The Ultimate Adversary: Artificial Intelligence, VR & 5G

http://kevbakershow.com/

The post THE ULTIMATE ADVERSARY: Artificial Intelligence, VR & 5G appeared first on Kev Baker Show.

THE KEV BAKER SHOW, EP#767

SPECIAL GUEST – ANTHONY PATCH

Welcome back & in this episode of The Kev Baker Show we continue our discussion on the topics of Artificial Intelligence (AI) & other technologies including Virtual Reality (VR) & the roll out of 5G.

THE ULTIMATE ADVERSARY.

Artificial Intelligence is advancing at break neck speed & this was demonstrated again over the weekend when Elon Musk’s OpenAI took on some of the worlds best players at online game Dota 2….. and won!

We previously covered, in depth, how Google Deepmind’s AlphaGo AI managed to outwit one of the worlds top Go players. Over the course of 3 games of the ancient Chinese game of Go, the Deepmind AI took on and won against a player regarded as one of the best, Ke Jie. Not only that, but it came up with moves previously unseen by human players. This achievement by AlphaGo was heralded around the world as an indication as to just how powerful these AI systems are becoming.

You might not think that this is all that special, but take into account that the game of Go is regarded as mans’s most complicated strategy game. The eventual goal of Deepmind is to create a machine that can “solve intelligence & make the world a better place“.

AI taking on players at Go & winning was one thing, but arguably, OpenAI just went one better.

OpenAI is the open source AI company set up by Elon Musk. Musk is not stranger to the world of AI & his remarks are usually met with controversy due to his pessimistic views on the subject. On this occasion, its the actual AI that has people talking.

OpenAI just stunned the gaming world by beating a professional player at the game of DOTA 2. It had been thought that an AI bot wouldn’t be able to even come close to beating a skilled human player.

OpenAI went up against Ukrainian Danylo “Dendi” Ishutin…. and won! What is even more impressive, or scary, is the fact that the AI only started learning the game TWO WEEKS AGO!

This marks an important step in the progress of AI technology.

So, how does an  AI bot learn how to play a game that is even difficult to explain to humans?

The answer to that questions is Adversarial Neural Networks (ANN).

ANN’s can be trained to recognise patterns in data including speech, text data, or visual images & are the the basis for a large number of AI under development in recent times.

We have talked in the past about “deep machine learning” and “recurrent neural networks”, but in the case of the AI bot that just won at DOTA 2, it the ANN – Adversarial Neural Network.

Think of it like this – there are 2 AI bots pitted against each other, in this case, the game of DOTA 2. Neither one of the AI start with any knowledge of the task at hand or the game they are playing. What happens is that through trial and error, and by always facing an opponent of the very same skill level, they learn from their mistakes and from each other, eventually leading to them developing an understanding of the task at hand.

The AI has one goal in mind, and that is to win, at any or all costs! It is designed to compete against, not work alongside, anything it encounters & that includes US!

Perhaps its just me but this technique used to train the AI puts us in a very disadvantageous position down the line. I say that due to the fact that the very nature of the machine is to win, win, win. There is no room for error on its part & will keep going until it achieves the only outcome it desires – victory. In the gaming arena this is fine, but what happens when this transfers into an artificial super intelligence, or as it also known “Strong” or “Wide” AI?

Recently we seen reports of both Google & Facebook having to pull the plug on AI chat bots because they had developed their own language that programmers couldn’t decipher. There are also stories about AI developing the ability to deceive its users, all to gain an advantage so that it can “win”.

All AI’s are driven by goals which are met with rewards, and the AI will go to any length to achieve its goal so that it attains the reward it seeks. Just how far will these machines go to reach their goals, that is the question?

This is why people such as Elon Musk, rightly in my opinion, warn about the existential threat posed by AI.

Anthony Patch & I discuss Musk’s warnings and offer our own theories as to why a man so concerned about AI is in fact one of the people bringing about its inception.

Musk has in the past spoken about the need for humans to integrate with the AI via something he calls the neural lace.

‘Nobody likes being regulated, but everything (cars, planes, food, drugs, etc) that’s a danger to the public is regulated. AI should be too,’ Musk also tweeted on Friday night.

In 2016, Musk said during a media conference that human beings may eventually need to add a digital layer of intelligence to our brains in order to prevent robots from turning us into ‘house cats,’ according to Business Insider.

‘I don’t love the idea of being a house cat, but what’s the solution? I think one of the solutions that seems maybe the best is to add an AI layer,’ he said. ‘A third, digital layer that could work well and symbiotically’ with the rest of your body.

Musk said that a ‘neural lace,’ a concept that proposes turning the human mind into a wireless computer-brain system, may be necessary in order to keep up with rapidly advancing AI technology.

SOURCE

Musk talks openly about “a third digital layer that could work symbiotically” with the human & machine. Contraversially, i do see where he is coming from. For the machine to truly appreciate & understand human emotion, the best way would be for it to “experience” these emotions via us. With us being a part of the AI system, that in itself would be a safety feature to prevent the potential for AI to wipe us out.

Let me make it abundantly clear that i am not advocating for merging with machines, im merely explaining my opinion on Elon Musk’s idea that we need something like the Neural Lace to prevent our downfall.

There is another, more nefarious, way of integrating us into the AI system, and it comes in the form of our DNA.

Anthony Patch has discussed the potential for a third strand of DNA to be utilised for just this function. We all may be unwittingly wired & ready to go already! The metal particulates that we are all subjected to via the aerosol spraying could be used for this very purpose.

The use of Nanotechnology could literally wire us for the matrix & all that awaits is the activation of this technology.

Could 5G by the switch that is about to be flicked & welcome us all into the matrix?

The post THE ULTIMATE ADVERSARY: Artificial Intelligence, VR & 5G appeared first on Kev Baker Show.

http://kevbakershow.com/