▲ 365 ▼ "participants who had access to an AI assistant wrote significantly less secure code" and "were also more likely to believe they wrote secure code" - 2023 Stanford University study published at CCS23 (arxiv.org) submitted 2 years ago by cypherpunks@lemmy.ml to c/technology@lemmy.ml 38 comments fedilink hide all child comments
[–] Hasherm0n@lemmy.world 7 points 2 years ago Until you find out those were also built by a junior using an llm to help 🙃 permalink fedilink source parent