‘Vibe coding’ using LLMs susceptible to most common security flaws
programming.dev·35w
Preview
Report Post

104

OpenAI’s models were most likely to produce vulnerable code in tests by Backslash Security.

You must log in or # to comment.

HotTopControversialNewOld

Chat

Who would have thought letting an AI shit out slop code would produce insecure, slop code? I’m shocked, I tell you. Shocked.

Just keep giving it more prompts and editing the output until the squiggles go away and it runs. It will be just fine, surely.

deleted by creator

Even classic coders doesn’t automatically write safe and secure code.

That’s the crux of it. People talk like most coders are some kind of know all security and best practices wise.

That’s just a lie or virgin innocence…

Similar Posts

Loading similar posts...