4 points | by noyesno 5 hours ago ago
4 comments
The methodology is a joke, this barely qualifies as research
TLDR;
”1. Chinese LLMs produce more vulnerable code when prompted with a U.S. government persona than without—and the vulnerabilities are highly obfuscated.
2. Chinese LLMs inject PRC-aligned political bias into both the answers and code they generate. ”
> we do not have proof at this point that code flaws are intentionally introduced.
If it is the way it is (unsecure), what does proof of intent matter?
The methodology is a joke, this barely qualifies as research
TLDR;
”1. Chinese LLMs produce more vulnerable code when prompted with a U.S. government persona than without—and the vulnerabilities are highly obfuscated.
2. Chinese LLMs inject PRC-aligned political bias into both the answers and code they generate. ”
> we do not have proof at this point that code flaws are intentionally introduced.
If it is the way it is (unsecure), what does proof of intent matter?