[emodendroket]

To be clear, my suspicion is that this is so unlikely to happen unintentionally that it does not represent a real risk. If the issue is that I can force it to generate infringing output if I really want to, it is an argument against the Web browser too, since I could just as easily use the copyright-unsafe "copy" feature.

[ghoward]

I don't entirely agree.

Whereas using the browser's copy feature requires the user to have intent to use it, getting Copilot to produce exact code does not. And proving that intent is not easy.

I think companies will see that such code can be exactly reproduced and decide to stay away from Copilot. I hope they do. In fact, I am less willing to take outside contributions for my own code, even for bug fixes, just because of the risk that that code came from Copilot.

[emodendroket]

That makes sense if you ignore the idea that such a thing would seem unlikely to happen without intent, which was the key thing in the post you’re replying to.

[ghoward]

Unlikely stuff will always happen with enough use. There are billions of lines of code in the world. There will be enough copyright violations. Even on single multi-million line codebases, there will be violations.