users-ai-assistant-wrote-less-secure-code-weremechanismsingle paper
Users with an AI assistant wrote less secure code and were more confident it was secure.
Capability: Writing secure code and dependencies
Sources
- Users with an AI assistant wrote less secure code and were more confident it was secure.
Disagreeing is the most useful thing you can do here. Both sides of every contested claim in this catalog were assembled by the same person, which is its weakest point.
Related claims
- Models complied with insecure completions a large fraction of the time, and more capable models were more likely to suggest insecure code.Writing secure code and dependencies