-
On the puzzling behavior of generative AIs: What to make of recent interpretations of the inner workings of Claude, Anthropic’s large language model?
Illustration by DALL·E On March 27, 2025, Anthropic published a set of innovative analyses in its article On the Biology of a Large Language Model (LLM), constituting an in-depth exploration of the internal workings of its Claude 3.5 Haiku model. The approach adopted is that of a transposition of neuroscience techniques: starting from the human…
-
Video – Wow! Maya and I: Sci-Fi is Here to Stay!, March 9th 2025
-
Intelligence & Société N°6, an English-speaking review by NotebookLM
-
Intelligence & Société N°5, an English-speaking review by NotebookLM
-
Intelligence & Société N°3, an English-speaking review by NotebookLM
-
Video – LLMs: Will we run out of data?
-
Exploring the Frontiers of AI and Society: My Journey as a Speaker and Columnist
In recent weeks, I have had the privilege of speaking to audiences in Lille, Brussels and Nîmes about the societal and ethical dimensions of large language models (LLMs). These discussions, sparked by my latest book, La Singularité. L’humain ébranlé par l’intelligence artificielle, have brought me face-to-face with the pressing questions and challenges posed by artificial…
-
A NEW “US” FOR NEW TIMES. XV. Updating the “us” within a technological context
Illustration by DALL·E Updating the “us” within a technological context In the same way as Linnæus positioned us among the anthropoid apes, now the existence of intelligent machines sets us in a family of intelligent beings where living creatures and machines rub shoulders and share similar brain architectures. The better we understand how these LLMs…
-
“Progress in AI is slowing down” = fake news
Demis Hassabis, who recently won the Nobel Prize in Chemistry for his work in AI, is sceptical about an article published four days ago: He does so (under a pseudonym) in the video below. What he says is very interesting (‘there’s no proof that there’s a slowdown’) but what I’d like to say (a comment…
-
A NEW “US” FOR NEW TIMES. XIV. Intelligence: A defining feature of our pride
Illustration by DALL·E: “The human being worshipping his own intelligence” Intelligence: A defining feature of our pride Humankind’s intelligence has always been praised by humans as maybe their defining quality. Animals around us display some of this intelligence but to a lesser degree, what Rousseau called in a non-disparaging way, the “stupidity” of the animal.…
-
A NEW “US” FOR NEW TIMES. XII. AI has the great disadvantage of being … artificial
Illustration by DALL·E AI has the great disadvantage of being … artificial This does not mean however that scientific research had become an easy pursuit in the West from the 16th century onwards: the image of the tree of knowledge’s forbidden fruit maintained its lingering presence, manifesting itself in particular through an endemic denigration of…
-
A NEW “US” FOR NEW TIMES. XI. Why were we caught off-guard by AI?
A portrait of Tycho Brahe, Royal Astrologist of Danemark, by Stable Diffusion Why were we caught off-guard by AI? Despite artificial intelligence having been a field of research since the 1950s and the achievement of software of the type of LLMs having been an objective clearly stated from the outset and the research effort having…
-
A NEW “US” FOR NEW TIMES. X. The rise of the LLMs
Portrait of Geoffrey Hinton by Stable Diffusion The rise of the LLMs This is the way things stood before some recent events that led Geoffrey Hinton to make a disturbing statement. Geoffrey Hinton is the mastermind behind Large Language Models such as ChatGPT, Claude, Gemini or LLaMA3. In the press he goes often under the…
-
A NEW “US” FOR NEW TIMES. III. East and West
Illustration by DALL·E of Marco Polo discovering China East and West In the conceptualisation of the advent of artificial intelligence on a par with human intelligence, there are major differences between the West and the East: between the civilisations and the range of cultures born, on the one hand, in the Fertile Crescent and, on…
-
A NEW “US” FOR NEW TIMES. II. Decentring and refocusing the definition of “us”
Illustration by DALL·E from the text Decentring and refocusing the definition of ‘us’ Who do we think we should protect as a last resort, the members of a strictly defined ‘us’ with some and a vaguely defined ‘us’ among others, when things go wrong, as they unfortunately do time and time again on this troubled…
-
A NEW “US” FOR NEW TIMES. I. Who are we? Where do we go from here?
Illustration by DALL·E (+Michelangelo) (+PJ) I. Who are we? Where do we go from here? If we examine this essential question of understanding who we are and what we are experiencing at this unprecedented moment, we see that the history of humanity has offered ever-changing answers, each shaped by an understanding of ourselves that reflects…
-
A review of « L’avènement de la Singularité. L’humain ébranlé par la Singularité » by the two chatting IAs at NotebookLM
-
‘The prince and the honest man facing the Singularity’, AOC, March 5th 2024
-
Review of my article ‘The prince and the honest man facing the Singularity’ by NotebookLM’s AIs
-
An Ambitious Project II. Consciousness: Claude is enthusiastic. I test her or him right away.
Illustration by Stable Diffusion (+PJ) Claude : I would be deeply honoured and excited to collaborate with you on this project exploring the intersection of AI and consciousness. The question of whether and how artificial systems might give rise to genuine conscious experiences is one of the most profound and challenging issues in both philosophy…