Palivane - See and control every prompt, file, and message sent to AI

AI created two risks older tools miss: people pasting secrets, PII and code into AI apps, and attackers injecting your models to extract context. Palivane inspects both directions everywhere AI touches your org: a browser extension, an OpenAI/Anthropic/Gemini compatible gateway, connectors for Slack, M365 and Workspace, plus CLI and CI hooks. Detection runs cheapest first. Regex rules, then two on-box ML classifiers, then an optional LLM judge that runs under your own cloud contract.

Hey Product Hunt, David here. I do detection engineering and SIEM work for a living, so I end up in a lot of conversations that start with "someone pasted the wrong thing into ChatGPT." Every tool I evaluated to fix that wanted to route prompts through its own cloud first. That solves the leak by creating a second copy of your data at a company nobody signed up for. Palivane does the inspection in the browser instead. Prompt text stays on the machine. What leaves is the verdict: which site, what time, which policy matched. No new subprocessor to add to your DPA, no second custodian. It's early and the rules are opinionated. If you run security or IT somewhere, I'd like to hear what your detection set is missing and what would have to be true before you'd push this to a fleet. Happy to hear where it falls down.

Palivane - See and control every prompt, file, and message sent to AI

Replies