How My Experience with Prompt Engineering Has Evolved from Google in the Early 2000s to ChatGPT Today
How has the use of prompt engineering changed from the early days of Google to today's language models like ChatGPT?
My conversation with ChatGPT about inputs and outputs.
What is prompt engineering?
ChatGPT: Response to what is prompt engineering
Input initial information
In order to "train" or engineer the result I have to provide some information or "context" to the bot. This could have been an earlier blog post of mine which would have more information as well as my writing style which the bot could also use in crafting the response later. This was a short and quick input.
My initial “training” information
ChatGPT responded with a "yes" and then I went to prompt it with the following.
My request to write a blog post
So, here is the blog post which I got back. Would I use this as my blog? No. It is way too dry and sterile. Perhaps if I trained it with more of my writing then I might get a better result. But then, this is Google in 2002 - it will only get better!
👇🏻👇🏻 Enjoy! 👇🏻👇🏻
How My Experience with Prompt Engineering Has Evolved from Google in the Early 2000s to ChatGPT Today
As someone who started learning how to better use Google in 2002, I have seen firsthand how the process of finding information has changed over the years. In the early days of Google, it was crucial to "engineer" my search query in order to get the best results. But as technology has advanced, the need for this complicated engineering has diminished. In this blog post, I will share my journey and how it relates to the use of prompt engineering in today's language models like ChatGPT.
Early Days of Google: The Importance of Query Engineering
When I first started using Google, I quickly learned that it was important to carefully craft my search query in order to get the most relevant results. This led me to adopt the title of "PhD. of Googleology" in my email signature for a few years. I spent a lot of time figuring out the best ways to word my queries and combining them with the right keywords in order to get the results I wanted.
My business email signature in 2002
How ChatGPT Works with Prompt Engineering Today
Today, I use language models like ChatGPT and have found that the process of finding information has become much more intuitive and accurate. ChatGPT works by predicting the probability distribution of the next word given the previous words in the input, also known as the prompt. By modifying the prompt, the user can control the content and style of the generated text. And just like with Google search, the more information you provide, the better the results you can get from ChatGPT.
The Future of Searching for Information
"The way we search for information will likely be forever changed with the advancement of technology." With language models like ChatGPT and the continued development of advanced techniques, the process of finding information is becoming more sophisticated and intuitive. It's exciting to think about how this technology will continue to evolve and improve in the coming years.
My experience with prompt engineering has come full circle from the early days of Google to today's advanced language models like ChatGPT. While the need for complicated query engineering has diminished, prompt engineering is still a useful tool for controlling the content and style of the generated text and steering the model towards a specific goal or task. And with the continued advancement of technology, the future of searching for information looks bright and full of endless possibilities.
