On 15/01/26 at 17:43 +0100, Sylvain L. Sauvage wrote: > Le jeudi 15 janvier 2026, 17:02:05 heure normale d’Europe centrale Lucas > Nussbaum a écrit : > >[…] > > If you haven't already, I would recommend getting first hand experience > > with LLM-assisted coding. Using it to modify existing code is very > > different from using it to do "Vibe coding", where you generate large > > amounts of code without even reading it. > > Okay, but you said: > “I let the agent write an initial version of the code, and then I review > it.” > > “Initial version,” even if that’s not “pure vibe coding,” it very much > smells like it.
I probably should have written "initial version of the patch/change" indeed. > Nonetheless, if you ask the tool to add a function to one of your files… > maybe it’s incremental… but maybe it’s just regurgitating almost verbatim > someone else’s code (changing their tab indentation and camel case > notation to your code’s 4-space indentation and snake case). My feeling is that it would not happen like that. I would first ask for a change in the code to add a missing feature, then as a second step I would ask for refactoring. Mainly because it's much easier to review the two steps separately. Since the new function would be a result of the refactoring, it's less likely to be code regurgitated from elsewhere. But I see some people writing about first designing a detailed spec of what they want the LLM to do, and then asking the LLM to implement. In that case, it might be more likely to regurgitate code captured elsewhere. I don't really know since that's not my workflow. > Maybe you won’t agree on “regurgitating,” maybe you’d call that > “reinventing”… but if a human studied a lot of codes (and was known to > have done that) and managed to write a (non-trivial) function exactly as > it was in that corpus (modulo cosmestics), they would be suspected of > plagiarism. > > 2. In any case, and more importantly, most of these tools have been > trained on other people’s work, regardless (and stripping them) of their > copyrights and licences. They are poisoned. That's true. I must admit I did not follow much the legal side of things. However, if the original code was free software, I think it can be argued that training falls under Freedom #0 or #1. Lucas
