it:ai
Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revision | |||
| it:ai [2026/10/22 14:07] – [Potential security risks of running local AI models] gary1 | it:ai [2026/10/22 14:07] (current) – [Potential security risks of running local AI models] gary1 | ||
|---|---|---|---|
| Line 176: | Line 176: | ||
| ***running a local AI is generally preferable to running a cloud AI model as your prompt or data is not sent to a cloud AI for its training data or worse, for someone to see your private data - you don't want patient data sent to the web unprotected** | ***running a local AI is generally preferable to running a cloud AI model as your prompt or data is not sent to a cloud AI for its training data or worse, for someone to see your private data - you don't want patient data sent to the web unprotected** | ||
| *if you have the latest software and you download models from verified platforms and you don't add in agents or give it file system access etc then you are reasonably safe from issues | *if you have the latest software and you download models from verified platforms and you don't add in agents or give it file system access etc then you are reasonably safe from issues | ||
| - | *local AI LLM models have several layers and each layer poses different risks which your virus protection systems may not be able to protect you from(([[https:// | + | |
| * **GGUF LLM model file** | * **GGUF LLM model file** | ||
| * this is the model file you download and it mainly just has safe tensor numbers which will tell the neural network how to work | * this is the model file you download and it mainly just has safe tensor numbers which will tell the neural network how to work | ||
it/ai.txt · Last modified: 2026/10/22 14:07 by gary1