All Aboard the ChatGPT-4 Express: One AI Journey You Can't Afford to Miss!

Overview

Unless you are living in a cave, you may have heard about an internet revolution: a conversational artificial intelligence tool called ChatGPT.

Chat GPT is an generative artificial intelligence tool that was developed and released by OpenAI team in November 2022.

In my 25 years of IT experience, I don’t think I’ve ever seen a tech tool that lit up the whole internet in a very short time as ChatGPT.?The closest that comes to my mind was the release of Google search to public in the late nineties.

Not very often a technically superior tool would land in the hands of the general public which helps them in unimaginable and unfathomable ways - from creating a travel itinerary for a summer trip to Athens, summarising a legal document in layman's words, developing a self-help plan for losing weight, analysing code for security and performance bugs, designing application’s data models, comparing and contrasting technologies for a specific use case, writing complaint letters to a Twitter CEO and so on - the list goes on.

Introducing ChatGPT

ChatGPT is a conversational agent (chatbot) built on GPT (Generative Pre-trained Transformer) architecture, and is capable of generating human-like text based on the user’s prompts. It is an instance of a Large Language Model (LLM), designed for conversation with a specific goal of generating relevant and safe content while engaging in a meaningful dialogue.?

The models are fed with vast amounts of text data and learn to predict the next word in a sentence. It is trained using a diverse range of internet text, but it can also be fine-tuned with specific datasets for various tasks. Through this process, the models learn parts and parcels of human language text: grammar, punctuation, syntax, facts about the world, and some degree of reasoning ability.?

LLM is a broad term that refers to any large-scale model trained to understand or generate human-like text. These models are characterized by their vast number of parameters and their ability to handle a wide range of natural language processing tasks. LLMs can be based on various architectures and training methods.

AI Race

With the release of ChatGPT to the public, a sudden race in the space of search using artificial intelligence emerged overnight. ChatGPT has become a disruptor to many industries and no less a threat to Google’s search. If not already, most AI backed up tools similar to ChatGPT will? disrupt many industries in the coming years. Mounting under tremendous pressure and possibly to save the search leader status, Google decided to unleash their version of conversational generative artificial intelligence agent called Bard? (https://bard.google.com) which was publicly available in May 2023.?

Microsoft, in the meantime, committed to invest $10bn in ChatGPT, over and above their initial investment of $3bn since 2019. Their Edge Browser is integrated with ChatGPT via Bing search engine, which was made publicly available in May 2023 as well. Microsoft also unleashed the AI powered Microsoft 365 Apps - which means the AI agent is available in Microsoft Word, Excel, Email and other tools in the su). Meta’s LLaMA is another one that started competing in the race of Generative Artificial Intelligence.?

The ChatGPT3 and 4 models were trained on billions of digital copies of books, articles, papers, blogs and so on. The ChatGPT-4 model was fed with the data up until September 2021 (meaning, it wouldn’t be able to retrieve any data after that date.?

Although there is no internet access for the GPT-4 to fetch real time information, as I write this section, OpenAI just released a beta version of Web Browser version for its Plus subscribers. So expect the internet-enabled generative AI assistant from Open AI soon to be available to the general public.

Similarly, Google released their AI agent to the world: Bard. Microsoft's ChatGPT-integrated-Bing search enabled Edge from Microsoft is another one of the players.

Play time

Before we move a bit further, let's play a bit with the Chat GPT-4 model on few simple cases.

We all know about Moore's law (if not, either play along or quickly browse the internet). I wanted to ask ChatGPT about it:

ME: 

What's Moore's law?

ChatGPT: 

Moore's Law is the observation made in 1965 by Gordon Moore, 
co-founder of Intel and Fairchild Semiconductor, that the number of 
transistors on an integrated circuit (microchip) would double 
approximately every two years. 


Moore's Law, though originally framed in terms of transistor count, 
has often been generalized in the technology industry to capture the 
idea that computers' speed, memory capacity, sensors, and even the 
number and size of pixels in digital cameras are improving at roughly 
exponential rates.

It's important to note that Moore's Law is not a physical or 
natural law, but rather an empirical observation and a projection 
of a trend that has held up for many decades. It has been kept alive 
by regular technological innovations in semiconductor fabrication. 
As of my last training cut-off in September 2021, there's been 
increasing discussion about the possible "end" of Moore's Law as 
physical limits on miniaturization are approached. However, new 
technologies, such as quantum computing, continue to push the 
boundaries of computational power and efficiency..        

That's an awesome answer, but may be I don't need three paragraphs of it. Let me ask my friend if he (!) can produce a summarised/condensed version:

Me: Can you condense it, please?

ChatGPT: Moore's Law, named after Intel co-founder Gordon Moore, 
predicts that the number of transistors on a microchip doubles 
roughly every two years, signifying exponential growth in computing 
power. While not a physical law, it's an empirical trend that has 
held true for decades. Despite approaching physical miniaturization 
limits as of 2021, innovations like quantum computing continue to 
push computational boundaries.        

That's an amazing summary. I want to take it a bit further:

Me: Ok, sounds good. But explain to me as if I'm a 10 year old kid

ChatGPT: You know how your video game console or computer gets 
faster and better every few years? That's kind of what Moore's Law 
is about. Gordon Moore, who helped start a big computer chip company, 
once said that the tiny parts that make computers work (called 
transistors) would fit twice as many on a chip about every two years. 
This means our gadgets can do more cool things, like run bigger games 
or show better videos, without getting any bigger themselves. 
It's like if you could fit twice as many legos on a baseplate every 
two years without the baseplate getting any bigger! But there's only 
so small these parts can get, so people are trying to find new ways 
to keep making our electronics better and faster.        

Not bad eh?!

I was explaining to my wife about this tool in the last few weeks. She wanted to test the ChatGPT with a question related to a J.M.W. Turner's classic: "Rain, Steam and Speed – The Great Western Railway". Here's the prompt and response:

This is almost an accurate analysis of what the Hare in front of a train symoolizes! Truth be told, I'd have struggled to come up this sort of answer!

The list goes on (I'll be producing a few more in the coming weeks).

Future of AI powered Search

One of the fundamental questions that search engineers were asked was how do the Generative AI agents change the course of search? Let’s answer this question by asking ChatGPT at how AI agents can compliment or help modern search or change their direction if they can. Let’s check the areas where tools like Generative AI will reshape the search space.?

Intuitive Search: Search queries will become more conversational and intuitive. Generative AI models, like GPT-4, have an advanced understanding of natural language, enabling them to interpret complex queries more effectively. Users will no longer need to rely on specific keywords or phrases; they can simply ask questions as they would in a conversation with another person. This will allow for more accurate and relevant search results, as the AI can better comprehend the context and intent of the query.

With the introduction and release of highly capable Generative AI agents and models, there is a greater scope to significantly reshape full-text search capabilities provided by modern search engines like Elasticsearch. We can expect to see several key changes that will redefine the search experience for both users and developers as this technology becomes increasingly integrated into search platforms.

Personalized Search: With the incorporation of Generative AI, search results can become more personalized and adaptive. The search engines will be able to learn a ton of valuable data points from users' preferences, behavior, and search history, which inturn helps the search engines tailor the results to meet individual user’s needs. As the AI gathers more data, it will continually refine its understanding of what users are looking for, leading to an increasingly customized search experience.

Predictive Search: Generative AI has the potential to make search engines more proactive in anticipating user needs. Instead of merely responding to queries, AI-driven search engines could predict what information users might be interested in, based on their previous interactions or current context. This would allow search platforms to proactively offer relevant suggestions, increasing the value of the search experience and reducing the need for users to perform additional queries.

Advanced Search: Generative AI will enable search engines to provide more diverse and rich search results. By understanding the context and semantics of a query, AI-driven search engines can generate content summaries, relevant visualizations, and even synthesize new information to help answer a user's question. This will lead to a more comprehensive and informative search experience that goes beyond merely linking to existing content.

Hence, in? my mind, the introduction of Generative AI will certainly revolutionise full-text search capabilities, making search engines more conversational, personalised, adaptive, and proactive. This will not only enhance the user experience but also offer new opportunities for businesses and developers to create innovative search applications and services. Teams are working hard to adopt the upcoming changes in the search space with the advent of artificial intelligence in the recent months. So, do expect a search space revolution in due course!

I'm starting a series of articles on these GPT Agents in the coming weeks - will dig deeper in explain the technology behind them along with putting it to work in various domains. Will work through the LLM modes, prompt engineering as well as plugins, including developing our own plugins.

Follow me on LinkedIn for upcoming articles.

Stay tuned!

Me @?Medium?||?LinkedIn?||?Twitter?||?GitHub