In this video, I show you how to run a ChatGPT-like AI chatbot completely in your local environment using TrueNAS. Public AI tools like ChatGPT or Google Gemini are useful, but many people forget that daily prompts can include personal or business data. So my goal here is simple: build your own AI chatbot on your own server, so your data stays under your control. I install Open WebUI from the TrueNAS Apps section and configure the required datasets with proper app permissions. During installation, I select the built-in Ollama image (and I also explain that if you have an NVIDIA GPU, you can enable CUDA support). After the app is up, I create the first-time admin account in Open WebUI, then pull an LLM model directly from the Ollama library—like DeepSeek-R1 (in my demo, I use the 5B parameter model). Once the model downloads, you can start chatting locally. I also cover how you can add multiple models, control model visibility, and create multiple users with permissions, so you can run a secure AI assistant inside your own environment.