# Why don’t we have national open AI models? — retained record

> Retained reddit record from reddit:LocalLLaMA.

Canonical: https://highsignal.app/data/records/e69fdc8ecddc99ee

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:LocalLLaMA)  published 2026-11-14 15:17 ingested 2026-11-15 02:36

[original source ↗](https://www.reddit.com/r/LocalLLaMA/comments/1xw4ue3/why_dont_we_have_national_open_ai_models/)

## retained text



```
Deepseek spent about $6m on 4.1 apparently. Published all of their work.

US gov says don’t trust China, blah blah…

As a US citizen why are we not training national open source?

The cost is nothing compared to the budget, solves security issues, and prevents monopolistic practices on token pricing
```



document fetched 2026-11-15 00:36 · canonical https://reddit.com/r/LocalLLaMA/comments/1xw4ue3/why_dont_we_have_national_open_ai_models
