
The free, open-source RemoveMacAI is out for Apple-silicon Macs on macOS 27. It switches off Apple Intelligence, deletes the downloaded AI models, and blocks their re-download.
Summaries like this, in your inbox every morning.
RemoveMacAI is published on GitHub under an MIT license, and its developer says it makes no external network connections and collects no user data. Beyond Apple Intelligence, it can also turn off Apple's usage analytics, personalized ads, Spotlight internet search suggestions, click-to-hide wallpaper behavior, Dock animations, and automatic text correction.
It includes storage-cleaning features as well, letting users review old macOS installers, downloaded video wallpapers, iPhone and iPad update files, unneeded Xcode data, and app caches for deletion. Background updaters registered by third-party apps such as Google's can also be disabled individually.
One practical caveat: because macOS itself performs the deletion of the model files, the freed disk space may not show up in storage figures right away after the models are removed. The changes can be made either from a normal Mac app or from the command line in Terminal, can be reviewed before being applied, and can be undone. The tool does not require disabling macOS System Integrity Protection (SIP) and is designed not to modify protected system areas.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
MangoBoost delivered open AI models powered by AMD, per the announcement "MangoBoost Delivers Open AI Models P…

A developer released Trading Dojo, an open-source MCP system that hands AI agents like Claude Desktop and Anti…

PC vendors are expanding memory capacity, and software platforms are making larger open-weight models easier t…

OpenAI put a broad range of new mathematical results from an internal frontier model on GitHub — covering 90 o…

Opper published jevman, an open-source Pac-Man benchmark that runs six AI models — among them TypeSafe AI's Je…

On a self-built 76-question Jev-format set, six trained 3B–9B open-weight models (Imajev-4B, Clef-Flash 9B, Je…
