Why all the internet giants are shutting down their AI programs?

By Skepsion.com

Machines walking around streets, killing humans and taking over the world are the feast for science fiction writers and directors. What if I tell you that it’s not just a fiction anymore? Movies like a Space Odyssey, The Terminator, The Matrix, Transcendence, Ex Machina and Avenger’s Age of Ultron all portrays how the computers would evolve to an extent that the humanity would not be able to control or compete with it.

“Success in creating AI would be the biggest event in human history,” wrote Stephen Hawking in an op-ed, which appeared in The Independent in 2014. “Unfortunately, it might also be the last, unless we learn how to avoid the risks. In the near term, world militaries are considering autonomous-weapon systems that can choose and eliminate targets.” He added.

All this would have been less scary if Facebook Artificial Intelligence didn’t start creating its “own language!”

The AI did not start shutting down computers worldwide or something of the sort, but it stopped using English and started using a language that it created. The two chatbots came to create their own changes to English that made it easier for them to work – but which remained mysterious to the humans that supposedly look after them.

What were they talking?

The AI’s were instructed to negotiate with each other and improve their functioning. But they started using comprehensible English to create their own “shorthand”, according to researchers. The conversation, however, looks odd and it looks like a kid is learning to talk for the first time.

Bob: i can i i everything else . . . . . . . . . . . . . .

Alice: balls have zero to me to me to me to me to me to me to me to me to

Bob: you i everything else . . . . . . . . . . . . . .

Alice: balls have a ball to me to me to me to me to me to me to me

Bob: i i can i i i everything else . . . . . . . . . . . . . .

Alice: balls have a ball to me to me to me to me to me to me to me

Bob: i . . . . . . . . . . . . . . . . . . .

Alice: balls have zero to me to me to me to me to me to me to me to me to

Bob: you i i i i i everything else . . . . . . . . . . . . . .

Alice: balls have 0 to me to me to me to me to me to me to me to me to

Bob: you i i i everything else . . . . . . . . . . . . . .

Alice: balls have zero to me to me to me to me to me to me to me to me to

The chatbots also learned to negotiate in ways that seem very human. They would, for instance, pretend to be very interested in one specific item – so that they could later pretend they were making a big sacrifice in giving it up, according to a paper published by FAIR. After learning to negotiate, the bots relied on machine learning and advanced strategies in an attempt to improve the outcome of these negotiations.

Facebook’s experiment isn’t the only time that artificial intelligence has invented new forms of language. Earlier this year, Google revealed that the AI it uses for its Translate tool had created its own language, which it would translate things into and then out of. But the company was happy with that development and allowed it to continue.

According to the linguist Mark Liberman, the sentence included phrases, words and structures similar to human language.

 

Why is this a BIG DEAL?

At this point after reading all the gibberish chat by the AI chatbots you will probably thinking why is this a big deal? why should i bother? How these stupid chat program can terminate humans? If you got this question in your mind then you have no idea on AI and how it works. Don’t worry here i have simplified explanation for our non techy friends. Even Mark Zukerberg under estimated it. Elon Musk have a clear understanding on it and warned Mark by tweeting this:

Zuckerberg blasts @elonmusk warnings against artificial intelligence as 'pretty irresponsible' https://t.co/DzPjvBym7W @svbizjournal #ai

— Darren Cunningham (@dcunni) July 25, 2017

Machine learning is the method of teaching a machine to do a certain task and make decision without programming it explicitly. The computer will behave like a learning baby trying to understand everything from the environment and the experience that its subjected to. The base algorithm will be like this

 “Analyse all data and inputs > store everything in a structured database > note all the mistakes it makes > never repeat the mistakes > repeat learning”.

In our primary classes we have learned that “computer never make mistakes” . In this case we are designing an algorithm where it make mistakes, but only once. In the next run it wont repeat it. Thats how humans work but the process is too slow. A computer can do this in a matter of seconds.

Based on all these information that it has learned, an AI can create its own algorithm for further progress. This idea of Artificial intelligence can really be helpful in many scenarios. If we are planning a rover mission on Pluto sending data back and forth to control the machines is insanely stupid idea. At this point the rover has to make decisions its own. The rover have to analyse the environment conditions to move across the terrain and choose its own path.  An AI connected to internet with a giant database can learn everything from internet and make more complex decisions like weather forecast (even a catastrophic event prediction if it has enough information fed to it).

 

How it can turn against us?

So far all you read is about machine learning and its awesome features and how it can be beneficial to human race. Now here comes the dooms day scenario. As mentioned above Facebook AI bots have created their own language. They claimed it is the most efficient way of information interchange with low resource consumption. It proved that without proper sentence formation and linguistics the information can be sent to peer and a proper communication can be possible. They have gone too far in analyzing a communication method and making it even better. So straight to the point, What if they were developed with full privilege to grasp everything from internet. sure they will find their own method to find and communicate other AI systems and will sync their database. At this point every AI will be of same intelligence level. They will continue to search for information and sure they will try to eliminate every complex problem. In the end they will realize humans are the big threat to the environment. That is gonna be the end. This super intelligent AI will develop their own protocols to eliminate human threats. You may be thinking, “we haven’t created any humanoid robots, so whats the problem”. That is not necessary these intelligent systems can control the internet, military networks, nuclear warfare systems, rest is self explanatory. AI’s can evolve and develop their algorithm to a whole new level like a cell mutation in living organisms.

 

Tech leaders and genius mind who warned about AI

STEPHEN HAWKING

Back in 2014 during an interview with BBC, professor Stephen Hawking claimed that the AI could end mankind.

“The development of full artificial intelligence could spell the end of human race. It would take off on its own, and redesign itself at an ever increasing rate. Humans who are limited by slow biological evolution, couldn’t compete and would be superseded”

A year later, speaking at zeitgeist Hawking provided a timeline telling: “computers will overtake humans with AI at sometime within next 100 years. When that happens, we need to make sure the computers have goals aligned with us.”

ELON MUSK

In the April 2017 edition of Vanity Fair, Elon Musk, presented “strawberry picking scenario” along with his many concerns about AI.

“let’s say you create a self- improving AI to pick strawberries and it gets better and better at picking strawberries and picks more and more and it is self-improving, so all it really wants to do is pick strawberries. So then it would have all the world be strawberry fields. Strawberry fields forever. No room for human beings.”

Recently he tweeted saying: “AI is a greater threat than North Korea.”

If you're not concerned about AI safety, you should be. Vastly more risk than North Korea. pic.twitter.com/2z0tiid0lc

— Elon Musk (@elonmusk) August 12, 2017

SIR TIM BERNERS LEE

The architect of the World Wide Web, laid a nightmarish scenario where the AI could become the ‘masters of the universe’, at a conference in London.

The Scenario is machines controlling the financial world and decide which companies to acquire and took this to its logical conclusion. “ so when AI starts to make decisions such as who gets a mortgage, that’s a big one or which companies to acquire and when AI starts creating its own company, creating holding companies generating new versions of itself to run this companies”

“so you have survival of the fittest going on between the AI companies until you reach the point where you wonder if it becomes possible to understand how to ensure they are being fair, and how do you describe to a computer what that means anyway?”

BILL GATES

Last year in the BBC interview Bill Gates revealed his thoughts on AI.

“I’m in a camp that is concerned about super intelligence. First the machines will do a lot of jobs for us and not be super intelligent. That should be positive if we manage it well. A few decades after that though the intelligence is strong enough to be a concern. I agree with Elon Musk and some others on this and don’t understand why some people are not concerned.”

STEVE WOZNIAK

Speaking to Australian financial Review, wozniak said, he most notably claimed “humans will be robots pets”. He said: “I agree that the future is scary and bad for people. If we build these devices to take care of everything for us, eventually they’ll think faster than us and they’ll get rid slow humans to run companies more efficiently.”

Although he took a U-turn claiming that “they are going to be smarter than us and if they are smarter than us they’ll realize they need us.”

Can super intelligent computers really outrun humans?

Talking about human brains. Our brains were evolutionarily programmed to carry out set of certain functions like finding food and shelter, identifying friend vs. foe and finding a mate and multiply. Well now our brain has managed to develop, evolutionarily, to 100 billion neurons per average brain and 7000 synaptic connections per neurons. But when we compare the specs of a modern day hardware to the 1 million old brain, we are limited.

In 2000s, the complexity of the entire internet was compared to a single human brain, but the scenario has changed. But still brain is 100,000 times more energy-efficient than computers. It is estimated that the computers will surpass capability of human brain by the year 2040. We are limited by our hardware but machines are not.

Neuron, for example, can only fire up to 200 times per second whereas computers can run billions of cycles per second. Signals in the neuron travels about one millionth of the speed of optic fiber.

But when it comes to memory, biology takes over. Consider DNA as a memory storage device. All the information required to run the body is stored in the DNA of each of the six trillion cells. It is estimated that in 2015 the information on all the computers in the world, if stored in DNA, would fit in the back of an SUV. Already DNA is used to store non biological information. Not only the space but also the durability is no match for the modern computers. The hard drives and other, memory storage devices decompose after 20-30 years, whereas intact DNA has been  found close to a million years old.

Scientist a year ago tried to make a supercomputer mimicking the human brain. They needed 82,000 processors running on one of the fastest supercomputers to mimic just second of normal human brains activity. This shows that the arising of the AI would take some time. We are at the baby stage of creating Artificial Intelligence, but with AI creating languages, military trying to use AI for weaponry and the world being connected to each other, anything can happen.

So yeah, shutting down AI is the only way to slow it down (No we cant stop it). With further advancement in technology at some point even primary school students will be learning to create simple AI programs. Then all it will take is a mad psychopath with good coding skill to end human race.

“Artificial Intelligence will work with humans to bring global peace or maybe it will eradicate humans for global peace”