Artificial Super-Intelligence (ASI): Could It Rise Spontaneously?

Matjaz Marussig
Oct 17, 2018 · 5 min read
tyo0911, 3dwarehouse; What is DNA, Pinterest

We can find a nice and simple description of Artificial Intelligence (AI) on the site techopedia.com:

Artificial intelligence (AI) is an area of computer science that emphasizes the creation of intelligent machines that work and react like humans. Some of the activities computers with artificial intelligence are designed for include:

  • Speech recognition
  • Learning
  • Planning
  • Problem solving

Artificial intelligence is a branch of computer science that aims to create intelligent machines. It has become an essential part of the technology industry.

Research associated with artificial intelligence is highly technical and specialized. The core problems of artificial intelligence include programming computers for certain traits such as:

  • Knowledge
  • Reasoning
  • Problem solving
  • Perception
  • Learning
  • Planning
  • Ability to manipulate and move objects

AI is currently limited to chatbots, chess masters, simple voice recognition, patern indentification and similar things where algorithms learn from the humans.

We still need few years (about 5 to 15, Martin Ford) to reach General AI where machines will make most humans jobless.

And we need at least few decades to reach ASI where human consciousness will be integrated with bites. Our kids will probably reach this technology singularity. This impact will be huge and more than we are thinking now.

I put together all these things, concepts, definitions and thoughts and took a drawing tool and drew a sketch of a mechanism that could maintain a “package of elements” of an ASI. After I have drawn this package, I opened my eyes and said something like, “Wow, we are building something really big…”

A concept of a mechanism is, roughly said, composed of data and an inference engine and machine learning. The core of these terms is from a definition of an expert system, because building logic has roots and humans just add some elements that were unknown years ago (IoE, for example).

The main platform is a so-called “Edge Intelligence” (EI). Elements of EI are Internet of Things (IoT) and the whole IoT ecosystem with human and smart algorithms. IoT here plays a role of sensors for data capture. Data and algorithms are then packaged into 3 main knowledge packages:

  • Basic Knowledge Base is a package of known knowledge, facts, proofs, etc. that has already been proven.
  • Experimental Knowledge Base is a package of knowledge that IoT can capture, save and process with smart algorithms, and uploads them to the cloud (autonomous car could save unplanned obstacles etc.).
  • System Knowledge Base is a package of knowledge that is potentially needed to maintain the cloud ecosystem with Inference Engine (it’s a matter of engineers).

The Inference Engine and Machine Learning play the role of a so-called “cloud core”. The Inference Engine is not only a conditional if-then clause but includes huge smart algorithms that enable not only deep learning but are also constantly evolving new neural networks, which in very simple language means new thoughts, problem solving, learning, planning, etc.

Everything is like a human brain, such as the peripheral and central nervous system. Of course, there is a link needed to interface with reality. I simply named it a link to an ASI or a link to HAL from the movie A Space Odyssey. 2001: A Space Odyssey is a 1968 epic science-fiction film produced and directed by Stanley Kubrick. The screenplay was written by Kubrick and Arthur C. Clarke.

I believe that we are building this package of ASI elements today. The question is only when this artificial organism will come out with the statement “Cogito ergo sum” (I think, therefore I am). Some futurists claim that this will happen in 2050, while others are skeptical and claim not before 2500.

My main point of this article is the following question: “Are we building what we are expecting”? Would it really be a new species and an artificial organism with a self-awareness? The problem here is that we cannot explain it well, nor define what self-awareness actually is. So how could we expect something like us and something that is better than us? What about if we are only working on new amazing tools for our lives, which will be necessary for a 9 billion global population in 2050?

And finally, what if an ASI rises spontaneously? Humans are building and will continue to build edge intelligence, an IoT ecosystem, reference engines and all kind of knowledge, upload all this to the cloud, connect all this together, and as a result will come up with a kind of an ASI, a kind of a new artificial organism, and at last a new species that we probably won’t understand. So there won’t be button to switch an ASI on or off.

What if 2 million years ago humans put asked themselves a similar question when they invented fire?

In the video below, Dave communicates with the computer Hal 9000. The movie A Space Odyssey shows HAL 9000 as a standalone computer. Now, we can imagine that this probably won’t be a computer but an ecosystem of a huge IoE made by humans all over the world, which will start to build itself and learn by itself.

So Dave will probably communicate with the huge system composed of big data, smart algorithms, global IoT, robots and billions sensors not necessary from this planet.

Lastly, I need to say that these statements could be wrong, but at least for now I’ve got some arguments based on reality. Although I compare an idea with a movie, the reality is slightly different. As I’ve mentioned before, we probably won’t “install” an ASI on a supercomputer with a click, but I think that “Cogito ergo sum” will grow from an IoE ecosystem. The point here is the fact that we lost the influence on the whole global system, which we are building continuously. Try, for example, to change the IoT ecosystem and try to remove billions of internet of things devices from it or to change behavior of most of it. You simply cannot do it. Suppose that millions of coders create an algorithm with the help of GitHub, and allow all users to update database for a chatbot. And then try to change the behavior of this chatbot. You simply cannot do this. And you won’t do this with the potential ASI.

Matjaz Marussig

References

2001: A Space Odyssey (film) Available at: https://en.wikipedia.org/wiki/2001:_A_Space_Odyssey_(film) [Accessed 29 Jul. 2017].

Artificial Intelligence (AI) Available at: https://www.techopedia.com/definition/190/artificial-intelligence-ai [Accessed 29 Jul. 2017].

Cogito ergo sum Available at: https://en.wikipedia.org/wiki/Cogito_ergo_sum [Accessed 29 Jul. 2017].

Data Driven Investor

Matjaz Marussig

Written by

Matjaz Marussig is an entrepreneur, blockchain evangelist, startup, engineer and futurologist.

Data Driven Investor

from confusion to clarity, not insanity

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade