What Is ChatGPT Doing and Why Does It Work? by Stephen Wolfram. Stephen Wolfram is a computer scientist and physicist. He was named a fellow of the American Mathematical Society and is currently an adjunct professor at the University of Illinois Department of Computer Science.
He holds a PhD in particle physics from the California Institute of Technology. Steven then began teaching at Caltech and was the youngest recipient of a MacArthur Fellowship at age 21.
In 1987 he created Wolfram Research and continues to serve as CEO developing both Mathematica and the Wolfram Alpha answer engine. Recently the company announced that Wolfram Alpha would integrate with ChatGPT.
In short, this book is perhaps the first worthy book to address ChatGPT with a full understanding of AI technology. Plus, Steven holds the ability to explain AI, algorithms and equations which will help the reader gain a much better (perhaps the best) understanding of ChatGPT.
Since the first public introduction in late 2022, many have simply asked how does this AI service function? Stephen in fact, lays out in very readable form, the background and basic understanding of neural net technologies. As a scientific discovery, this AI technology has certainly holds great potential to change computing with a potential huge downstream impact to society.
It’s Just Adding One Word at a Time
Since the first public introduction of ChatGPT in late 2022, many have simply asked how does this AI service function? Stephen lays out in very readable form, the background and basic understanding of neural net technologies. As a scientific discovery, this AI technology has seemingly great potential to change computing with a potential huge downstream impact to society.
Perhaps many readers will also be surprised by how ChapGPT actually builds — literally one word at a time. Fascinating beginning to the book. And Models for Human-Like Tasks explains how computers recognize images. In fact, humans can still recognize handwritten numerals better than computers. I would also recommend How Smart Machines Think by Sean Garrish.
From this beginning Steven moves into Neural Nets. This provides an excellent overview to the computational base of generative AI. Steven explains how Generative AI can create classifications via tokens. He demonstrates how this process mirrors human cognition.
The Training of ChatGPT
Once the base of ChatGPT is revealed, the reader will begin to learn how to train ChatGPT. To this end, many of the early criticism is focused upon the source of OpenAI resources. Here, Steven introduces the ability for ChatCGP to tap into Wolfram Alpha.
Since ChatGPT generates linguistic output, it can only follow the data patterns on the web and onto internet that ChatCPT has ingested and becomes a knowledge-base. So what happens when Wolfram Alpha can provide ChatGPT with answers? It becomes a trusted source. So imagine for example The Associated Press providing the same source of trust for knowledge as Wolfram Alpha? This becomes very interesting.
In conclusion, this is simply a must-read for everyone who engages the internet. So, this would be everyone on the planet.