Imagine a kid picking up a guitar in 2010.

The path ahead was fairly predictable.

You spent years fumbling through chord charts in your bedroom. You built calluses. You learned songs badly, then slightly less badly. Eventually, you became good enough to play a local gig, film a YouTube tutorial, teach someone else, or perhaps even make some money from the skill.

For many hobbies, the path from curiosity to opportunity looked something like this:

Interest → Practice → Skill → Credibility → Public Proof → Opportunity

Then AI changed what counts as public proof.

Today, you don't need ten years of guitar calluses to publish a polished guide to music theory. AI can help script the lesson, generate backing tracks, design the thumbnail and write the caption in minutes.

The ability to produce content that looks like expertise is no longer scarce.

That matters.

For years, publishing useful tutorials, thoughtful essays or polished educational videos served two purposes. The content itself was useful, but producing it also signalled something about the person behind it:

This person probably knows what they're talking about.

That signal is getting weaker.

AI doesn't make real expertise worthless.

It makes the appearance of expertise cheap.

And when polished output becomes easier to manufacture, we need better ways to prove that a real human actually earned what they know.