Did Facebook almost invent a new language?

Are robots smart enough to code?

With AI you can have a robot do a lot of things, coding is one of them. Facebook saw the opportunity to have robots do the coding for them. Sounds like a great idea. There are a lot of benefits by letting a robot do the coding for you.

They can code 24/7 without complaining

They make no syntax errors (unless the instructions given by the humans are wrong)

They work fast then humans

Those 3 points alone are already enough the replace some coders by robots, right? Facebook did that and they let the robots communicate with each other in coding language. So robots are definitely smart enought to code for Facebook and any other company in the world. You can teach a robot anything you want and with the combination of AI they become faster and smarter everytime they preform the task. Amazing! This is the future :)

Did they start with coding? Or was this learned by them?

The robots of Facebook started communicating in Englisch with each other. But after sometime they made their own language (which is not allowed ofcourse) and used that to communicate. This was made illegal by the robots so Facebook shut off the experiment. But why? Doesn't it make more sense to talk in coding langauge?

You could let them talk in English, Arabic or Dutch but when the robots will discuss the coding language, they have to do that in coding language. Makes sense? If we need to explain an HTML line, we use the coding line and explain what it means. Robots already know what is means, but they have to communicate the written rule of code. If they need to switch to English after that again, it will lose some time.

AI will learn what it wants

There are some down sides. AI will learn of the things they do, but what if they learn things you dont want them to learn? Like a languague which we don't understand? This is what happend with Facebook AI robots. They developed a new language which they comminucated with to each other. Even the Facebook developers where not able to decrypt the langauge the robots developed. Facebook then decided to stop the experiment. Was this a smart move?

I don't think so. They could have stopped the robots activily develop code for the website, but isn't this a great way to experiment how robots with AI will work if we let them develop their own language? What I would love to see is that they continue the experiment in a project to see how this will develop. This will give us some insight in how robots deal with problems, how they develop new things and maybe how we could work do more work in less time.

Hopefully some other big company is willing to put a group of AI robots together and let them develop their own language, let them develop new tools for us or even make new rules.

Isn't it dangerous?

Could be. If it is done in a closed down, monitored area I think we will be safe. If we do the experiment in the open world where the robots could do what they want, yes it will be dangerous then. For now I think Facebook or Google or Apple or any other company with the ability to test this, should definitly do that. Maybe they will both develop the same language, which could be a new, universal coding language? Or they develop a new way to have a car run on water?

This is the future

This way of coding, learning or anything is the future. AI will become the next big thing in the coming years. Lets hope we will use the AI technology for the best of the people and earth instead of using it for terrorism or any other bad cause.

Just think about the possibilities of leting a robot do tasks for you. Cooking diner, vacuumclean the floor, get groceries, walk the dog.. This all could be done with AI and we are on the edge of an very interesting development with robots. This could be your new how robots who will do any task you give them, great!!