It seems to me there are some really fundamental questions about copyright and use posed here that have been submerged in the background of the internet rush to

It seems to me there are some really fundamental questions about copyright and use posed here that have been submerged in the background of the internet rush to digitize all forms of content and expression. Even though these generative tools may simplify the path to verbatim code (or image, or text) copying without attribution, and complicate determining what, if any, attributions are required, they didn't invent it. A browser coupled with a search engine facilitates such copying all the time. Like the generative models, the search engine is a tool that has read, for its own purposes, vast swaths of content, remembered key details about it, and serves a means to render the content up to a user. It's not the copying per se that causes trouble - creators make their material on the internet with the expectation and hope that it will be copied, by all those tools, into search engines and ultimately onto screens, so I can see it. It's that the chain of attribution is lost (or with OS code, the license imprint, which is really just another form of attribution), and with the attribution, any hope that the creator will benefit from the copying. But actually carrying attribution forward is going to be hard. Things that come out of a generative model that are substantially identical to a particular input do so for one of two reasons: one particular training input to the model is overwhelmingly the "best" source of response to the prompt, or a training input has been repeated over and over again in the training set so as to become the consensus response to one or more prompts. The first might reasonably feasible to track down, although it's bound to be computationally expensive. The second ... really tough, since it forces the model to "know" which of those many training sources is the one that should attributed. The widely circulated example of an image generation model reproducing Steve McMurray's famous photo of Sharbat Gula, a green-eyed Afghani girl that appeared on National Geographics cover in 1985 shows the problem. Do a Google search for "green-eyed Afghan girl" and you'll find hundreds of copies of varying resolution and definition, and hundreds more derivative versions, of McMurray's photo. A model spitting out yet another derived, but nearly identical version is likely drawing from hundreds of those images itself, not some original root, copyrighted golden copy. Which should it attribute?