# Push to prohibit open source models or is there actually some truth to it? — retained record

> Retained reddit record from reddit:ClaudeCode.

Canonical: https://highsignal.app/data/records/96ca41e5201e27ed

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


source: [reddit](https://highsignal.app/data/reddit) (reddit:ClaudeCode)  published 2026-09-13 16:28 ingested 2026-09-14 02:38

[original source ↗](https://www.reddit.com/r/ClaudeCode/comments/1wfcax3/push_to_prohibit_open_source_models_or_is_there/)

## retained text



```
I work with Claude Code every day - excessively. In light of the recent drama about slowing down AI development because “it’s going to destroy us”: Honestly, I don’t get it. The frontier models are so unbelievably bad at coding. You need to supervise over every PR on the micro level (don’t write 4 gigantic lines of comment for no reason) and need to spec out the exact architecture and methodology before every task.  
Now granted, integration testing (unit testing as well) has never been better or easier. But as a software engineer, this still doesn’t feel nowhere near something that could replace me.  
To me, it’s essentially a hyperactive junior that is permanently online and doesn’t evolve past their stage. But once specced out to the teeth, it will do what you want.  
How is that even close to AGI?

Are these statements by Amodei, and now Altman + Musk just a big push to prohibit open source models? Or what the hell am I missing?
```



document fetched 2026-09-14 00:35 · canonical https://reddit.com/r/ClaudeCode/comments/1wfcax3/push_to_prohibit_open_source_models_or_is_there
