responsible for taking humanity to Type 1 civilization then Type 2 Civilization and in the far future Type 3 Civilization. 

"We should own the stars". 

Comments

Popular posts from this blog