Are AI coding tools creating more builders, or just more prototypes?

Something I've been thinking about recently:

AI coding tools have dramatically lowered the barrier to building software. A person who couldn't code a year ago can now create a working MVP in a weekend.

But I wonder if we are entering a new bottleneck.

Building the first version is becoming easier, but:

• How many of these projects actually reach real users?

• How many founders know what to improve after launching?

• Is the difficult part still coding, or is it product thinking, distribution, and understanding users?

I've been exploring different AI coding workflows recently, and it feels like the future is less about "AI writes code for you" and more about "AI helps you build, test, and iterate faster."

Curious to hear from people who have actually shipped something with AI coding tools:

What was the biggest challenge after your first MVP was ready?