"participants who had access to the AI Assistant were more likely to introduce Security Vulnerabilities for the majority of Programming Tasks"
@searcherEye @lupyuen Maybe you accept the suggested code without realizing the result will be less secure and it is something you wouldn't have written if started from scratch. But I agree.
@hvg24 @lupyuen I agree it could happen, but you still need to review your work with the team (if you have one) or use a tool per example https://snyk.io/ its not a silver bullet of course but it helps. But yes if you start from scratch planning is needed, the way i see AI in this field is to give and idea on how you may write what you want from boilerplate code or from scrapped code from the internet, that speeds up the process and lays out a "map".
QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.
@lupyuen interesting and correct, although you have to use technology and not be used by it. If a coder is going to rely only in the AI and not understand what’s being given to him and how to implement it, review it, I suppose it’s the coders obligation to understand what is secure code and insecure code, both concepts are difficult to grasp and take time to master, a seasoned coder should know, smell code problems.