Media & education, without the hype

Learn to read the internet before it reads you.

MG teaches social media literacy, how AI quietly shapes your feed, and how outlets use AI and prediction markets to build the narratives you scroll past — plus the practical skills to build something of your own.

"A rising tide raises all boats."

What we teach

Four things worth understanding right now.

Each is a lens on the same problem: information moves faster than judgment does. We slow it back down.

01

Social media literacy

Why your feed shows you what it shows you — the incentives, the ranking signals, and how to read a post the way its algorithm does.

Foundations
02

AI in your daily life

Where AI is already making decisions on your behalf — your search results, your applications, your recommendations — and how to notice it.

Awareness
03

Narratives, AI & prediction markets

How newsrooms and platforms use AI-driven trends and prediction markets to decide what's "the story" — and how that becomes your reality.

Systems
04

Build something real

Starting a business, writing your first real code, and using AI as a tool rather than a crutch — practical skills, not vibes.

Practice

How it works

No gurus. No hype. Just a method.

01

We show our sources

Every claim links to the primary source — the paper, the filing, the dataset. You can check our work, and we want you to.

02

We name the incentive

Before explaining what a platform or outlet did, we explain why it was profitable for them to do it that way.

03

We end with a skill

Understanding a problem is step one. Every piece ends with something concrete you can go do differently.

Patrick, founder of MG
Patrick — founder, MG

Who's behind this

I'm Patrick. I built MG because I got tired of watching smart people get played by feeds they didn't understand.

I spent years working with the systems this site explains from the inside — the ranking models, the trend cycles, the way a newsroom decides what counts as news on a given morning.None of that is secret. Most of it just isn't explained plainly.

MG exists to close that gap: sourced, specific, and built so a skeptical reader can check every claim themselves. If something here reads like hype, that's a bug — tell us and we'll fix it.

Sourced
every claim links to a primary source
No sponsors
we don't take money to cover a topic
Public record
corrections logged, not memory-holed