Feels like we’re hitting an interesting shift.
Last year, everything was:
→ “add AI”
→ “wrap GPT”
→ “ship faster than everyone else”
Now I’m seeing more indie builders do the opposite.
They’re not competing on AI capability anymore.
They’re competing on where AI fits in the workflow.
Because the problem wasn’t:
“users don’t have AI”
It was:
“AI shows up at the wrong moment”
Example patterns I’m noticing:
tools that remove steps entirely (instead of adding an AI step)
products that run in the background instead of chat interfaces
features that trigger automatically vs waiting for prompts
The winners aren’t the ones with “better prompts”
They’re the ones answering:
“where does this actually save time in a real workflow?”
Feels like we’re moving from:
AI as a feature → AI as invisible infrastructure
Curious if others are seeing the same shift
or still finding success shipping visible AI features?
The novelty phase is ending.
Users don’t care that AI exists, they care that something became faster, easier, or disappeared entirely.
Exactly. The novelty premium fades fast.
Users rarely buy “AI.” They buy saved time, reduced friction, better outcomes, or fewer steps.
That’s why invisible AI often wins. When the workflow improves, nobody asks what model powered it.
Well said. A lot of technology markets mature the same way, first people sell the ingredient, later they sell the outcome. AI feels like it’s entering that second phase now, where usefulness matters more than visibility.