Google and AI are completely different, and understanding how they work changes how you use them. *To be specific: by “AI” here I mean Large Language Models, or LLMs. Google has technically always been AI – its PageRank algorithm, which decides which links to return to you, has always... been a form of AI. And yes, Google is now augmented with LLMs. What we’re comparing in this post is traditional Google (using PageRank) and LLMs.* Lets break down the differences in plain English & answer some of the most commonly asked questions below (so you know when to use which tool): Q: What’s the biggest difference between Google and LLMs like ChatGPT? Google searches. LLMs predict. Google scans the internet and returns the most relevant websites using a ranking system. LLMs don’t search. They learn from massive amounts of text — and then predict what word comes next based on probability. Q: So are LLMs just looking things up? No. They’re word generators. They learn patterns in language, and then generate text that follows those patterns. That’s why they can sound confident — even when they’re wrong. Many - crucially, not all!!! - models today have been augmented with the ability to pull some data from the internet and insert it into its context window (i.e. the input into the LLM, the prediction machine) which reduces the risk of “hallucination,” (i.e. making things up) but even so, the model is still a predictor. It can (and often does) still mess up EVEN with web search added. Q: What’s the biggest tradeoff to watch out for? Hallucinations. LLMs are generators (& not a fact checkers). So they can make things up. That’s why tools like web-browsing and retrieval-augmented generation (RAG) are so important. They anchor generation to real data. But they’re not foolproof. Bottom line: If Google is a librarian, LLMs are improv actors who’ve read the entire library. Sometimes brilliant. Sometimes also wrong. But when you understand *how* they think — you can use them way more effectively. Leave your questions or comments in the section below 🥰👇 & share to spread the word!! This concept is simple and yet so important. You guys rock, sending love to you all!show more

Useful distinction. Search retrieves existing documents, while an LLM generates a fresh answer from patterns, so it can be fluent and wrong with no source to check. Treating it as a drafting partner rather than a lookup tool is the habit that matters.

So true. I pay for Grok, Claude, ChatGPT and Gemini, and run my own Hermes agents on a personal server. Different jobs need different AI, and knowing which one is improvising is half the skill.

Also, if you're new here, hi! I'm Harper. I've spent a decade building AI as a Stanford computer scientist and coding AI systems at Meta & at a startup acquired by NVIDIA. I've always loved to teach, and now I get to create educational videos to share the skills, understanding, and intuition I've worked years to develop with you. For my (free) 10 Days of AI Basics series + more:

Btw I mean Google Search and not the whole company! Google has lots of AI products, and of course now Google Search has the AI summary at the top, which uses LLMs + sources. What we’re getting at here is how agentic AI = LLMs & the harness around them, which enables tool use (for example, web search) in addition to token generation.

Google gives you sources to evaluate; LLMs give you answers to scrutinize. Knowing when you need evidence versus a useful starting point makes all the difference.

@grok is PageRank an "AI algorithm"?

The real skill is knowing when to ask AI and when to verify the answer with actual sources.

This explainer lands at a weird moment. Google itself now runs AI Overviews and leans on LLMs, and a whole industry has sprung up just to help brands rank inside ChatGPT and Gemini answers. Meanwhile Gen Z already trusts ChatGPT over Google for product research, per a Forbes survey in late September. So the line you're drawing between search and prediction is real, but the two tools are blurring fast. Knowing which one you're talking to matters more than ever.

Btw I mean Google Search and not the whole company! Google has lots of AI products, and of course now Google Search has the AI summary at the top, which uses LLMs + sources. What we’re getting at here is how agentic AI = LLMs & the harness around them, which enables tool use (for example, web search) in addition to token generation.

The search versus generation split is the part most people never learn. Once you know a model is predicting text, you stop trusting its confident tone.

A simple way to use both: use search when you need a fact, a date, or a price, because it points to pages that already exist. Use a chat tool when you need a first draft or a plain explanation, then check any numbers or names it gives you against a source you trust.

The librarian vs improv actor line is a great way to put it. Improv is brilliant for a first draft. For the answers people actually act on, what's your favorite habit for bringing the librarian back in?

Wait. I thought PageRank was a deterministic algorithm that always featured the same links for anybody for the same query and at a given point in time. Isn't that the case? (not talking about personalized Google search.)

Still not convinced this isn’t an AI person. Which is a compliment. 🤣