Table Of Contents
AI Basics Everyone's Talking About
Setting the scene
Hi there!
Today we'll explore: AI Basics Everyone's Talking About (in Plain English)
AI is everywhere. Powering chatbots, generating images, even in your workflow tools. But the jargon can feel overwhelming. Terms like tokens, RESTful APIs, and agentic AI are key to understanding what's happening right now. While you're perhaps stuck in back-to-back meetings today, here's a quick guide that could come in handy:
What's what?
Machine Learning (ML)
This is the foundation and science behind AI. These are the systems that learn from data rather than being explicitly programmed. Models improve as they process more examples, and these can power everything from spam filters to recommendation engines and genAI.
Generative AI (genAI)
GenAI uses ML to create text, images, code, and content. It does not just analyze data, it drafts copy, designs visuals, automates workflows. Think ChatGPT.
Large Language Models (LLMs)
LLMs use massive generative AI models trained on huge datasets. They understand and produce human-like text, powering tools like ChatGPT's GPT-4 model or Claude. Their strength lies in their ability to (at least seemingly) converse, summarise, and ‘reason'. They're excellent at predicting patterns, but their limitation is that they don't truly ‘know' facts.
Agentic AI
This is the shift from AI as a tool to AI as a teammate. Where ‘old AI' follows the “summarize this -> done” approach, agentic AI searches, compares, summarises, and even emails you. All autonomously.
Tokens
These are the building blocks of AI language. AI models process your text search into small chunks of words. For example, your input of ‘why is artificial intelligence cool?' is broken down into [why] [is] [artificial] [intelli] [gence] [cool]. These tokens impact cost, speed, and context length and vary by AI model in terms of input and output results.
Restful APIs
APIs are how apps talk to AI. A RESTful API lets you send a request (e.g. 'Summarize this') and get structured data back. It's the glue for embedding AI into products and connecting APIs to apps.
Now why does this all matter? Tokens manage costs. ML and LLMs explain how AI learns and scales. Generative AI shows what it creates. APIs connect it all. And agentic AI points to a future of autonomous digital workers.
Outro
That's a wrap for today. Every ~10 days I'll drop something useful in your inbox (think sharp how-to guides, a peek under the hood of tools we've built, or quick takes on key topics). Want more? Explore other articles from our team or swing by our site to see what we're building.
Scale wisely,
Laurent
