AI for Beginners

If you're just a regular, non-technical person and are curious about AI, read on. The sole purpose of this post is to get you to dip your feet in the water and familiarize yourself with it. Have some fun!

Artificial Intelligence (AI) is no longer the stuff of science fiction, it's now a part of our daily lives going forward. Many are fearful of it, perhaps not without reason. In future posts, I may look at the implications of AI and how it will affect society in general. For now, let's settle on satisfying our curiosity and not being afraid of something we have no control over. Like anything else, it can be used for good or bad.

What is AI? It's a branch of computer science that enables machines to perform tasks that would typically require human intelligence. These tasks include understanding natural language, recognizing patterns, solving problems, and making decisions. In short, AI allows computers to learn from data and improve their performance over time, much like how humans learn from experience. These models are effectively trained over time through a process known as machine learning.

It's pretty easy to get started. There are several AI engines available today, such as ChatGPT by OpenAI and Claude by Anthropic. These tools, known as Large Language Models (LLMs) are text-based language models, and allow you to interact with AI in a conversational way, making it a fun and accessible experience. They are browser-based tools – just sign up for a free subscription and start your adventure! My personal preference is ChatGPT. It's free, but I use it a lot and pay a $20 a month subscription which gives me free access to image models like DALL-E (used to create the images in this post).

Once you’ve signed up, you can start chatting with the AI chatbot. Begin with simple prompts like “Tell me a fun fact about the Irish” or “What’s a good recipe for dinner tonight?” Remember, AI is a tool to assist and augment your own thinking. It's not perfect and can make mistakes, so always use your judgment and verify important information. I can't stress this enough.

To get the most from it, you will find yourself refining your initial prompt a lot. In fact, this has given rise to a whole new field of study called Prompt Engineering. Let's keep this simple. The book "Prompt Engineering for Generative AI" suggests five generally accepted conventions to use as a basis for constructing your prompts:

(1) Give Direction: Describe the desired style in detail, or reference a relevant persona
(2) Specify Format: Define what rules to follow, and the required structure of the response
(3) Provide Examples: Insert a diverse set of test cases where the task was done correctly
(4) Evaluate Quality: Identify errors and rate responses, much like a feedback loop of continuous improvement
(5) Divide Labor: Split tasks into multiple steps, chained together for complex goals

The GIGO (Garbage In - Garbage Out) principle truly applies here. Over time, you can experiment and improve the quality of your prompts. AI is powerful, but as I pointed out earlier, it can make mistakes. However, by using these prompts and exploring different AI tools, you'll quickly become more comfortable with this exciting technology.