GitHub Copilot Generated Insecure Code in 40% of Circumstances During Experiment

Re:GitHub Copilot Generated Insecure Code in 40% of Circumst... by lampe3 on Friday September 03, 2021 @08:02AM (#28403042)

I'm using copilot now for some time and yeah it's more a toy than real help right now.

The only time it really helped when I needed to create a named list of char codes.

When it comes to more complex code than checking the code of copilot takes the same time as writing it. 90% of the time I needed to correct copilot.

For me, tools like linters are way more helpful then. If I could only use ESLint or copilot, I would go 100% of the time with ESLint.

[ Parent | on HN ]

Re:GitHub Copilot Generated Insecure Code in 40% of Circumst... by harlekein on Friday September 03, 2021 @10:48AM (#28404933)

I think another risk with getting Copilot to start out, is that it might nudge you into a direction you wouldn't have gone into otherwise.

Whether that is better or not, I suppose, it depends.

[ Parent | on HN ]

Re:GitHub Copilot Generated Insecure Code in 40% of Circumst... by lampe3 on Friday September 03, 2021 @04:38PM (#28409114)

I'm working on a HTML Tokenizer in Deno/Typescript.

Copilot only helps with boilerplate code which could be handled by good intellisense.

When it tries to generate a function from the function name it fails so hard that it is more in your way then helpful.

[ Parent | on HN ]