← All episodes
What It Really Takes to Run AI Locally — and the Security You Have to Manage
Running AI locally means running a whole stack. Earlier this year, one crafted model file made Ollama leak API keys before anyone typed a prompt. So is Ollama safe, and what else around the model needs securing? This episode maps everything around a local model and marks where attacks have landed.
Coming Tuesday 27 October.
Why a model file is input to a program. What rides along with the weights: code, chat templates, conversions and LoRA adapters. The runner defaults and parser bugs a CVE scanner misses. The unauthenticated Ollama API, exposed twice in two years. And three checks to run on your own setup.
For a running example of a local model wired into real work, see the Mini SOC.