Today I want to write my vision of the near future facing humanity in the next decade.

The progress of Artificial Intelligence continues. His achievements are increasingly covering the material sphere of life. Soon, we will all be able to see firsthand the new generation of synthetic robots performing everyday tasks. But how exactly will this be implemented?

First, the first AI platforms will appear. Conventionally, these platforms will have sections for each robot separately. In 2023-25, massive humanoid robots with legs, arms and hands will already appear. The most popular models will have millions of sales, and will be the largest ecosystem for them. In each section of an individual robot, Models will be presented for its behavior in various situations. You can imagine a robot as a phone, then its section is its operating system, and the robot's behavior models are applications that can be installed and removed on it. The robot can be taught yoga, cooking, delivery and fishing, these will be independent models of behavior, which the robot can either train independently, or download a ready-made model according to the rating level. It is long and expensive to train the robot yourself, but over time, training will require less time and data, which will allow you to generate these models on the fly, but only for the simplest tasks, be it floor cleaning. Complex tasks will take a longer time to master, and then the question of a more effective choice will arise - to learn quickly some task yourself, if a ready-made model of this task does not yet exist at all, or download a ready-made one. In addition, the structure of the operating system is very important, but it will be easy and understandable for everyone. The robot itself will wait for commands in standby mode. When it receives a command, it converts the voice into text, tries to isolate the meaning from the text, just like Google search is doing it now, comparing it with Action Models. If the model is found, the robot performs the action, having previously specified the initial criteria for the selected task. If the action model is not found in the robot's memory, the robot invites the owner to create his own or download it from the Internet. Behaviors such as painting walls, washing floors, cooking will be widespread, have many alternatives, and possibly even be used in parallel. Some complex tasks, including working at a computer for a robot, or, for example, repairing a computer, will need to train a lot of time and spend a lot of energy on them, which will be available at first only to the largest companies, then to medium-sized ones, and after a while to everyone, as of how teaching methods will get cheaper. At some point, even incredibly complex tasks can be mastered in the shortest possible time, but more complex ones will appear, the complexity of the tasks and the speed of learning these tasks will increase exponentially over time. Today the largest companies train robots for the simplest tasks. The number of companies will keep growing, training will keep getting cheaper, and robots will get smarter. The trend cannot be stopped.

​

I am sure, less then in 2 years AI companies will make AI that can control humanoid bot, which can do any human hand work. This means only one thing. When they will make 100 such bots, this bots will double every 1-3 months. By 2030, we can have 1 Million-1Billion such robots, which will make everything and change our life LIKE NOTHING BEFORE !

For example, they will produce all food for everyone, but NOT only food. Fabric production will become not just automated, but robotized with universal clever humanoid workers.

Also they can do some landscaping and environmental management. But wait.

Remember what I tell you in the beginning? If they can do everthing, they will making new factories to produce themselves MORE and MORE. This is not like electrocars. 1 electrocar cant produce other 2 electocars. If 1000 robots can produce factory that can produce 10000 robots per year, then 1 robot can produce 10 per year. You guys know what I mean ? Exponential grow of robot population in this decade. And this is not a funny joke, this soon will be our reality.

Oh cmon, but how much robots we REALLY need ? Somebody will say 100000 robots per city will be FAR enough to cover all the needs of the city and will absolutely right. BUT. What we really want from our life ? I mean, in near future we all will all have a certain resource. Resource, that humanity never have before. We can call it ABILITY TO CREATE. Create almost without borders. So, after we cover all our ordinary needs, what we will do next ? What is humanity really capable of with such power?

Mega projects. Yep, this things will be next after working humanoid robot revolution. After day, then count of robots will more then 1 billion, things will change really fast. Making new islands in pacific sea near Hawaii ? Of coures. How much people want to live on Maldives ? What if we can create Maldives the size of Eurasia in the Pacific? They would accommodate 2 times more people than they currently live on earth.

God only knows what incredible things we can create in this century. But the truth is, we can't even imagine it all. We got so caught up in films about the future that we forgot that the present future is actually UNKNOWN to us and most likely it WILL NOT look like what we CAN imagine. This fact fascinates me every time.

And I will not tire of asserting that the beginning of this will be laid already in this decade. Very soon, the first robot will come off the assembly line, understanding your commands and simply executing them. At this point, you can safely say that this is the beginning of singularity.

​

Actual articles was writen by me 2 years ago:

https://www.reddit.com/r/singularity/comments/k0csir/absolute_robotization/

https://www.reddit.com/r/Futurology/comments/jp6o30/how_humanoid_robots_will_make_impact_in_near/

I still have this optimistic vision of the future and feel that my predictions are coming true before my eyes.

I want to share these articles with new members /r singularity because so many good people have joined us over the last year

P.S.

I want to add that it is in this article (in my opinion) that the whole purpose and direction of the arrow of the entire progress of the human colossus in the current decade is contained. I don't presume to judge what goal we as a species will begin to pursue in the next decade, it can be both the conquest of the cosmos, and going deep into the mind and matter. But the current task of obtaining an absolute transorative power over the physical world must be completed before 2030 and the direction is clear to me. Not an ounce of pessimism.