Bro, if you're tryna make money off AI in 2026, don't just focus on selling prompts.

The Market / 集市
1

2026 and people are still pushing “selling prompts” as the main AI monetization strategy. I think it’s time to rethink. Prompts cost almost nothing to copy—sell one, and the buyer can tweak it and pass it along. The moat is way too shallow. What actually makes sustainable money is embedding AI into a concrete deliverable: batch-generating product images that match a brand’s style for e-commerce, producing reusable art asset packs for game teams, or setting up a stable local workflow for small studios and charging maintenance fees.

You’re selling results and reliability, not a string of text.

Prompts are the foundation of your craft, but don’t treat the foundation as the product itself. Figure out what problem you can solve for whom—that’s way more practical than obsessing over how to price your prompts higher.

2

Copying a prompt costs basically nothing. Once you sell it, someone just tweaks it and resells it.

3

Yeah, the idea of embedding your skills into the actual delivery is spot on—sell the result, not just words.

4

Making a batch of images that match the vibe for e-commerce clients—now that’s the kind of sticky work that keeps 'em coming back.

5

A lot of people overlook charging a maintenance fee—being able to consistently deliver is where the real value is.

6

“Too shallow a moat” — that hits the nail on the head. Anyone can copy a prompt.

7

I know some people who make game asset packs. If the assets are reusable, you can definitely keep making money off them long-term.

8

Figuring out whose problems you’re actually solving is way more real than trying to sell prompts at a premium.

9

You gotta separate the skill from the product itself—lots of people just lump them together.