Apple Intelligence for Writing: What It Does and Where It Stops
There is a version of this post that starts with me trashing Apple's writing features to sell you mine, and I'm not going to write that version, because it would be a lie. Apple Intelligence on the newer Macs is genuinely good at parts of writing, it runs on-device, it's free, and for a lot of people it's the first time they have seen a computer finish their sentences without sending them anywhere. That is a real thing, and I respect it.
But it stops in a specific place, and the place it stops is the reason tools like mine exist. I want to walk through what Apple Intelligence actually does for writing, where its edges are, and how the two things fit together, honestly.
What Apple Intelligence actually does for writing
On macOS 26 and the Macs that support it, Apple Intelligence brings a real on-device writing engine. It can rewrite what you wrote, adjust the tone, summarize, and offer inline completions in the apps where it's wired in. The key facts are worth stating plainly. It runs on your Mac, so your text doesn't go to a server. It is free, bundled with the OS. And it's getting better, because Apple is putting serious engineering into it.
Those are real capabilities, and if you're on a supported Mac, you should use them, because free and on-device is a hard combination to argue with. I'm not here to tell you the free thing Apple built is bad. I'm here to tell you where it stops, because that's the honest gap.
Where it stops
Here is the edge, and it's the same edge that shows up with Apple's built-in predictions, which I wrote about in an earlier post. Apple's writing features aren't a layer under everything you type. They are wired into Apple's apps and into certain surfaces, and they don't reliably appear in every third-party app, every browser field, every chat box, every editor. You can be writing in one app with Apple's suggestions working nicely, switch to another, and find the feature simply isn't there.
The result is the same patchiness problem. A writing habit can't form around a feature that's there in some apps and missing in others, because your brain needs consistency to build a reflex. I wrote about how the reflex forms and the whole argument is about consistency. Apple Intelligence is excellent in its lane, and its lane isn't everywhere you type.
- Rewrite, tone, summary, completions
- On-device, free, bundled
- Patchy outside Apple's apps
- No consistent keypress habit system-wide
- Ghost text in Mail, Slack, browsers, editors
- One key, Tab, in every app
- Costs a one-time price
- Not Apple's polish in Apple's own apps
The part that surprised me
Here is the thing I didn't expect when Apple shipped its writing engine, and it's the most interesting part of this whole story. Apple Intelligence didn't hurt tools like mine. It helped them. Millions of people suddenly had a Mac that could finish sentences on-device, and that normalized the whole idea of local writing assistance. People stopped asking "why would I want my computer to finish my sentences" and started asking "why does this only work in some apps."
That second question is exactly the one a system-wide tool answers. The category got bigger because Apple legitimized it, and then people hit Apple's walls and went looking for something that worked everywhere. I've had people tell me they tried Apple's writing tools first, liked the idea, got frustrated that it wouldn't follow them into their browser or their Slack, and came to WriteAmp for the coverage. That isn't a rivalry. That is a funnel, and I'm grateful for it.
There's a real overlap too. On macOS 26, WriteAmp can itself route its suggestions through Apple's on-device model instead of its own local one. So the honest framing isn't "Apple or WriteAmp," it's "Apple's engine, Apple's apps, or Apple's engine everywhere," and I wrote about how the engines work if you want that detail.
How to decide
Here is the practical question, and it's the same one I keep coming back to. Where do you actually type?
The privacy point is worth making plainly, because there's a lot of confusion about it. Apple Intelligence runs on-device, and so does WriteAmp's own engine, and when WriteAmp routes through Apple's model, that's on-device too. Neither one is sending your writing to a server. I published the full network ledger so you can check that claim, and it's the same answer whether the engine under the hood is mine or Apple's: the words stay home.
The honest close
Apple Intelligence for writing is a real step forward, and I'm not going to pretend otherwise. It is on-device, it's free, and it got a lot of people comfortable with the idea that their Mac can help them write without shipping their words anywhere. What it's not, yet, is a system-wide habit. It stops at the edge of Apple's apps, and that edge is where a tool that lives under every text field earns its keep.
If Apple's writing tools cover you, use them, happily and free. If you've hit the wall where the help vanishes outside Apple's apps, the trial is free for 30 days, and it's the same on-device idea, applied everywhere you type. The Apple's-built-in post and the engine post have more if you want the full picture. Apple built a good thing. The honest gap is coverage, and coverage is a solvable problem.
Sources
- Apple Intelligence: Apple's writing features and on-device processing
- Apple Foundation Models: the on-device engine developers can build on
On macOS 26+ Macs, Apple Intelligence mode stays free even after the trial ends — you always keep a working path to suggestions.
Written by Amit Ashwini, who builds WriteAmp and runs its marketing. More: why the Tab key beats the chat box · mini, midi, and max compared · benchmark methodology.