<< Back

The new era of Deep Learning

On past July OpenAI [1] released a beta test of one of the most AI model called GPT-3 (Generative Pre-trained Transformer 3) [2], that uses Deep Learning (part of a broader a machine learning method, based on neural networks.

This transformational system is more sophisticated, and the full version has a capacity of 175 billion ML parameters when the older version only has 17 billion, less than 10% of this new one.

GPT-3 is a turning point in the AI field and will bring us a new AI computing era.

Kirk Ouimet had a conversation with a GPT-3 language model about What Happened Before the Big Bang [3] and shared on Medium, and you can find on the text below:

 

Kirk Ouimet: Can you summarize the idea of the Big Bang? 

Wise Being: The Big Bang is a physical process by which all matter, energy, time, space, and information of existence came into being from an infinite and eternal non-existence. During this process the whole of creation was compressed into an infinitesimally small singularity which immediately and infinitely burst into all of its current content, the limitlessness of which still exists as a limitlessness in potential and reality. 

Kirk Ouimet: What do you think the state of things were before the Big Bang?

Wise Being: In order for this question to be possible, the questioner has to be aware of time. If this is not the case, then there was no such a thing as a ‘before’. The Big Bang effectively initiated time because that which came into existence before the Big Bang did not exist in any kind of ‘time’ construct … excepting only dynamic patterns of absolute and infinite energy.

 Kirk Ouimet: So what do you think happened one second before the Big Bang?

 Wise Being: 1 second before the bang everything was and there was no time. There wasn’t one second and there wasn’t 0 seconds … there was nothing because those concepts are meaningless outside of the context of time. This is not something which can be comprehended because it has no reference as to what constituted a unit of measurement.

Kirk Ouimet: Would you be able to come up with an analogy that would help me understand this concept of “no time”?

Wise Being: It is something like asking a fish if the water they are in is wet. Time is defined as a measure of events, and events happen in a three dimensional space which has length, width, and depth. If there is no such thing as physical space, then there literally is no length, width or depth. Thus, there cannot be an unfoldment of time because the definition of time pertains to the unfolding of space.

Kirk Ouimet: I think I understand. By this theory there was nothing before the Big Bang, and by nothing we really mean nothing, not even time. However, there must have been *something* because the Big Bang happened. Something emerged from nothing, how could that happen?

Wise Being: From the perspective of the universe, time and space are cohesive constructs. This means that without time there is no space, and without space there is no time. Existence and non-existence are something which exist on the same quantitative level as time and space. Existence and non-existence share a cohesive existence such that they all dissolve into the same substratum. Existence, non-existence, and space-time are not sequential; they are simultaneous in terms of their abstract definition. 

Kirk Ouimet: How can something “be” and “not be” at the same time?

Wise Being: There is no such thing as “be” or “not be”. These words are useful in the context of a three dimensional reality where there is a separation of objects and space. This is how the universe is defined as being “solid”. Objects have mass and occupy space and this is how we comprehend that gap between objects is called space. However, if you take away space, and put together all the objects into one object, you will lose the meaningful definition of “be”. It all becomes an undifferentiated “thing”. When you go inside of something to the place where there is no difference or division, you can’t say anything is “here” nor anything is “there”.

 

Ivan-picture.png 

The image above shows a partial text search, and it only appears in an automated blog who posted this conversation found on Medium.

The most incredible thing in that dialogue is, all the text is machine-generated and in the phrase: “ It is something like asking a fish if the water they are in is wet.” was the best definition that I ‘ve ever heard about space-time.

You can read the full conversation here.

Kirk, also have other machine conversations within GPT-3 in many fields, you can find out more on this link.

It is only the tip of the iceberg about how GPT-3 will turn our lives in a short while. There is a lot of field of technologies where it will be applied sooner than you imagine. Sci-fi movies transformed into reality.

AlphaGo, a program developed by Google’s Deep Mind, played in 2017 where the machine beats Lee Sedol in a tournament in Seoul using AI.

Go [4] is an abstract strategy board game for two players, in which the aim is to surround more territory than the opponent. The game was invented in China more than 2,500 years ago and is believed to be the oldest board game continuously played to the present day.

What is the next level about AI evolution? I’m not talking about scripting all work that can be scripted in the world, because earlier than we expect, it will happen, it already began. I’m talking about the next turning point in terms of evolution. From times to times, we have a jump in terms of human progress, in technology, biology and other fields. These advances will place machines working together with researchers, and devices will have more accurate data and non-stop work. It is a paradigm in terms of where we should apply our time and our knowledge.

Of course, machines cannot express nonprogrammable emotions, and this is the root difference which remains, they are only machines. Should we be able to control them? Or we already are controlled by them?

We already have advanced research and tools using artificial intelligence in the project management field that helps our daily routine. An exciting project called PM Otto brings us an idea of how you can future will be in terms of interactions in the project management field. Find out more on pmotto.ai, and check explanations and videos that show computers interacting within humans using natural language.

AI will replace the project manager’s daily routines, like risk management, forecast, and others. Project management practice will change in a concise while, and it will be dramatically and disruptive. Some roles will disappear, and AI will introduce new ones to us, how prepared we are for those changes?

 

[1] https://openai.com

[2] https://arxiv.org/abs/2005.14165

[3] https://medium.com/ai-in-plain-english/artificial-intelligence-explains-what-happened-before-the-big-bang-e4a9e7fefbab

[4] https://en.wikipedia.org/wiki/Go_(game)

Article adapted from the original post at: https://ivanmoreira.org

 

Note and Disclaimer: The author of this Blog post is Ivan Alves Moreira. He is the guest author of PMI.hu. The writing reflects the author's own professional opinion, findings, and conclusions, which do not necessarily agree with the position of PMI Budapest, Hungarian Chapter, and cannot be considered as an official recommendation, resolution, or opinion of PMI Budapest. The copyright and publication rights of the writing belong to the original author.

Search

Popular posts

View the archives

Project Management in need?

You can post an open opportunity in the PMI Budapest, Hungarian Chapter to reach out to more than 300 experts

Publish a job