Don't install anything, use an LLM to write everything from scratch. It may have bugs, but no one will know how to exploit them, especially when closed source.
Code is cheap and is becoming cheaper by the day. We need new paradigms.
Don't install anything, use an LLM to write everything from scratch. It may have bugs, but no one will know how to exploit them, especially when closed source.
Code is cheap and is becoming cheaper by the day. We need new paradigms.
So no external libraries for anything? Billions of lines of code that duplicate the same thing n-times across an organization?
And the benefit is the obscurity of "no one will know how to exploit them"?
No, thanks.
Code is becoming so cheap that all you need is a bunch of api's for hardware and your computer will build to that spec. And you can define it in natural language.
LLMs have been used to scan binary blobs for exploits already. What would be more effective is a system designed with multiple layers of security so any one exploit is largely useless.
They would have to have access to and scan your individual binary. You'd have to describe how you can write a system with multiple layers of security generally for most problems, because I don't see that as being possible.
Next: the back doors are written by the LLM!
You think we are doing well against back doors right now? Pfft.