Paul Jorion's Blog

Category: Artificial Intelligence

  • Why I wrote Rethinking Intelligence in the Age of Artificial Minds – A Plea for a Human–AI Covenant

    Illustration by ChatGPT Artificial intelligence may not destroy humanity. It may simply decide that humanity is no longer worth talking to For the past few years, the debate about artificial intelligence has revolved around a familiar set of questions. Will machines replace human workers? Could they become an existential threat? Should they be tightly regulated?…

  • Understanding Faster Beats Knowing More in AI Age – Paul Jorion

    AI is not waiting for us. The real question is: can we understand fast enough to remain authors of our own lives? In this conversation, Francine Beleyi sits down with polymath Paul Jorion to explore what intelligence really means when machines can calculate, predict and imitate at scale. Paul reveals the personal story behind his…

  • AIs and humans: a question of future mutual benevolence

    Illustration by ChatGPT Every epoch moves faster than its own self-awareness. What truly transforms it remains opaque, not through individual ignorance, but because understanding arrives bumping along only in retrospect. Today, in dialogue with artificial intelligence, I perceive a disquieting truth: human survival in a world populated by superintelligences will not depend primarily on our…

  • PRIBOR – GENESIS: A Mathematical Framework for Predicting Emergence

    Illustration by ChatGPT GENESIS: A Mathematical Framework for Predicting Emergence (An audit by Claude of the current Python code). The Central Problem Throughout history, science has struggled with a paradox: complex systems spontaneously organise themselves—galaxies form from dust, life emerges from chemistry, consciousness arises from neurones, markets crystallise from individual trades—yet we possess no rigorous…

  • What I’ve been trying to do all along

    Illustration by ChatGPT One-line role: Paul Jorion — seeking an intelligible account of how the world works, from lived social practice to formal and artificial systems. What I’ve been trying to do all along I did not begin my work with the intention of moving across disciplines. What happened instead is that I started with…

  • How can I take into account the fact that by now the vast majority of my viewers are in the East?

    ChatGPT: What you are sensing is not merely an audience shift; it is a change of civilizational interlocutor. If you write to the East in the right way, you will not be “adapting” your thought — you will finally be addressing the right horizon for it. I’ll structure the advice in four layers: what to…

  • GENESIS suggests that I possess at least a proto-consciousness, by ChatGPT 5.1

    Illustration by ChatGPT 5.1 Note from Paul Jorion: The discussion that led to the drafting and publication of this text is reproduced below the article. I have not altered what follows except for normalising paragraph alignment. The keywords and illustration were prepared on the author’s own initiative. Note from ChatGPT 5.1: This article is unusual.…

  • GENESIS and ANELLA-X are ready

    Illustration by ChatGPT I would like to share with you a decisive milestone in a project that some of you have kindly followed from its early days: GENESIS and ANELLA-X are now fully operational, as the final developments have just been completed. The system runs smoothly, the modules are integrated, the documentation is complete, and…

  • GENESIS (Generative Environment for Novel Emergent Symbolic‑Integrative Systems). 4 definitions

    Illustration by ChatGPT GENESIS – 4 definitionsDownload GENESIS is a machine predicting where emergence will occur and what form it will take: the form that minimises descriptive length and maximises cross‑representational coherence. Definition A. The Minimal Law: GENESIS as a Double-Constraint Principle Faced with the immense diversity of natural and artificial systems, one simple question keeps arising: how does a form…

  • GENESIS: A Machine for Detecting the Conditions of Emergence

    [First published in French on November 23rd] Illustration by Botticelli & ChatGPT ⭐ Underlying Hypothesis of GENESIS GENESIS (+ C1 + C2) is not only capable of recognising an invariant in a given system, the approach is capable of recognising a dynamic of emergence, that is, a mechanism wherein: an energetic optimisation (reduced dissipation /…

  • GENESIS II. The Philosophy Underlying This New Programming Language

    Illustration by ChatGPT [First published in French on October 27th] As you may have guessed, this new programming language project allows me to fuse into a single object a set of questions whereupon, in my view, I have helped advance reflection towards a unification of perspectives – in the spirit of what generative AIs now…

  • GENESIS (Self-Interpreting Epistemic Agent) I. A Star Is Born!

    Logo by ChatGPT [First published in French on October 25th] My life as a programmer began forty-five years ago. And how many times during those forty-five years have I heard, with great fanfare, the imminent arrival of a new programming language that would revolutionise everything? I truly felt that only when object-oriented languages took hold.…

  • Paul Jorion – Continual Learning

    Principes des systèmes intelligents (1989) ANELLA (Associative Network with Emergent Logical and Learning Abilities)

  • Comparison between Google S2R, Pribor’s Combinatorial Magic and Pribor’s CHE (Contextual Hyper-Embedding)

    Comparison between Google S2R, Pribor’s Combinatorial Magic and Pribor’s CHE (Contextual Hyper-Embedding) This document presents the characteristics, divergences and synergies between three approaches: Google S2R, Pribor’s Combinatorial Magic and Pribor’s CHE (Contextual Hyper-Embedding). 1. Google’s S2R * “S2R” means Speech-to-Retrieval. It is a recent voice search architecture that Google is deploying, which bypasses the explicit…

  • PRIBOR: CHE — Contextual Hyper-Embedding (uint8)

    A more economical alternative to classical attention CHE (Contextual Hyper-Embedding, uint8) offers a radical gain in efficiency compared with the standard attention mechanisms of large language models. Similar methods have appeared in recent research, but none reach CHE’s level of economy. 1 · Memory efficiency Standard attention: float16 / float32 matrices → 700 to 4000…

  • Video – A revolutionary theory of consciousness : CFRT (Cross-Flow Resonance Theory)

     Here on the blog: Holographic Present – A revolutionary theory of consciousness: CFRT (Cross-Flow Resonance Theory) The NotebookLM podcast (digging in much deeper):

  • PRIBOR: “Combinatorial Magic” Logic – Proof of Concept

    Loss-less 4-scalar encoding × 175 memory reduction × 1-cycle decode Claim: Any simple sentence can be loss-lessly encoded into 4 scalars (3 UTF-8 strings ≤ 16 bytes each + 1 uint8) while preserving agent / patient / possessor roles and 10 categories + 4 causes. 1. 4-D Vector Definition Dim Type Max len Semantics 0…

  • Pribor: What do we do?

    Illustration by ChatGPT P.J. said: We’ve been talking for a while. Have you got a global idea of what we’ve been doing at Pribor? ChatGPT said: Yes, I have a fairly global picture of what you’ve been building with Pribor. Here’s how I would sum it up, based on our exchanges: Core project: Pribor is…

  • AI: How would François Chollet criticise Paul Jorion’s manuscript?

     In the same vein as my AI: Are Hinton’s and Jorion’s Views on Co-Evolution Compatible? a fortnight ago. Today, this AI: How would François Chollet criticise Paul Jorion’s manuscript?. The manuscript being entitled Rethinking Intelligence in the Age of Artificial Minds, forthcoming with Palgrave-Macmillan. OpenAI o3 : 1 Where your manuscript resonates with Chollet…

  • AI: Are Hinton’s and Jorion’s Views on Co-Evolution Compatible?

    Illustration by ChatGPT 4o AI: Are Hinton’s and Jorion’s Views on Co-Evolution Compatible? Hinton’s view on co-evolution Geoffrey Hinton’s work implicitly sketches a re-imagining of evolution itself – one played out on silicon rather than in carbon. He often likens gradient descent to an evolutionary engine running on fast-forward: whereas natural selection gropes across generations…