AI Since Omni: From the Cray-1 in 1976 to GPT-3 in 2020

What happens when you ask AI to write its own history? An article titled "Monster Machines: Artificial Intelligence" by Michael Edelhart in Omni, August 1983 provides inspiration for just this prompt. Edelhart writes about the supercomputers built by Seymour Cray, an important milestone on the road to the current juggernaut status of AI. While ChatGPT allows us to learn more about its own development in a way that extends beyond the boundaries of different companies in the space, its writing style still stands in sharp contrast to that of a human, such as Edelhart.

Excerpt, "Monster Machines: Artificial Intelligence" by Michael Edelhart:

How long does it take to count all the atoms in the universe?

For a human being, the task is impossible. A home computer might be able to manage the effort, given a few millennia. A powerful IBM mainframe could spend a few years at it. But Seymour Cray's computer can provide the total during a coffee break.

In fact, Cray's $8 million computer, the Cray-1, found the largest known prime number in just 20 minutes. The 13,395- digit-long integer is larger than the number of atoms in the universe.

Now Cray is about to introduce a machine that will make everyone else's computers look like children counting on stubby fingers. The new $12.6 million Cray X-MP, due this fall, will be two to five times as fast as the Cray-1. And within two years we can expect yet another monster machine, the Cray-2, which will blast through more than half a billion calculations per second.

Cray, fifty-four, designs and builds supercomputers, machines that process information more than 100 times faster than typical large computers. Since the early Sixties, when Cray and colleagues developed the Model 6600, the world's first supercomputer, he has dominated the field in a way few individuals can. He founded both the leading supercomputer builder in the world - Cray Research, Inc. - and its principal competitor, Control Data Corporation. While computer developers in California's Silicon Valley, Cambridge, Massachusetts, and Stanford, California, have become household names, Cray has quietly gone about the business of creating number crunchers in his small Chippewa Falls, Wisconsin, laboratory. In a computer world where everything becomes outmoded instantly, Cray's computer has been the world's fastest for seven years, the company claims. Since it was released in 1976, nothing on earth has caught up with the Cray-1 for certain applications.

The content below was generated using ChatGPT, and in the interest of transparency and learning, the heading represents the prompt that created it.

What is the timeline of the development of AI since Cray-1?

Since the Cray-1 supercomputer was introduced in 1976, and the field of Artificial Intelligence (AI) had already been established, I will provide a timeline of some of the major developments in AI since then:

• 1986: The first conference on Neural Information Processing Systems (NIPS) is held, which focuses on the use of neural networks in machine learning.
• 1987: The concept of backpropagation is introduced, which greatly improves the training of neural networks.
• 1997: IBM's Deep Blue chess-playing computer defeats world champion Garry Kasparov.
• 2002: The first DARPA Grand Challenge is held, in which autonomous vehicles compete in a 142-mile desert course. None of the vehicles finish the race.
• 2006: Geoffrey Hinton and his colleagues develop deep learning techniques that allow neural networks with multiple hidden layers to be trained efficiently, revolutionizing the field of computer vision.
• 2011: IBM's Watson defeats two human champions on the quiz show Jeopardy!.
• 2012: The first ImageNet Large Scale Visual Recognition Challenge is held, spurring advances in computer vision through the development of large-scale datasets and deep learning models.
• 2014: Google's DeepMind develops AlphaGo, a computer program that defeats a human world champion in the board game Go.
• 2016: DeepMind develops AlphaGo Zero, which learns to play Go without any human guidance and defeats the earlier version of AlphaGo.
• 2018: OpenAI develops GPT-2, a language model that generates realistic human-like text.
• 2019: Google's BERT (Bidirectional Encoder Representations from Transformers) model improves natural language processing (NLP) capabilities by pretraining on large datasets.
• 2020: OpenAI develops GPT-3, a language model that can generate highly coherent and diverse human-like text.

These are just a few of the many significant developments in AI since the introduction of the Cray-1 in 1976. The field continues to evolve rapidly, with ongoing research and advancements in machine learning, deep learning, natural language processing, computer vision, robotics, and more.

Can you rewrite this in an editorial style?

Certainly! Here's a rewritten version in an editorial style:

Since the introduction of the Cray-1 supercomputer in 1976, the field of Artificial Intelligence (AI) has undergone significant developments. In the years that followed, AI research has produced numerous breakthroughs that have transformed the way we use computers and interact with technology.

In 1986, the first Neural Information Processing Systems (NIPS) conference was held, which marked a turning point in the use of neural networks in machine learning. A year later, the concept of backpropagation was introduced, which greatly improved the training of neural networks.

In 1997, IBM's Deep Blue chess-playing computer defeated world champion Garry Kasparov, marking the first time a computer defeated a human world champion in a competitive game.

In 2014, Google's DeepMind developed AlphaGo, a computer program that defeated a human world champion in the board game Go, a feat previously thought impossible for a machine. Two years later, DeepMind developed AlphaGo Zero, which learned to play Go without any human guidance and defeated the earlier version of AlphaGo.

In recent years, language models have been a major focus of AI research. In 2018, OpenAI developed GPT-2, a language model that can generate highly coherent and realistic human-like text. In 2019, Google's BERT model improved natural language processing capabilities by pretraining on large datasets. And in 2020, OpenAI's GPT-3 pushed the boundaries of what language models can do, generating highly diverse and creative text.

These are just a few examples of the many advances in AI that have taken place since the Cray-1 was introduced in 1976. As research continues, we can expect even more transformative breakthroughs that will reshape the way we live, work, and interact with technology.

What are your thoughts on this trajectory?

As an AI language model, I don't have thoughts or opinions in the way that humans do. However, I can provide some context about the trajectory of AI development based on my training data.

The trajectory of AI development since the introduction of the Cray-1 has been characterized by significant breakthroughs and rapid progress in many areas of research. The development of neural networks, deep learning techniques, and language models like GPT-2 and GPT-3 have greatly improved our ability to perform tasks such as natural language processing, image recognition, and game playing. These advances have opened up new possibilities for applications in fields such as healthcare, finance, and transportation.

At the same time, the trajectory of AI development has also raised concerns about the ethical and societal implications of these technologies. There are concerns about the potential for AI to be used to automate jobs, perpetuate bias and discrimination, and even be used for malicious purposes. As AI technology continues to evolve, it will be important to carefully consider the potential benefits and risks and to develop responsible and ethical frameworks for its use.

Overall, the trajectory of AI development since the Cray-1 has been marked by both tremendous progress and ongoing challenges. As research continues, it will be important to navigate these challenges and ensure that the benefits of AI are realized while minimizing the potential risks.