I've been thinking a lot about fear lately—specifically, the fear that keeps you from hitting "accept" on an AI suggestion.
You're writing a post. You've got your raw thoughts down. Then you ask an AI to expand a paragraph, polish a phrase, or rewrite a section entirely. The suggestion comes back. It's good. Maybe even better. But what if the original was actually the voice you wanted? What if you lose something important by accepting?
So you reject it. You stick with what you had. You never know if you made the right call.
This is the problem I've been wrestling with at Jottings.
The Cost of Commitment
When you're writing by hand—pen and paper—revision is natural. You cross out. You rewrite. You live in the gray area of possibility. But with digital writing, especially when AI is involved, there's something paralyzing about the binary choice: accept or reject.
It feels like commitment. Finality. And finality breeds caution.
I realized this watching early Jottings users work with AI suggestions. They'd hover over the "Accept" button. Stare at the suggestion. Compare it to their original. Then click away without deciding, leaving the post in limbo.
It wasn't laziness. It was fear of loss.
What If You Could Have Both?
Then I thought: what if you didn't have to choose?
What if accepting an AI suggestion created a new version of your post, but kept the previous one safe? What if you could try something out, live with it for a moment, and then instantly jump back if it wasn't right?
Undo/redo, but not the mechanical "Ctrl+Z" kind. More like a revision history where you can see every iteration—the original, the AI suggestions you accepted, the paths you took and the ones you abandoned.
This is what I've been building.
How It Works
The idea is simple but powerful:
You accept an AI suggestion. Your original post gets archived as a version. The new version becomes your current post.
You can see the diff instantly. What changed? Highlighted, side-by-side. No confusion about what the AI actually did.
You can jump back anytime. Not "undo." Literally just switch back to a previous version with one click. The newer versions stick around too—you can flip forward again.
Your whole experiment is visible. You can see the journey from raw thought → AI polish → final post. Or you can see the five different AI suggestions you tried before picking the one that felt right.
The beautiful part? You never lose anything. Not a version, not a thought, not a path you considered and rejected.
The Real Benefit
This isn't really about having a safety net, though that's part of it.
It's about permission.
When you know you can try something, live with it, and instantly revert without losing anything, you become bolder. You say "yes" to AI suggestions you'd normally reject. You experiment. You compare the original to the polished version side-by-side and actually feel the difference instead of imagining it.
You become a better writer because you're not paralyzed by the fear of making the wrong choice.
I've been testing this in my own writing, and the difference is striking. I accepted an AI suggestion to tighten a paragraph. It was good—cleaner, more direct. But then I went back to the original for comparison. The original had this conversational warmth the tightened version lost. So I kept the original and rejected that suggestion.
But here's the thing: I actually knew what I was rejecting now. I wasn't guessing. I wasn't making a fearful choice in the abstract. I was making an informed choice based on what I actually wanted to say.
That's powerful.
Beyond Safety
The revision history isn't just a safety feature. It's a learning tool.
You can see exactly what the AI changed. Bold text showing additions, strikethrough showing deletions. Over time, you learn what kinds of suggestions work for you. Maybe you like when AI tightens prose but not when it removes specificity. Maybe you love AI-generated headlines but prefer to write your own opening paragraph.
The history becomes data about your own writing preferences. You become more intentional about when you use AI and when you don't.
The Anti-AI Angle
I know Jottings has always been a little skeptical of AI. We're not trying to automate your writing. We're not selling you on the idea that AI makes you a better writer.
But I think the truth is more nuanced.
AI suggestions are tools. Like any tool, they can hurt you if you're not careful. They can flatten your voice, remove the specific weirdness that makes your writing yours. But they can also help. They can suggest phrasing you wouldn't have thought of. They can prompt you to reconsider something you dismissed too quickly.
The key is keeping you in control. Keeping your voice, your choice, your judgment in the center of the decision.
Revision history does that. It makes AI a tool you experiment with rather than something you have to commit to. It puts you back in the driver's seat.
Trying It Out
We're still testing this. Not everything is in production yet. But if you're a Jottings user and you're interested in being part of the experiment, let me know. I'd love to hear what you think about the idea of writing with a safety net—not because you need it, but because it frees you to be braver.
The best writing happens when you're not afraid. When you can try something, feel it out, and change your mind without losing the path you came from.
That's what I'm building.
What do you think? Is revision history something that would change how you use AI with your writing? I'm genuinely curious. Feel free to send thoughts my way—I read everything.
Or if you want to explore Jottings and see what we're building, give it a shot. No signup required to look around. Come write without fear.