Mozilla just wired Mistral’s open AI models into Firefox’s Smart Window, and the pitch is different from every other AI browser: your conversations aren’t stored, and there’s a zero-data-retention promise behind it. Here’s what it does, how the privacy claim actually works, and how it compares to the cloud AI browsers and truly local setups.
For two years the AI story was ‘bigger is better.’ That’s quietly changing. A new wave of small models is fast, cheap, private, and good enough for real work — and you can run them on your own laptop or phone.
RAG is the trick that lets an AI look things up before it answers — pulling in real documents instead of guessing from memory. Here’s what it is, how it works, and why it’s the difference between a chatbot that makes things up and one you can trust with your own files.
You don’t have to send your documents to a cloud to get AI’s help with them. We installed Ollama on an ordinary Windows PC, pointed it at our own files, and got a private, offline, no-API-key assistant. Here’s the honest story – what it’s great at, and where a normal computer hits its limits.
