Thanks, you saved me the effort of pointing out that the 'big plagiarism machine' label is not a thing: nothing more than misunderstanding (except when it's deliberate propaganda), with no real basis in reality.
There are certainly issues with big companies using copyright material without permission, but they are far more nuanced and nothing like the label implies. > > On 11 Jul 2026 at 23:58, Gregory Casamento <[email protected]> wrote: > > > > > Riccardo, > > > > I meant to add what I am saying below as a postscript. > > > > > On Sat, Jul 11, 2026 at 2:36 PM Gregory Casamento <[email protected]> > wrote: > > > > > > > > > On Sat, Jul 11, 2026 at 8:43 AM Riccardo Mottola > > <[email protected]> wrote: > > > > > > > > > Hi, > > > > > > Foreword 1: when for brevity I am referring to AI, I am referring to the > > > latest trend of the "big plagiarism machine" to cite David (or I prefer > > > Chomsky's wording "High Tech Plagiarism"): large systems that suck code > > > from everywhere, closed and open source, that scavenge every available > > > resource and site (including our own) and digest everything. Well aware > > > that AI could be local, integrated in Lisp, neural networks in chips and > > > a lot of other usages of AI and Neural Networks all branded of "AI". > > > > > > > > Here's the thing... Chomsky's assessment demonstrates a fundamental > misunderstanding of how modern AI systems operate. Also, anyone who > espouses this position displays a similar misunderstanding; here's why: > > > > > Why the "Plagiarism" Label Fails > > > Novel Output: AI generates entirely original combinations of words rather > than copying existing text. > > Dynamic Synthesis: Large language models merge thousands of disparate > concepts to create unique insights. > > Lossless Storage Myth: Models do not store database copies of their > training data to copy from. > > Lossy Compression: The AI internalizes abstract concepts and semantic > rules rather than specific phrases. > > > If all the LLM represents is a statistical prediction machine, then I could > understand your position on this point, but it isn't. It utilizes a neural > network to support the above functionality. This means it can combine > concepts it is exposed to in novel ways, so again. > > > > The following citations support my argument: > Mitchell, M., & Wu, X. (2026). LLMs, reasoning and plagiarism. arXiv. > https://arxiv.org/html/2601.02380v5 > Liberman, M. (2026). Chomsky and the origins of AI research. > Language Log. https://languagelog.ldc.upenn.edu/nll/?p=72547 > Delétang, G. (2023). What Noam Chomsky gets wrong about AI. MLPowered. > https://www.mlpowered.com/posts/chomsky-gets-wrong/ > > > Yours, GC > -- > > > > > Gregory Casamento > GNUstep Lead Developer / Black Lotus, Principal Consultant > http://www.gnustep.org - http://heronsperch.blogspot.com > > https://www.openhub.net/languages/objective_c > > > > >
