Qwen3-Coder writes less secure code when told it's for a US persona

Re:Qwen3-Coder writes less secure code when told it's for a ... by stymaar on Wednesday October 21, 2026 @09:27AM (#50180600)

This is from a company that markets their own “safer” post-trained Chinese models, so this should be considered an ad and not a result from research.

[ Parent | on HN ]

Re:Qwen3-Coder writes less secure code when told it's for a ... by hk__2 on Wednesday October 21, 2026 @09:31AM (#50180663)

It can be both at the same time. An interesting result from a good research can be a free ad for the company.

[ Parent | on HN ]

Re:Qwen3-Coder writes less secure code when told it's for a ... by kennywinker on Wednesday October 21, 2026 @11:03AM (#50182247)

Sure, but based on the obvious biases (cherry picked model, cherry picked prompts) I don't think it's good research.

[ Parent | on HN ]