For decades, artificial intelligence has been used to perform specific functions, like self-driving a Tesla or auto-processing an insurance claim. These single purpose AI and machine learning technologies have sat silently behind the scenes, just doing their thing, with little awareness from the likes of you and I.
Then came ChatGPT, the first of a new and disruptive form of artificial intelligence that is a general purpose technology in the hands of everyone. Known as “generative AI,” this new breed of AI is different to the single purpose AI that went before it.
Because this form of generative AI is all about words. And the use of language that’s on par with human levels of communication.
Quite simply, generative AI is a statistical prediction machine, also known as an LLM or Large Language Model, because that’s exactly what it is! The LLM has read millions of pieces of human writing from across the internet, and broken every sentence and word down into their smallest units or fractions of words, for example, “football” becomes “foot” and “ball.”
The AI then creates a massive multi-dimensional map of the way that words are combined to form many many different meanings. Think of how many ways you can use the words “foot” and “ball” in a different context.
But, there’s more of generative AI than just building a mega spreadsheet of word combinations. What unlocked their abilities to write as fluently as they do today is a revolutionary AI tool called “the Transformer.”
The Transformer model was first published by a group of eight AI researchers at Google in June 2017. Before Transformer, AI development focused on trying to replicate the way the human brain works with neural networks. But Transformer changed everything.
Transformers are the ability of an AI system to process an entire sequence of words, analysing all its parts, breaking it down and “understanding” the meaning of the words. By that I mean that the transformer is able to work out what parts of a sentence or article are the most important in defining the meaning of it.
A key concept of the Transformer architecture is called “self-attention”. This is what allows LLMs to understand the relationships between words. Self-attention looks at each word, or token, in a body of text and decides how they all relate to each other and which are most important to understanding its meaning.
I asked Claude to explain how this works using a simple example:
The quick brown fox jumps over the lazy dog.
This sentence would first be broken down into individual tokens, or words. So the model would see:
[“The”, “quick”, “brown”, “fox”, “jumps”, “over”, “the”, “lazy”, “dog”]
The model then uses self-attention to understand how these words relate to each other. It creates numerical vectors for each word, and maps out their relationships based on context clues. So it would map that “fox” and “jumps” go together, as do “dog” and “lazy”.
With this vector map of word relationships, the model can then start generating new sentences based on its understanding. It might take the core of “fox jumps over dog” and create a new sentence like: “The playful kitten leapfrogged over the napping puppy.”. The model is able to mix words around contextually and correctly due to building up its knowledge from the vector mappings created by self-attention. It can continuously improve at natural language generation the more examples it has to learn from.
➜ Here’s The Thing: Google’s 11-page research paper in 2017 marked the start of the generative AI era. Without it, there would be no ChatGPT!
FREE: The Beginner’s Guide To ChatGPT
Every week, 15k get Wiser! the newsletter about AI, emerging technologies and the future of work.
Join the mailing list for FREE and download The Beginner’s Guide to ChatGPT. PLUS, a bonus download of The Utility Of Emerging Technologies, containing over 25 case studies of consumer brands adopting emerging technology to build customer engagement.
Sign Up For FREE Here …and get your Beginner’s Guide to ChatGPT TODAY.
The Rise of Generative AI
OpenAI were not the only ones building large language models, but they were the first to market with a chatbot that allowed you and I to talk directly to the AI.
OpenAI is now valued at $80 billion, up from $30 billion just six months ago. In less than a year, ChatGPT has opened up a whole new market sector that is forecast to be $1.3 trillion by 2032. To the novice and AI newcomer, you’d think that ChatGPT is the be all and end all of generative AI technologies. But you’d be wrong.
There are many alternatives to ChatGPT. A leading alternative is called Claude from an AI startup called Anthropic. Claude does much the same thing as ChatGPT but it is also different. Primarily it’s positioned to offer a safer, more responsible alternative to other AI models like ChatGPT.
Anthropic was founded in 2021 by a brother and sister duo, Dario and Daniela Amodei, who were both senior employees at OpenAI. They made no secret of their concerns about the direction of travel for OpenAI’s AI research and eventually left to create a more responsible AI platform, which is Claude.
So, what is Claude?
Claude is a versatile AI model capable of various tasks, including writing, answering questions, and collaboration. Claude outperforms most other AI models on standardised tests.
One of the key differentiators of Claude is its focus on safety. Anthropic has implemented safety guardrails to address issues of bias, inaccuracy, and unethical behaviour. This is achieved by, what they call, their Constitutional AI system. It’s an AI model built on a set of principles and values that discourages toxic or biased responses.
Claude self-learns right from wrong by testing itself against these principles, much like a human does when they develop their own moral compass. This is different to the traditional approach of using human moderators to manually mark and correct “bad” content.
Apart from safety, Claude offers several other advantages to ChatGPT. The biggest feature difference is that it can handle up to 100K tokens per prompt, equivalent to around 75,000 words or the length of The Great Gatsby novel.
What this means is that you can give Claude 10 times more information to process in a single go, making it capable of managing complex multi-step instructions and large amounts of content. With ChatGPT, you have to break this down into multiple steps to achieve the same end.
Ok, if you’re now thinking, how do I get my hands on Claude, there’s a couple of ways to do it.
First, go straight to Anthropic’s home page and hit the “Talk To Claude” button.
Or third, which is my recommendation for you, is to head over to Poe. This is a hub for chatbots and generative AI. There are literally dozens of different AI systems you can access, including multiple versions of Claude. They’re all slightly different depending on what you want to use them for. With Poe, you can try them all, using the same prompt to see how the answers differ.
Poe.com is the best place to try out all of the Claudes, and many other generative AI systems, including image generation and recognition, not just words. It’s free to signup and you will find dozens and dozens of different AI models to play around with.
About The Author
Rick Huckstep has worked in technology his entire career, as a corporate sales leader, investor in tech startups and keynote speaker. From his home in Spain, Rick is thought leader in artificial intelligence, emerging technologies and the future of work.
🤔 Join The Mailing List and Get Wiser! every week (and your free eBook)