The real value is not the editor. It is the shortcut. KI-Anonymizer cleans up the text already sitting in your clipboard before it is pasted into AI chatbots, email, documents, or other tools.
AI chat workflows are exactly where sensitive information is still copied too quickly and checked too late. That is where the app fits: on-device, fast, no cloud, no account, no extra process. It protects the same moment in which copy and paste can otherwise turn into a privacy problem.
For Mac users who want to clean up sensitive text locally before AI use, review, sharing, or email.
Not later in the archive. Not only at the moment of sending. Right in the instant text gets copied into chatbots, email, tickets, or review threads. Names, addresses, phone numbers, and identifiers often move faster than anyone checks them.
Prompts are assembled fast, but sensitive details are often left untouched. KI-Anonymizer sits right before that step and reduces the risk of carrying personal data into the prompt.
The same workflow also makes sense for email, approvals, documentation, internal reviews, or sharing text externally. The product is positioned around AI, but it is useful in any copy-and-paste moment.
The editor is the visual control layer. The actual product core is the global shortcut. It turns an ordinary copy-and-paste routine into a safer intermediate step without forcing users into a new process.
You copy a sensitive passage from email, a document, the browser, a note, or a draft prompt for an AI chat.
With one global shortcut, the app anonymizes the clipboard locally on your Mac. No cloud, no account, no detour.
You paste the anonymized text where it is needed. If necessary, you verify the result in the editor before using it.
The app detects classic personal data in a structured, local way. Fixed patterns such as IBANs, dates, and email addresses are handled differently from names and places. That layered approach is exactly what makes the tool reliable.
Personal names, common name contexts, and consistent placeholder mapping across repeated mentions.
Street lines, house numbers, ZIP codes, and location references are grouped as address information.
Email addresses plus phone and fax numbers in common formats.
IBANs, case references, social insurance numbers, and similar formal identifiers.
Numeric and written date formats for local pre-anonymization.
Terms or fixed phrases can always be anonymized even when no standard pattern applies.
Specific terms can intentionally be kept out of anonymization when they should stay visible.
The app supports the workflow. Final responsibility still stays with the person reviewing the output.
KI-Anonymizer works fully without Apple Intelligence. When Apple's local system models are available on the device, the app can use them as a third detection layer: not for rigid patterns, but for additional person and place references that fixed rules or classic language detection may miss.
Ideal for email addresses, dates, phone numbers, IBANs, case references, and other clearly structured patterns.
Locally detects typical person and place entities through Apple NaturalLanguage.
Can add local person or place references that do not fit neatly into rigid patterns. It does not replace the other layers. It only fills likely gaps.
The screenshots stay intentionally compact. They show the real app rather than a marketing mockup: editor, settings, and the legal notice. Each view can be enlarged and explained in more detail.
One-time purchase. No subscription. Local text anonymization for sensitive content before AI use, email, review, or sharing.
These answers cover the questions that almost always come up around a local Mac app for anonymization: AI chatbots, local processing, and the additional value of Apple Intelligence.
No. The most common trigger is copy and paste into AI chatbots. The same local step also makes sense before email, approvals, documentation, internal reviews, or sharing text with third parties.
The anonymization step itself runs locally on the device. The app is designed to clean up sensitive content before it gets pasted into other systems in the first place.
If Apple Intelligence is available locally, it adds semantic extra hits. Hard patterns like IBANs or email addresses stay rule-based; Apple Intelligence only helps where people or places are described in a less rigid way.
By default, KI-Anonymizer follows a fast workflow: copy, anonymize, paste. Optional restore mapping takes that workflow one step further.
You can send anonymized text to an AI tool, copy the revised result back, and then restore the original references with a second shortcut. The mapping is optional, limited to the most recent session, stored in memory only, and automatically cleared after 5 minutes.
As an AI consultant, author, and instructor, I kept seeing the same mistake in practice: content was moved into chatbots, email, or other tools too quickly and without removing sensitive data first. KI-Anonymizer was built for exactly that problem: a small, fast tool at the copy-and-paste moment itself.
Yes. There is also a browser-based web app that can be used offline in the browser. It does not include the global shortcuts or the Apple Intelligence enhancement of the macOS version.
