> Sent: Monday, March 30, 2026 at 11:54 AM > From: "Jean Louis" <[email protected]> > To: [email protected] > Cc: "Björn Kettunen" <[email protected]> > Subject: Re: [emacs-tangents] Is org-mode accepting AI-assisted babel ob- > code updates? > > On 2026-03-29 13:11, Björn Kettunen wrote: > >> Who Does That Server Really Serve? - GNU Project - Free Software > >> Foundation: > >> https://www.gnu.org/philosophy/who-does-that-server-really-serve.html > >> > >> The GNU philosophy piece "Who Does That Server Really Serve?" warns > >> against exactly the kind of dependency you're describing—but it also > >> assumes a world where users are forced to interact with software as a > >> service. That assumption is increasingly outdated. Today, I can run > >> Qwen, Llama, DeepSeek, or any number of open‑weight models entirely > >> locally on my own hardware. Hugging Face, Allen AI, IBM, Apertus, and > >> others are making this the norm. When I generate code, it's on my > >> machine, with models that are publicly available, often under > >> permissive or free software licenses. The "proprietary service" > >> framing doesn't apply when the user controls the tool end‑to‑end. > > > > The assumption isn't so outdate when users predominantly interact with > > SAS LLM's such as Claude. > > Sidenote: We should not call them AI but LLM. The former obfuscates > > what these actually are. > > If you mean Allen AI I have mentioned, that is name of the company. > > An LLM (Large Language Model) is fundamentally a statistical model — a > massive set of learned parameters (weights) that predict the next token > in a sequence based on patterns in its training data. > > The GNU project (and people like Richard Stallman) are correct to push > back against casually calling a raw LLM "artificial intelligence." It > can feel like marketing hype that overstates what it is. A bare model is > more like a very advanced lookup/completion tool than true intelligence.
Precisely. > A single LLM call (prompt → output) is limited and often brittle. > > But when you wrap it in agentic workflows, tool use, memory, planning > loops, multi-step reasoning, self-correction, external tools (search, > code execution, calculators, APIs), etc., the overall system can exhibit > behaviors that reasonably match many classical and modern definitions of > "artificial intelligence." > > So there are you are, try it out, you will understand it. This is why > companies and researchers increasingly talk about LLM-based AI systems > or AI agents rather than just "the LLM." The model is the core engine, > but the surrounding architecture is what makes it intelligent in > practice. > > -- > Jean Louis > > --- > via emacs-tangents mailing list > (https://lists.gnu.org/mailman/listinfo/emacs-tangents) > --- via emacs-tangents mailing list (https://lists.gnu.org/mailman/listinfo/emacs-tangents)
