Plenty of AI assistants promise to "respect your privacy." Few people stop to ask what that actually changes about how the tool is built. For LiveAI, it has very concrete consequences.
No proxy by default
The most defining decision: LiveAI does not route your messages through a central server by default. With your own API key, the app talks directly to the AI provider you chose. In practice, that means LiveAI technically cannot read the content of your conversations in this mode — not as a promise, but by construction.
The trade-off: fewer server-side "magic" features
This choice comes at a cost: no automatic cloud sync across devices, no centralized history browsable from a web page, no analysis of your habits to "personalize" the experience. Anything that requires a server to see your data is, on principle, off the table — unless you explicitly opt into the built-in LiveAI option, where the opposite trade-off is clearly spelled out upfront.
What still lives server-side, regardless
An account (name, email) is still required to sign in and, if you use it, to manage built-in LiveAI credits. This information is spelled out in full in our privacy policy — the goal isn't to claim an absolute privacy that's impossible to deliver, but to be precise about what's genuinely local and what isn't.
Why it matters
Most mainstream AI tools do the opposite: centralize everything by default, and offer privacy as a paid add-on or a line in the legal fine print. For a tool that potentially reads your files and your most personal conversations, we think that order of priorities should be reversed.