Lumo
Proton's hosted AI assistant with zero-access encryption of saved chats and a strict no-logs policy.
Private alternatives to ChatGPT, Google Gemini, Microsoft Copilot, Perplexity, vetted against our public criteria.
Grouped by threat level
Proton's hosted AI assistant with zero-access encryption of saved chats and a strict no-logs policy.
DuckDuckGo's private AI chat that lets you talk to multiple models with no account and no training on your prompts.
Infomaniak's Swiss-hosted AI assistant that pledges no training on user data and operates under GDPR and Swiss data law.
Apple's AI. On-device where it can, with verifiable Private Cloud Compute for the rest.
Ollama lets you download and run large language models locally via a simple CLI and REST API. Supports a growing library of open models including Llama, Mistral, and Gemma on Windows, Mac, and Linux with no data sent to the cloud.
A Mozilla project that packages a complete LLM and its runtime into a single executable file. Download one file, run it on Windows, Mac, or Linux with no installation, no dependencies, and no network connection required.
Offline AI chat from Ente that runs models entirely on your device, with no network calls required.
KoboldCpp is a self-contained local AI inference tool with a built-in web UI. It runs GGUF text, image, and speech models with no installation beyond a single binary, primarily aimed at creative writing and roleplay workflows.
No matches for those filters.
Mainstream AI assistants log your prompts, can train on what you type, and tie every conversation to your account. There are two private ways out, and both are below. Hosted assistants still send your prompt to a server, but to providers that contractually do not train on your chats and keep little or nothing. Local tools run an open model directly on your own machine, so the text never leaves your computer and works with no account and no internet. The picks run from the easiest swap to the most private, so you can choose how far you want to go.
A hosted assistant is the one-click replacement. Your prompt still travels to a server, so you are trusting a provider’s no-training pledge and its retention settings, but a good one collects almost nothing and operates under privacy-friendly law. A local model is the stronger posture: inference happens on your device, there is no server to log, leak, or train on what you type, and it keeps working with the network off. The trade is convenience and raw capability for control. Both map onto our threat levels: the hosted picks sit at Covered, the local runners at Hardened.
For a hosted assistant, the provider’s commitments are what protect you: a clear no-training pledge, minimal logging, no advertising profile built from your questions, and ideally an open-source client and a strong privacy jurisdiction. For a local tool, look for one that runs open-weight models offline, is itself open source, and does not phone home. Hardware matters for the local route, since bigger models want more RAM and a capable GPU, so start with a smaller model and scale up to what your machine handles comfortably.
Point your everyday questions at one of the hosted picks and use it the way you used the mainstream tool. For anything sensitive, drop down to a local model where the text never leaves your computer. The honest catch is that the largest mainstream models still lead on the hardest problems, so you may keep one mainstream account for those, and simply stop feeding it the rest of your life.