🤖 AI 4 Beginners
100% free No code No homework

Don't just read about AI. Poke it. Play with it. Beat it.

The home of @ai.4.beginners — AI news explained simply, plus tiny experiments you can poke: guess against a language model, crank a chaos dial, shred a sentence into tokens. You'll understand AI because you touched it.

it takes 20 seconds ↓ Start playing
🎮 Game No. 1
Beat the model

A language model writes by predicting the next word. Think you can predict it too? Five rounds.

winner gets bragging rights →

Which word does the model pick next?

You 0 · Model matched 0 of 0
🌶️ Experiment No. 2
The chaos dial

“Temperature” controls how adventurous an AI gets. Low = predictable. High = unhinged poetry. The bars are computed live — this is the real math.

drag it all the way, we dare you →

“Describe a cat.” → 

0.0 · Librarian2.0 · Jazz poet

temperature = 0.2

✂️ Toy No. 3
The word shredder

Models don't read words — they read tokens, chunks about ¾ of a word long. Type anything. Watch it get shredded.

try your name! →

0 tokens ·

The blog

AI news, explained like you're human.

One author, no guest posts — the same stories as @ai.4.beginners on Instagram, with a little more room to breathe.

The word collection

Eight words that unlock every AI article.

large language modelnoun /ˌel ˌel ˈem/ · abbr. LLM

A system trained on vast amounts of text to predict the next word. The thing you just beat (or didn't) in Game No. 1.

promptnoun /prämpt/

The request you give a model. The single biggest lever on answer quality you control — master it and every AI tool gets better.

tokennoun /ˈtō-kən/

A chunk of text, roughly three-quarters of a word — the unit a model actually reads and writes in. You shredded some in Toy No. 3.

temperaturenoun /ˈtem-pər-ˌchu̇r/

The dial that controls how adventurous a model's word choices are. You turned it in Experiment No. 2. Handle with care above 1.5.

hallucinationnoun /hə-ˌlü-sə-ˈnā-shən/

A confident, fluent, entirely made-up answer. Not a lie — a plausible guess presented as fact. See: the blog post about fake sources.

training datanoun /ˈtrā-niŋ ˈdā-tə/

The text, images, and examples a model learned from. It shapes — and limits — everything the model can do.

neural networknoun /ˈnu̇r-əl ˈnet-ˌwərk/

Layers of simple math, loosely inspired by neurons, that turn examples into ability. The machinery under every modern model.

agentnoun /ˈā-jənt/

An AI that doesn't just answer but acts — browsing, writing files, using tools in sequence to finish a task.

Fair questions

Asked before, answered plainly.

Who writes this? +
Aditya Dutta — a 12th-grade student passionate about AI and tech, and the same person behind @ai.4.beginners on Instagram. Every blog post here and every reel there comes from the same keyboard. No content farm, no ghostwriters.
Can anyone else post on the blog? +
No. The blog has exactly one author, and there's no submit button to find — posts are published directly with the site. What you read here is always first-hand.
Do I need to know how to code? +
No. Nothing on this site involves code. If you can drag a slider and type a sentence, you meet the prerequisites, because there are none.
Is it actually free? +
Yes — the blog, the games, and the word collection. The experiments run right on this page in your browser. Nothing to unlock because nothing is locked.
Can I suggest a topic? +
Absolutely — send a DM to @ai.4.beginners. If a question makes three people go “wait, really?”, it probably becomes a post.
New posts whenever the news is worth it

You've already played with an AI today. Now follow along.

psst — the button does a thing 🎉