📊 Veilles thématiques

📰 Intelligence Artificielle ✕ Effacer le filtre

How the Pope’s Magnifica Humanitas offers a template for individuals to meet the AI moment

Intelligence Artificielle
Pope Leo XIV’s new encyclical on artificial intelligence includes a statement that warrants serious attention from technologists and policymakers: “Technology is never neutral.” Magnifica Humanitas (“Magnificent Humanity”) is a clarion call to all people to act with courage and s…

Profiling in PyTorch (Part 1): A Beginner's Guide to torch.profiler

Intelligence Artificielle
Back to Articles Profiling in PyTorch (Part 1): A Beginner's Guide to torch.profiler Published May 29, 2026 Update on GitHub Upvote 7 +1 Aritra Roy Gosthipaty ariG23498 Follow Sayak Paul sayakpaul Follow Sergio Paniego sergiopaniego Follow Rémi Ouazan Reboul ror Follow Pedro Cuen…

Claude Opus 4.8 : Anthropic met l’accent sur un modèle plus honnête face à ses propres erreurs

Intelligence Artificielle
Sommaire L'amélioration mise en avant par Anthropic pour Claude Opus 4.8, publié le 28 mai 2026, tient en une métrique: le modèle serait, selon l'éditeur, quatre fois moins susceptible que son prédécesseur de laisser passer sans les signaler des défauts dans le code qu'il a produ…

Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL

Intelligence Artificielle
Back to Articles Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL Published May 27, 2026 Update on GitHub Upvote 1 Amine Dirhoussi aminediroHF Follow Quentin Gallouédec qgallouedec Follow Kashif Rasul kashif Follow Lewis Tunstall lewtun Follow Edward Bee…

GPT plus confiant sur les tâches difficiles où ils se trompe le plus, selon un preprint USC/Berkeley

Intelligence Artificielle
Sommaire La confiance affichée par GPT-4o, ChatGPT et GPT-o3 dépasse leur précision réelle, et l'écart se creuse précisément sur les tâches difficiles - sur les tâches faciles, à l'inverse, les modèles se sous-évaluent. Cet effet hard-easy quantifié dans un preprint sous révision…

Harness, Scaffold, and the AI Agent Terms Worth Getting Right

Intelligence Artificielle
Back to Articles Harness, Scaffold, and the AI Agent Terms Worth Getting Right Published May 25, 2026 Update on GitHub Upvote - Sergio Paniego sergiopaniego Follow Aritra Roy Gosthipaty ariG23498 Follow When a field evolves quickly, its vocabulary often evolves faster than its sh…

Specialization Beats Scale: A Strategic Variable Most AI Procurement Decisions Overlook

Intelligence Artificielle
Back to Articles Specialization Beats Scale: A Strategic Variable Most AI Procurement Decisions Overlook Team Article Published May 22, 2026 Upvote - Erick Lachmann ErickvL Follow Dharma-AI Pimenta de Freitas Cardoso GabrielPimenta99 Follow Dharma-AI When a model’s training histo…

Anthropic loue Colossus 1 à 1,25 Md$/mois sur un parc xAI qui plafonne à 11% de capacité

Intelligence Artificielle
Sommaire Anthropic vient de signer un contrat à 1,25 milliard de dollars par mois jusqu'en mai 2029, soit plus de 40 milliards de dollars sur la durée, pour réserver un parc de GPU dont l'opérateur lui-même qualifiait, quelques semaines plus tôt, le taux d'utilisation effectif de…

Spain’s Former PM Investigated in Airline Bailout Probe

Intelligence Artificielle
Reported by Antonio Baquero OCCRP Former Spanish Prime Minister José Luis Rodríguez Zapatero is under investigation for influence peddling and potential money laundering in connection with a government airline bailout, a court spokesperson confirmed to OCCRP Tuesday. Zapatero, a …

AI chatbots are giving out people’s real phone numbers

Intelligence Artificielle
People report that their personal contact info was surfaced by Google AI—and there’s apparently no easy way to prevent it. A Redditor recently wrote that he was “desperate for help”: for about a month, he said, his phone had been inundated by calls from “strangers” who were “look…

AlphaEvolve: Gemini-powered coding agent scaling impact across fields

Intelligence Artificielle
May 7, 2026 ScienceAlphaEvolve: How our Gemini-powered coding agent is scaling impact across fieldsAlphaEvolve team Your browser does not support the video tag.A year ago, we introduced AlphaEvolve, a Gemini-powered coding agent for designing advanced algorithms. We showed that A…

Recent books from the MIT community

Intelligence Artificielle
Priority Technologies: Ensuring US Security and Shared ProsperityEdited by Elisabeth B. Reynolds, professor of the practice of urban studies and planning and former executive director of the MIT Task Force on the Work of the Future MIT PRESS, 2026, $24.95 The Shape of Wonder: How…

World models

Intelligence Artificielle
10 Things That Matter in AI Right NowSee the full list AI systems have already gained impressive mastery over the digital world, but the physical world is still humanity’s domain. As it turns out, building an AI system that can compose a novel or code an app is far easier than de…