Jean Bonnenfant

<start of post>

The "AI is taking our jobs" crowd is looking at the wrong chart.

Most people think we're at the peak of the wave. That the market is flooded. That if you haven't built your "AI wrapper" yet, you've missed the boat.

They're wrong.

92% of mid-market firms have "AI" in their 2026 strategy

Less than 4% have a single agent running in production

Most "implementations" are just ChatGPT tabs left open in Chrome

We aren't in a bubble of over-adoption. We're in a bubble of over-talking.

For context: at Tinct, we're seeing this first-hand. When we talk to design partners, they don't want "more AI." They want the 10 hours back they spend manually mapping LinkedIn profiles to landing page copy.

They don't care if the engine is Claude, GPT, or a room full of magic squirrels. They care about the output.

One partner told me: "I don't want a co-pilot. I want an autopilot for the boring stuff."

That's the shift.

We're moving from "AI as a feature" to "AI as the infrastructure." The companies that win won't be the ones with the best prompts. They'll be the ones who hide the tech so well that the user just thinks the product is "fast."

Building this isn't easy. We've broken the onboarding flow three times this week trying to simplify the 'magic' moment. (shoutout to Fabien for fixing it at 2 AM)

But that's the price of being early.

If you're waiting for the "perfect time" to start building, you're waiting for the grey dots to turn red. By then, the margin is gone.

The best time to build was last year. The second best time is now.

PS: we just opened 50 more spots for the beta. Link in the first comment.

<end of post>