# Ollama urges security teams to run open models locally or privately

> Original title: Defensive work shouldn’t mean sending sensitive data somewhere you don’t control. You can run open models locally with Ollama, or on Olla...

- Source: Ollama
- Published: 2026-10-23T21:07:25.000Z
- Bad Influence: https://badinfluence.news/items/fhwptjho6od6oplxor6ziq2f5
- Original: https://x.com/ollama/status/2113739093398241739

## Summary

Ollama says defenders can run open models on their own machines or on Ollama's cloud with zero data retention, keeping sensitive data under their control. Its co-founder Michael Chiang argues that defensive work such as exploit analysis and red teaming needs models that closed providers often restrict to special access programs. He adds that attackers already use open models, so defenders should have them too.

## Post

Defensive work shouldn’t mean sending sensitive data somewhere you don’t control. You can run open models locally with Ollama, or on Ollama’s cloud with zero data retention.

## Quoted post @mchiang0610

> More security teams are coming to @ollama to run open models for defensive work.
> 
> Defending a system means understanding how attackers get in, which means using models that will help with things like exploit analysis and red teaming. Most closed model providers only allow that through special access programs. Open models can be used by any team, on their own machines or on servers they control.
> 
> Open models can be misused and teams should run them with care. But attackers already have them, so defenders should too.

https://x.com/mchiang0610/status/2113738632083427568