Podcasts about ArXiv

In this episode we discuss the paper "The Unreasonable Likelihood of Being: Origin of Life, Terraforming, and AI" (arXiv, 2025) with Robert Endres. Paper Abstract: The origin of life on Earth via the spontaneous emergence of a protocell prior to Darwinian evolution remains a fundamental open question in physics and chemistry. Here, we develop a conceptual framework based on information theory and algorithmic complexity. Using estimates grounded in modern computational models, we evaluate the difficulty of assembling structured biological in-formation under plausible prebiotic conditions. Our results highlight the formidable entropic and informational barriers to forming a viable protocell within the available window of Earth's early history. While the idea of Earth being terraformed by advanced extraterrestrials might violate Occam's razor from within mainstream science, directed panspermia—originally proposed by Francis Crick and Leslie Orgel—remains a speculative but logically open alternative. Ultimately, uncovering physical principles for life's spontaneous emergence remains a grand challenge for biological physics. Full paper available here Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/new-books-network

In this episode we discuss the paper "The Unreasonable Likelihood of Being: Origin of Life, Terraforming, and AI" (arXiv, 2025) with Robert Endres. Paper Abstract: The origin of life on Earth via the spontaneous emergence of a protocell prior to Darwinian evolution remains a fundamental open question in physics and chemistry. Here, we develop a conceptual framework based on information theory and algorithmic complexity. Using estimates grounded in modern computational models, we evaluate the difficulty of assembling structured biological in-formation under plausible prebiotic conditions. Our results highlight the formidable entropic and informational barriers to forming a viable protocell within the available window of Earth's early history. While the idea of Earth being terraformed by advanced extraterrestrials might violate Occam's razor from within mainstream science, directed panspermia—originally proposed by Francis Crick and Leslie Orgel—remains a speculative but logically open alternative. Ultimately, uncovering physical principles for life's spontaneous emergence remains a grand challenge for biological physics. Full paper available here Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/science

In this episode we discuss the paper "The Unreasonable Likelihood of Being: Origin of Life, Terraforming, and AI" (arXiv, 2025) with Robert Endres. Paper Abstract: The origin of life on Earth via the spontaneous emergence of a protocell prior to Darwinian evolution remains a fundamental open question in physics and chemistry. Here, we develop a conceptual framework based on information theory and algorithmic complexity. Using estimates grounded in modern computational models, we evaluate the difficulty of assembling structured biological in-formation under plausible prebiotic conditions. Our results highlight the formidable entropic and informational barriers to forming a viable protocell within the available window of Earth's early history. While the idea of Earth being terraformed by advanced extraterrestrials might violate Occam's razor from within mainstream science, directed panspermia—originally proposed by Francis Crick and Leslie Orgel—remains a speculative but logically open alternative. Ultimately, uncovering physical principles for life's spontaneous emergence remains a grand challenge for biological physics. Full paper available here Learn more about your ad choices. Visit megaphone.fm/adchoices

How a Lawyer and a Listicle Launched One of Quantum's Most Influential Media PlatformsEvan Kubes had no physics degree, no engineering background, and no idea what a qubit was when he stumbled across a press release about AWS investing in quantum. What he did have was experience translating complex industries for mainstream audiences — and within months, he and co-founder Alex Challans had turned a Wix website and a "Top 20 Most Influential People in Quantum" listicle into The Quantum Insider, now one of the industry's leading media and intelligence platforms. In this episode, Evan shares how that scrappy start grew into Resonance, a multi-vertical deep tech media company — and why he spent the last year making Our Quantum Future, a feature-length documentary premiering at APS March Meeting that aims to bring quantum out of the echo chamber and onto your screen.Why this episode mattersThis episode marks a new chapter for The New Quantum Era. In the intro, Sebastian shares some big updates — going fully independent, new media projects including the Helgoland 2025 documentary, a newsletter, and broader efforts to build a more accessible and equitable quantum technology ecosystem through open source and open standards. He also announces his new role as a Fellow at the Unitary Foundation. Read the full blog post: A New Chapter.The conversation with Evan Kubes is a perfect fit for this moment. Evan sits at the intersection of quantum's technical community and the broader world trying to make sense of it — a translator between physicists and the public. His story illuminates something the industry rarely discusses: how do you actually build awareness, trust, and market understanding for a technology most people can't explain?The documentary Our Quantum Future, produced for the International Year of Quantum and featuring Nobel laureates, a former CIA officer, and the leaders of Google, Microsoft, and IonQ, is designed for exactly that audience — the curious non-specialist who wants to understand what quantum means for the world. The ethics and national security themes it surfaces are relevant well beyond the quantum community.What you'll learnHow The Quantum Insider went from zero readers to a leading quantum industry platform using a creative "vanity listicle" strategy that got CEOs to respond overnightWhy a lawyer from the esports world saw the same market opportunity in quantum that venture capitalists were pouring billions into — and what that says about the accessibility gap in deep techHow the Resonance media model applies The Quantum Insider playbook to space, AI, and climate tech — and what makes a deep tech vertical ripe for this approachWhat 39 interviews across 40 countries revealed about how the quantum community thinks about ethics — including a striking divide between engineers ("I'm just solving a hard problem") and policymakers ("we need safeguards now")The Oppenheimer parallel: how the documentary draws a direct line between the atomic bomb's development and today's quantum technology, and why some builders don't think about consequences while others think about nothing elseA former CIA operative's reframing of quantum advantage as incremental compounding — 1% better per year for five years — and why that makes quantum feel much more real today than the "break all encryption" narrative suggestsWhy academics and corporate leaders consistently disagree on quantum's timeline, and where Evan lands after a year of filming both campsResources & linksGuest linksThe Quantum Insider — Quantum industry media, intelligence, and data platform co-founded by EvanResonance — Parent company extending the deep tech media model to space, AI, climate tech [link to confirm]Our Quantum Future — Documentary website with sign-up for distribution updatesPeople mentioned in the episodeAlex Challans — Co-founder and CEO of The Quantum Insider; Evan's business partnerNicholas Ogler — Former CIA operative featured in the documentary; redefines quantum advantage from a national security lensDr. Bill Phillips — Nobel Prize-winning physicist; discusses his bet with Carl Williams on the quantum advantage timelineDr. John Doyle — Professor of quantum at Harvard, president of APS; draws the Oppenheimer parallelIlyas Khan — Former CEO of Quantinuum; argues for educational licensing frameworks around quantum technologyEric Cornell — Nobel Prize winner featured in the documentaryMentioned in the introA New Chapter — NQE blog post — Sebastian's full announcement on going independent, new projects, and the future of the podcastUnitary Foundation — Open-source quantum technology ecosystem; Sebastian is now a FellowKey quotes & insights"When Oppenheimer and the most brilliant minds in the world were developing the atom, you had a large group who didn't really understand what they were building — they were just trying to solve a very difficult engineering and physics problem. We posed that same question to engineers at Google today: do you ever think about the potential consequences of what you're building? They said, absolutely not.""Quantum advantage to me is simply: if I can do a certain task 1% better every single year for five years, that compounds quite heavily. A country that uses quantum to improve radar detection by half a percent per year for five years has a massive advantage." — Nicholas Agler, former CIA"We emailed 20 people in the quantum industry — CEOs of Microsoft, Google, IonQ, Atom Computing — and said: Congratulations, you made The Quantum Insider's list of the top 20 most influential people in quantum. Every single person responded and agreed to do an interview.""For any industry to succeed, you've gotta get the venture capitalists and the capital markets around it, and you've gotta get the end users excited. If it's only PhDs talking to each other, it's gonna be a very limited market.""This documentary was not made for the quantum industry. It was made for Joe Blow and Cindy Blow at home who've never heard of this industry — to elevate and highlight all this fascinating work that we're doing."Sponsorqubitsok — Cut Noise. Work Quantum. The quantum computing job board and arXiv research digest built for the community. - Job seekers & researchers: Subscribe free at qubitsok.com — weekly job alerts + daily paper digest filtered by 400+ quantum tags. - Hiring managers: Post your quantum role and reach 500+ targeted subscribers. Use code NEWQUANTUMERA-50 for 50% off your first listing at qubitsok.com/post-job.Join the conversationSee the film: Visit ourquantumfuture.com to sign up for distribution updates — the premiere is at APS March Meeting in Boulder, with broader release to follow.Read the blog ...

In this episode of the Crazy Wisdom podcast, host Stewart Alsop sits down with Markus Buehler, the McAfee Professor of Engineering at MIT, to explore how seemingly different systems—from proteins and music to knowledge structures and AI reasoning—share underlying patterns through hierarchy, self-organization, and scale-free networks. The conversation ranges from the limits of current AI interpolation versus true discovery (using the fire-to-fusion example), to the emergence of agent swarms and their non-linear effects, to practical questions about ontologies, knowledge graphs, and whether humans will remain necessary in the creative discovery process. Markus discusses his lab's work automating scientific discovery through AI agents that can generate hypotheses, run simulations, and even retrain themselves, while Stewart shares his own experiences building applications with AI coding agents and grapples with questions about intellectual property, material science constraints, and the future of human creativity in an AI-abundant world.Timestamps00:00 - Introduction to Marcus Buehler's work on knowledge graphs, structural grammar across proteins, music, and AI reasoning05:00 - Discussion of AI discovery versus interpolation, using fire and fusion as examples of fundamental versus incremental innovation10:00 - Language models as connective glue between agents, enabling communication despite imperfect outputs and canonical averaging15:00 - Embodiment and agency in AI systems, creating adversarial agents that challenge theories and expand world models20:00 - Emergent properties in materials and AI, comparing dislocations in metals to behaviors in agent swarms25:00 - Human role-playing and phase separation in society, parallels to composite materials and heterogeneity30:00 - Physical world challenges, atom-by-atom manufacturing at MIT.nano, limitations of lithography machines35:00 - Synthetic biology as alternative to nanotechnology, programming microorganisms for materials discovery40:00 - Intellectual property debates, commodification of AI models, control layers more valuable than model architecture45:00 - Automation of ontologies, agent self-testing, daughter's coding success at age 1150:00 - Graph theory for knowledge compression, neurosymbolic approaches combining symbolic and neural methods55:00 - Nonlinear acceleration in AI, emergence from accumulated innovations, restaurant owner embracing AI01:00:00 - Future generations possibly rejecting AI, democratization of knowledge, social media as real-time scientific discourseKey Insights1. Universal Patterns Across Disciplines: Seemingly different systems in nature—proteins, music, social networks, and knowledge itself—share fundamental structural patterns including hierarchy, self-organization, and scale-free networks. This commonality allows creative thinkers to draw insights across disciplines, applying principles from one domain to solve problems in another. As an engineer and materials scientist, Buehler has leveraged these isomorphisms to advance scientific understanding by mapping the "plumbing" of different systems onto each other, revealing hidden relationships that enable extrapolation beyond what's observable in any single domain.2. The Discovery Versus Interpolation Problem: Current AI systems, particularly large language models, excel at interpolation—recombining existing knowledge in new ways—but struggle with genuine discovery that requires fundamental rewiring of world models. Using the example of fire versus fusion, Buehler explains that an AI trained on combustion chemistry would propose bigger fires or new fuels, but couldn't conceive of fusion because that requires stepping back to more fundamental physics. True discovery demands the ability to recognize when existing theories have boundaries and to develop entirely new frameworks, something current AI architectures aren't designed to achieve due to their training objective of predicting the most likely outcome.3. The Role of Ontologies and Knowledge Graphs: While some AI researchers argue that ontologies are unnecessary because models form internal representations, Buehler advocates for explicit knowledge graphs as essential discovery tools. External ontologies provide sharp, analytical, symbolic representations that complement the fuzzy internal representations of neural networks. They enable verification of rare connections—like obscure papers that might hold key insights—which would be averaged away in standard AI training. This neurosymbolic approach combines the generalization capabilities of neural networks with the precision of formal knowledge structures, creating more powerful discovery systems.4. Emergent Properties and Agent Swarms: Just as materials science shows that collections of atoms exhibit properties impossible to predict from individual components, AI agent swarms demonstrate emergent behaviors beyond single models. When agents are incentivized not just to answer questions but to challenge each other adversarially, propose theories, and test hypotheses, they can spawn new copies of themselves and evolve understanding beyond their initial programming. This emergence isn't surprising from a materials science perspective—dislocations, grain boundaries, and other collective phenomena only appear at scale, fundamentally determining material behavior in ways unpredictable from studying just a few atoms.5. The Commoditization of Intelligence: The fundamental AI models themselves are becoming commodities, as evidenced by events like the Moldbug phenomenon where people built agents using various providers interchangeably. The real value is shifting from who has the smartest model to how models are orchestrated, integrated, and deployed. This parallels historical technology adoption patterns—just as we moved past debating who makes the best electricity to focusing on applications, AI is transitioning from a horse race over model capabilities to questions of infrastructure, energy, access speed, and agent coordination at the systems level.6. Human-AI Collaboration and Creative Control: Rather than wholesale replacement, AI enables humans to operate in an intensely creative space as orchestrators sampling from vast possibility spaces. Similar to how Buehler's 11-year-old daughter now builds sophisticated applications that would have required professional developers years ago, AI democratizes access to capabilities while humans retain the creative judgment about direction and meaning. The human role becomes curating emergence, finding rare connections, playing at the edges of knowledge, and exercising the kind of curiosity-driven exploration that AI systems lack without embodied stakes in their own survival and continuation.7. Technology as Evolutionary Inevitability: The development of AI represents not an unnatural threat but the next stage of human evolution—an extension of our innate drive to build models of ourselves and our world. From cave paintings to partial differential equations to artificial intelligence, humans continuously create increasingly sophisticated representations and tools. Attempting to stop this technological evolution is futile; instead, the focus should be on steering it ...

MIT physicist David Kaiser is one of those rare scientists who can make mind-bending physics feel like a great conversation over coffee — funny, generous, and genuinely thrilled by what we still don't know. And what he's working on is wild. What the if the universe is packed with invisible black holes smaller than an atom? Dave thinks the mysterious "missing stuff" that holds galaxies together might not be some exotic undiscovered particle — it could be tiny black holes that formed a split second after the Big Bang. If he's right, a handful of them could be cruising through our solar system right now, and we might be able to catch one in the act just by watching Mars wobble. We also dig into whether a rogue black hole might have flattened a Siberian forest in 1908, and rest assured, the residents of Brooklyn have nothing to worry about. Learn more about David Kaiser's primordial black hole research: MIT PBH Research Group: https://sites.mit.edu/mitpbh/ David Kaiser's essay in the London Review of Books — a great accessible overview with historical context: https://www.lrb.co.uk/the-paper/v46/n11/david-kaiser/black-hole-flyby The ultrahigh-energy neutrino paper (open access): https://journals.aps.org/prl/abstract/10.1103/vnm4-7wdc "Close Encounters of a Primordial Kind" — the Mars wobbles paper: https://journals.aps.org/prd/abstract/10.1103/PhysRevD.110.063533 (also available open access on arXiv: https://arxiv.org/abs/2312.17217) Primordial black holes with QCD color charge (open access): https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.132.231402 Full list of press coverage: https://sites.mit.edu/mitpbh/news/ Selected press coverage: MIT News — Exotic black holes could be a byproduct of dark matter: https://news.mit.edu/2024/exotic-black-holes-could-be-dark-matter-byproduct-0606 MIT News — Mars wobble could be dark matter: https://news.mit.edu/2024/mars-wobble-could-be-dark-matter-mit-study-finds-0917 MIT News — Could a primordial black hole explain a mysteriously energetic neutrino?: https://news.mit.edu/2025/could-primordial-black-holes-last-burst-explain-mysteriously-energetic-neutrino-0918 CNN — Black holes and dark matter: https://www.cnn.com/2024/06/17/science/black-holes-dark-matter-scn/index.html LA Times — Tiny black holes zipping through the solar system: https://www.latimes.com/science/story/2024-09-17/tiny-black-holes-zipping-through-the-solar-system Scientific American — Dark matter black holes could fly through the solar system once a decade: https://www.scientificamerican.com/article/dark-matter-black-holes-could-fly-through-the-solar-system-once-a-decade/ Quanta Magazine — Monster neutrino could be a messenger of ancient black holes: https://www.quantamagazine.org/monster-neutrino-could-be-a-messenger-of-ancient-black-holes-20260123/ APS Physics — "The Solar System as a Black Hole Detector" (Mars wobbles): https://physics.aps.org/articles/v17/s98 APS Physics — "New Suspect for Neutrino Signals" (neutrino paper): https://physics.aps.org/articles/v18/s124 --- Check out our membership rewards! Visit us at Patreon.com/Whattheif Got an IF of your own? Want to have us consider your idea for a show topic? Send YOUR IF to us! Email us at feedback@whattheif.com and let us know what's in your imagination. No idea is too small, or too big! Keep On IFFin', Philip, Matt & Gaby

Teletransportación cuántica en BerlínLa fibra comercial de Berlín logró teletransportar información cuántica con 90 % de fidelidad promedio realPor Félix Riaño @LocutorCoDeutsche Telekom y Qunnect prueban teletransportación cuántica sobre fibra óptica comercial en red activa urbana.En Berlín ocurrió algo que suena a película de ciencia ficción, pero no lo es. La empresa alemana Deutsche Telekom y la empresa estadounidense Qunnect lograron teletransportar información cuántica a través de 30 kilómetros de fibra óptica comercial que ya estaba en uso.Sí, escuchaste bien. No estaban en un laboratorio aislado. Estaban usando la misma infraestructura por donde circulan llamadas, mensajes y datos de internet. Y lo hicieron con una fidelidad promedio del 90 %. En algunos momentos alcanzaron el 95 %.Pero calma. No estamos hablando de personas viajando de un punto a otro. Estamos hablando de qubits. ¿Qué significa eso realmente? ¿Y por qué importa que esto ocurra fuera de un laboratorio?La física es frágil fuera del laboratorioPrimero vamos a entender qué pasó.Un qubit es la unidad básica de información en computación cuántica. A diferencia de un bit tradicional, que solo puede ser cero o uno, el qubit puede estar en una combinación de ambos estados al mismo tiempo. Eso permite cálculos que un computador clásico tardaría miles de años en resolver.La teletransportación cuántica no mueve materia. Lo que hace es transferir el estado cuántico de una partícula a otra distante, usando un fenómeno llamado entrelazamiento cuántico. Dos partículas quedan conectadas de tal forma que lo que le pasa a una afecta a la otra, aunque estén separadas por kilómetros.En enero de 2026, en una red activa de Berlín, los investigadores teletransportaron qubits generados por una fuente coherente débil a lo largo de 30 kilómetros de fibra óptica comercial. Esto ocurrió mientras el cable transportaba tráfico normal de internet.La plataforma utilizada fue el sistema Carina de Qunnect, que genera pares de fotones entrelazados y corrige automáticamente perturbaciones ambientales.Y aquí está el detalle interesante: la teletransportación se realizó a una longitud de onda de 795 nanómetros. Esa frecuencia es útil para conectar computadores cuánticos de átomos neutros, relojes atómicos y sensores cuánticos.Ahora bien, lograr esto en la vida real no es sencillo.Las redes cuánticas son extremadamente sensibles. Vibraciones, cambios de temperatura, interferencias eléctricas, incluso el paso de un camión pesado pueden alterar la señal en la fibra. En una ciudad como Berlín, con infraestructura subterránea compleja, eso es un desafío constante.Además, la fidelidad del 90 % significa que en promedio 9 de cada 10 estados cuánticos llegaron correctamente reconstruidos. Eso es alto, pero todavía no es perfecto. Para aplicaciones comerciales masivas, se necesitará acercarse mucho más al 100 %.También hay otra duda razonable. La computación cuántica todavía no está lista para uso general. Muchos prototipos requieren enfriamiento criogénico extremo. Entonces, ¿para qué teletransportar qubits si aún no hay redes globales de computadores cuánticos funcionando?Por otro lado, la seguridad. Se dice que los computadores cuánticos del futuro podrían romper los sistemas de cifrado actuales. Algunos expertos impulsan la criptografía poscuántica. Otros creen que se necesitarán redes basadas en física cuántica, como la distribución de claves cuánticas. Pero esa tecnología tiene límites de distancia cercanos a 200 kilómetros y ha recibido críticas por posibles vulnerabilidades técnicas.Así que la pregunta es: ¿esto es un avance sólido o es todavía una demostración experimental con muchas piezas faltantes?Lo que cambia el panorama es el contexto.No fue una simulación. Fue una red comercial operativa. Eso significa que los equipos de Qunnect se instalaron en racks reales, bajo control del operador, y convivieron con tráfico digital normal.Según declaraciones del equipo, el objetivo es avanzar hacia configuraciones con múltiples nodos y mayores distancias. Eso permitiría conectar centros de datos cuánticos en distintas ubicaciones y repartir carga de cálculo.En paralelo, en Nueva York, Qunnect trabajó con Cisco Systems para demostrar intercambio de entrelazamiento en una red metropolitana de 17,6 kilómetros. Allí lograron más de 1,7 millones de pares entrelazados por hora en pruebas locales y 5.400 pares por hora sobre fibra desplegada, con fidelidad superior al 99 %.Eso valida algo importante: estas tecnologías pueden funcionar en ciudades reales, con ruido real y cables reales.¿Significa que mañana tendremos internet cuántico en casa? No. Pero sí indica que las bases técnicas están saliendo del laboratorio y entrando en infraestructura existente.El concepto de internet cuántico busca conectar computadores cuánticos, sensores y relojes atómicos para que trabajen como una red distribuida. Esto permitiría, por ejemplo, sincronización ultra precisa, redes de sensores ambientales de alta sensibilidad y comunicaciones resistentes a ciertos tipos de espionaje.En Europa existe un interés estratégico en este campo. Se habla de soberanía tecnológica y de reducir dependencia de infraestructuras externas.En Estados Unidos, varias empresas también exploran arquitecturas tipo “hub and spoke”, donde un nodo central gestiona el entrelazamiento y otros nodos funcionan con equipos a temperatura ambiente, reduciendo costos.La publicación técnica del experimento de Berlín está disponible en arXiv, lo que permite que la comunidad científica revise los datos. Eso es relevante porque no se trata solo de un comunicado corporativo.Todavía faltan componentes como memorias cuánticas estables y repetidores cuánticos para ampliar distancias de cientos a miles de kilómetros. Pero estos ensayos muestran que la fibra óptica ya desplegada puede ser parte del futuro cuántico.Berlín logró teletransportar qubits por 30 kilómetros de fibra comercial con 90 % de fidelidad. No es ciencia ficción. Es infraestructura real probando redes cuánticas.¿Confías más en criptografía matemática o en seguridad basada en física cuántica? Cuéntamelo.Sigue el pódcast en Spotify:Flash DiarioBibliografíaDeutsche TelekomLight ReadingTom's HardwareQuantum Computing ReportReutersConviértete en un supporter de este podcast: https://www.spreaker.com/podcast/flash-diario-de-el-siglo-21-es-hoy--5835407/support.Apoya el Flash Diario y escúchalo sin publicidad en el Club de Supporters. 

Welcome to Nerd Alert, a series of special episodes bridging the gap between marketing academia and practitioners. We're breaking down highly involved, complex research into plain language and takeaways any marketer can use. In this episode, Elena and Rob explore how privacy first advertising changes digital marketing. They reveal that when individual tracking disappears, platforms must rely on user groups instead. This shifts advertising toward probabilistic targeting, like how TV has always worked. Topics covered: [01:00] "Reach, Measurement, Optimization and Frequency Capping and Targeted Online Advertising Under K Anonymity"[01:45] Privacy forces less tracking, more thinking[02:50] How K Anonymity groups users by shared traits[04:35] Simulating the trade-off between privacy and performance[06:00] Privacy pushes reach-first thinking  To learn more, visit marketingarchitects.com/podcast or subscribe to our newsletter at marketingarchitects.com/newsletter. Resources: Gao, Y., & Qiao, M. (2025). Reach measurement, optimization and frequency capping in targeted online advertising under k-anonymity. arXiv preprint arXiv:2501.04882. Get more research-backed marketing strategies by subscribing to The Marketing Architects on Apple Podcasts, Spotify, or wherever you listen to podcasts. 

Emmanuel et Guillaume discutent de divers sujets liés à la programmation, notamment les systèmes de fichiers en Java, le Data Oriented Programming, les défis de JPA avec Kotlin, et les nouvelles fonctionnalités de Quarkus. Ils explorent également des sujets un peu fous comme la création de datacenters dans l'espace. Pas mal d'architecture aussi. Enregistré le 13 février 2026 Téléchargement de l'épisode LesCastCodeurs-Episode-337.mp3 ou en vidéo sur YouTube. News Langages Comment implémenter un file system en Java https://foojay.io/today/bootstrapping-a-java-file-system/ Créer un système de fichiers Java personnalisé avec NIO.2 pour des usages variés (VCS, archives, systèmes distants). Évolution Java: java.io.File (1.0) -> NIO (1.4) -> NIO.2 (1.7) pour personnalisation via FileSystem. Recommander conception préalable; API Java est orientée POSIX. Composants clés à considérer: Conception URI (scheme unique, chemin). Gestion de l'arborescence (BD, métadonnées, efficacité). Stockage binaire (emplacement, chiffrement, versions). Minimum pour démarrer (4 composants): Implémenter Path (représente fichier/répertoire). Étendre FileSystem (instance du système). Étendre FileSystemProvider (moteur, enregistré par scheme). Enregistrer FileSystemProvider via META-INF/services. Étapes suivantes: Couche BD (arborescence), opérations répertoire/fichier de base, stockage, tests. Processus long et exigeant, mais gratifiant.   Un article de brian goetz sur le futur du data oriented programming en Java https://openjdk.org/projects/amber/design-notes/beyond-records Le projet Amber de Java introduit les "carrier classes", une évolution des records qui permet plus de flexibilité tout en gardant les avantages du pattern matching et de la reconstruction Les records imposent des contraintes strictes (immutabilité, représentation exacte de l'état) qui limitent leur usage pour des classes avec état muable ou dérivé Les carrier classes permettent de déclarer une state description complète et canonique sans imposer que la représentation interne corresponde exactement à l'API publique Le modificateur "component" sur les champs permet au compilateur de dériver automatiquement les accesseurs pour les composants alignés avec la state description Les compact constructors sont généralisés aux carrier classes, générant automatiquement l'initialisation des component fields Les carrier classes supportent la déconstruction via pattern matching comme les records, rendant possible leur usage dans les instanceof et switch Les carrier interfaces permettent de définir une state description sur une interface, obligeant les implémentations à fournir les accesseurs correspondants L'extension entre carrier classes est possible, avec dérivation automatique des appels super() quand les composants parent sont subsumés par l'enfant Les records deviennent un cas particulier de carrier classes avec des contraintes supplémentaires (final, extends Record, component fields privés et finaux obligatoires) L'évolution compatible des records est améliorée en permettant l'ajout de composants en fin de liste et la déconstruction partielle par préfixe Comment éviter les pièges courants avec JPA et Kotlin - https://blog.jetbrains.com/idea/2026/01/how-to-avoid-common-pitfalls-with-jpa-and-kotlin/ JPA est une spécification Java pour la persistance objet-relationnel, mais son utilisation avec Kotlin présente des incompatibilités dues aux différences de conception des deux langages Les classes Kotlin sont finales par défaut, ce qui empêche la création de proxies par JPA pour le lazy loading et les opérations transactionnelles Le plugin kotlin-jpa génère automatiquement des constructeurs sans argument et rend les classes open, résolvant les problèmes de compatibilité Les data classes Kotlin ne sont pas adaptées aux entités JPA car elles génèrent equals/hashCode basés sur tous les champs, causant des problèmes avec les relations lazy L'utilisation de lateinit var pour les relations peut provoquer des exceptions si on accède aux propriétés avant leur initialisation par JPA Les types non-nullables Kotlin peuvent entrer en conflit avec le comportement de JPA qui initialise les entités avec des valeurs null temporaires Le backing field direct dans les getters/setters personnalisés peut contourner la logique de JPA et casser le lazy loading IntelliJ IDEA 2024.3 introduit des inspections pour détecter automatiquement ces problèmes et propose des quick-fixes L'IDE détecte les entités finales, les data classes inappropriées, les problèmes de constructeurs et l'usage incorrect de lateinit Ces nouvelles fonctionnalités aident les développeurs à éviter les bugs subtils liés à l'utilisation de JPA avec Kotlin Librairies Guide sur MapStruct @IterableMapping - https://www.baeldung.com/java-mapstruct-iterablemapping MapStruct est une bibliothèque Java pour générer automatiquement des mappers entre beans, l'annotation @IterableMapping permet de configurer finement le mapping de collections L'attribut dateFormat permet de formater automatiquement des dates lors du mapping de listes sans écrire de boucle manuelle L'attribut qualifiedByName permet de spécifier quelle méthode custom appliquer sur chaque élément de la collection à mapper Exemple d'usage : filtrer des données sensibles comme des mots de passe en mappant uniquement certains champs via une méthode dédiée L'attribut nullValueMappingStrategy permet de contrôler le comportement quand la collection source est null (retourner null ou une collection vide) L'annotation fonctionne pour tous types de collections Java (List, Set, etc.) et génère le code de boucle nécessaire Possibilité d'appliquer des formats numériques avec numberFormat pour convertir des nombres en chaînes avec un format spécifique MapStruct génère l'implémentation complète du mapper au moment de la compilation, éliminant le code boilerplate L'annotation peut être combinée avec @Named pour créer des méthodes de mapping réutilisables et nommées Le mapping des collections supporte les conversions de types complexes au-delà des simples conversions de types primitifs Accès aux fichiers Samba depuis Java avec JCIFS - https://www.baeldung.com/java-samba-jcifs JCIFS est une bibliothèque Java permettant d'accéder aux partages Samba/SMB sans monter de lecteur réseau, supportant le protocole SMB3 on pense aux galériens qui doivent se connecter aux systèmes dit legacy La configuration nécessite un contexte CIFS (CIFSContext) et des objets SmbFile pour représenter les ressources distantes L'authentification se fait via NtlmPasswordAuthenticator avec domaine, nom d'utilisateur et mot de passe La bibliothèque permet de lister les fichiers et dossiers avec listFiles() et vérifier leurs propriétés (taille, date de modification) Création de fichiers avec createNewFile() et de dossiers avec mkdir() ou mkdirs() pour créer toute une arborescence Suppression via delete() qui peut parcourir et supprimer récursivement des arborescences entières Copie de fichiers entre partages Samba avec copyTo(), mais impossibilité de copier depuis le système de fichiers local Pour copier depuis le système local, utilisation des streams SmbFileInputStream et SmbFileOutputStream Les opérations peuvent cibler différents serveurs Samba et différents partages (anonymes ou protégés par mot de passe) La bibliothèque s'intègre dans des blocs try-with-resources pour une gestion automatique des ressources Quarkus 3.31 - Support complet Java 25, nouveau packaging Maven et Panache Next - https://quarkus.io/blog/quarkus-3-31-released/ Support complet de Java 25 avec images runtime et native Nouveau packaging Maven de type quarkus avec lifecycle optimisé pour des builds plus rapides voici un article complet pour plus de detail https://quarkus.io/blog/building-large-applications/ Introduction de Panache Next, nouvelle génération avec meilleure expérience développeur et API unifiée ORM/Reactive Mise à jour vers Hibernate ORM 7.2, Reactive 3.2, Search 8.2 Support de Hibernate Spatial pour les données géospatiales Passage à Testcontainers 2 et JUnit 6 Annotations de sécurité supportées sur les repositories Jakarta Data Chiffrement des tokens OIDC pour les implémentations custom TokenStateManager Support OAuth 2.0 Pushed Authorization Requests dans l'extension OIDC Maven 3.9 maintenant requis minimum pour les projets Quarkus A2A Java SDK 1.0.0.Alpha1 - Alignement avec la spécification 1.0 du protocole Agent2Agent - https://quarkus.io/blog/a2a-java-sdk-1-0-0-alpha1/ Le SDK Java A2A implémente le protocole Agent2Agent qui permet la communication standardisée entre agents IA pour découvrir des capacités, déléguer des tâches et collaborer Passage à la version 1.0 de la spécification marque la transition d'expérimental à production-ready avec des changements cassants assumés Modernisation complète du module spec avec des Java records partout remplaçant le mix précédent de classes et records pour plus de cohérence Adoption de Protocol Buffers comme source de vérité avec des mappers MapStruct pour la conversion et Gson pour JSON-RPC Les builders utilisent maintenant des méthodes factory statiques au lieu de constructeurs publics suivant les best practices Java modernes Introduction de trois BOMs Maven pour simplifier la gestion des dépendances du SDK core, des extensions et des implémentations de référence Quarkus AgentCard évolue avec une liste supportedInterfaces remplaçant url et preferredTransport pour plus de flexibilité dans la déclaration des protocoles Support de la pagination ajouté pour ListTasks et les endpoints de configuration des notifications push avec des wrappers Result appropriés Interface A2AHttpClient pluggable permettant des implémentations HTTP personnalisées avec une implémentation Vert.x fournie Travail continu vers la conformité complète avec le TCK 1.0 en cours de développement parallèlement à la finalisation de la spécification Pourquoi Quarkus finit par "cliquer" : les 10 questions que se posent les développeurs Java - https://www.the-main-thread.com/p/quarkus-java-developers-top-questions-2025 un article qui revele et repond aux questions des gens qui ont utilisé Quarkus depuis 4-6 mois, les non noob questions Quarkus est un framework Java moderne optimisé pour le cloud qui propose des temps de démarrage ultra-rapides et une empreinte mémoire réduite Pourquoi Quarkus démarre si vite ? Le framework effectue le travail lourd au moment du build (scanning, indexation, génération de bytecode) plutôt qu'au runtime Quand utiliser le mode réactif plutôt qu'impératif ? Le réactif est pertinent pour les workloads avec haute concurrence et dominance I/O, l'impératif reste plus simple dans les autres cas Quelle est la différence entre Dev Services et Testcontainers ? Dev Services utilise Testcontainers en gérant automatiquement le cycle de vie, les ports et la configuration sans cérémonie Comment la DI de Quarkus diffère de Spring ? CDI est un standard basé sur la sécurité des types et la découverte au build-time, différent de l'approche framework de Spring Comment gérer la configuration entre environnements ? Quarkus permet de scaler depuis le développement local jusqu'à Kubernetes avec des profils, fichiers multiples et configuration externe Comment tester correctement les applications Quarkus ? @QuarkusTest démarre l'application une fois pour toute la suite de tests, changeant le modèle mental par rapport à Spring Boot Que fait vraiment Panache en coulisses ? Panache est du JPA avec des opinions fortes et des défauts propres, enveloppant Hibernate avec un style Active Record Doit-on utiliser les images natives et quand ? Les images natives brillent pour le serverless et l'edge grâce au démarrage rapide et la faible empreinte mémoire, mais tous les apps n'en bénéficient pas Comment Quarkus s'intègre avec Kubernetes ? Le framework génère automatiquement les ressources Kubernetes, gère les health checks et métriques comme s'il était nativement conçu pour cet écosystème Comment intégrer l'IA dans une application Quarkus ? LangChain4j permet d'ajouter embeddings, retrieval, guardrails et observabilité directement en Java sans passer par Python Infrastructure Les alternatives à MinIO https://rmoff.net/2026/01/14/alternatives-to-minio-for-single-node-local-s3/ MinIO a abandonné le support single-node fin 2025 pour des raisons commerciales, cassant de nombreuses démos et pipelines CI/CD qui l'utilisaient pour émuler S3 localement L'auteur cherche un remplacement simple avec image Docker, compatibilité S3, licence open source, déploiement mono-nœud facile et communauté active S3Proxy est très léger et facile à configurer, semble être l'option la plus simple mais repose sur un seul contributeur RustFS est facile à utiliser et inclut une GUI, mais c'est un projet très récent en version alpha avec une faille de sécurité majeure récente SeaweedFS existe depuis 2012 avec support S3 depuis 2018, relativement facile à configurer et dispose d'une interface web basique Zenko CloudServer remplace facilement MinIO mais la documentation et le branding (cloudserver/zenko/scality) peuvent prêter à confusion Garage nécessite une configuration complexe avec fichier TOML et conteneur d'initialisation séparé, pas un simple remplacement drop-in Apache Ozone requiert au minimum quatre nœuds pour fonctionner, beaucoup trop lourd pour un usage local simple L'auteur recommande SeaweedFS et S3Proxy comme remplaçants viables, RustFS en maybe, et élimine Garage et Ozone pour leur complexité Garage a une histoire tres associative, il vient du collectif https://deuxfleurs.fr/ qui offre un cloud distribué sans datacenter C'est certainement pas une bonne idée, les datacenters dans l'espace https://taranis.ie/datacenters-in-space-are-a-terrible-horrible-no-good-idea/ Avis d'expert (ex-NASA/Google, Dr en électronique spatiale) : Centres de données spatiaux, une "terrible" idée. Incompatibilité fondamentale : L'électronique (surtout IA/GPU) est inadaptée à l'environnement spatial. Énergie : Accès limité. Le solaire (type ISS) est insuffisant pour l'échelle de l'IA. Le nucléaire (RTG) est trop faible. Refroidissement : L'espace n'est pas "froid" ; absence de convection. Nécessite des radiateurs gigantesques (ex: 531m² pour 200kW). Radiations : Provoque erreurs (SEU, SEL) et dommages. Les GPU sont très vulnérables. Blindage lourd et inefficace. Les puces "durcies" sont très lentes. Communications : Bande passante très limitée (1Gbps radio vs 100Gbps terrestre). Le laser est tributaire des conditions atmosphériques. Conclusion : Projet extrêmement difficile, coûteux et aux performances médiocres. Data et Intelligence Artificielle Guillaume a développé un serveur MCP pour arXiv (le site de publication de papiers de recherche) en Java avec le framework Quarkus https://glaforge.dev/posts/2026/01/18/implementing-an-arxiv-mcp-server-with-quarkus-in-java/ Implémentation d'un serveur MCP (Model Context Protocol) arXiv en Java avec Quarkus. Objectif : Accéder aux publications arXiv et illustrer les fonctionnalités moins connues du protocole MCP. Mise en œuvre : Utilisation du framework Quarkus (Java) et son support MCP étendu. Assistance par Antigravity (IDE agentique) pour le développement et l'intégration de l'API arXiv. Interaction avec l'API arXiv : requêtes HTTP, format XML Atom pour les résultats, parser XML Jackson. Fonctionnalités MCP exposées : Outils (@Tool) : Recherche de publications (search_papers). Ressources (@Resource, @ResourceTemplate) : Taxonomie des catégories arXiv, métadonnées des articles (via un template d'URI). Prompts (@Prompt) : Exemples pour résumer des articles ou construire des requêtes de recherche. Configuration : Le serveur peut fonctionner en STDIO (local) ou via HTTP Streamable (local ou distant), avec une configuration simple dans des clients comme Gemini CLI. Conclusion : Quarkus simplifie la création de serveurs MCP riches en fonctionnalités, rendant les données et services "prêts pour l'IA" avec l'aide d'outils d'IA comme Antigravity. Anthropic ne mettra pas de pub dans Claude https://www.anthropic.com/news/claude-is-a-space-to-think c'est en reaction au plan non public d'OpenAi de mettre de la pub pour pousser les gens au mode payant OpenAI a besoin de cash et est probablement le plus utilisé pour gratuit au monde Anthropic annonce que Claude restera sans publicité pour préserver son rôle d'assistant conversationnel dédié au travail et à la réflexion approfondie. Les conversations avec Claude sont souvent sensibles, personnelles ou impliquent des tâches complexes d'ingénierie logicielle où les publicités seraient inappropriées. L'analyse des conversations montre qu'une part significative aborde des sujets délicats similaires à ceux évoqués avec un conseiller de confiance. Un modèle publicitaire créerait des incitations contradictoires avec le principe fondamental d'être "genuinely helpful" inscrit dans la Constitution de Claude. Les publicités introduiraient un conflit d'intérêt potentiel où les recommandations pourraient être influencées par des motivations commerciales plutôt que par l'intérêt de l'utilisateur. Le modèle économique d'Anthropic repose sur les contrats entreprise et les abonnements payants, permettant de réinvestir dans l'amélioration de Claude. Anthropic maintient l'accès gratuit avec des modèles de pointe et propose des tarifs réduits pour les ONG et l'éducation dans plus de 60 pays. Le commerce "agentique" sera supporté mais uniquement à l'initiative de l'utilisateur, jamais des annonceurs, pour préserver la confiance. Les intégrations tierces comme Figma, Asana ou Canva continueront d'être développées en gardant l'utilisateur aux commandes. Anthropic compare Claude à un cahier ou un tableau blanc : des espaces de pensée purs, sans publicité. Infinispan 16.1 est sorti https://infinispan.org/blog/2026/02/04/infinispan-16-1 déjà le nom de la release mérite une mention Le memory bounded par cache et par ensemble de cache s est pas facile à faire en Java Une nouvelle api OpenAPI AOT caché dans les images container Un serveur MCP local juste avec un fichier Java ? C'est possible avec LangChain4j et JBang https://glaforge.dev/posts/2026/02/11/zero-boilerplate-java-stdio-mcp-servers-with-langchain4j-and-jbang/ Création rapide de serveurs MCP Java sans boilerplate. MCP (Model Context Protocol): standard pour connecter les LLM à des outils et données. Le tutoriel répond au manque d'options simples pour les développeurs Java, face à une prédominance de Python/TypeScript dans l'écosystème MCP. La solution utilise: LangChain4j: qui intègre un nouveau module serveur MCP pour le protocole STDIO. JBang: permet d'exécuter des fichiers Java comme des scripts, éliminant les fichiers de build (pom.xml, Gradle). Implémentation: se fait via un seul fichier .java. JBang gère automatiquement les dépendances (//DEPS). L'annotation @Tool de LangChain4j expose les méthodes Java aux LLM. StdioMcpServerTransport gère la communication JSON-RPC via l'entrée/sortie standard (STDIO). Point crucial: Les logs doivent impérativement être redirigés vers System.err pour éviter de corrompre System.out, qui est réservé à la communication MCP (messages JSON-RPC). Facilite l'intégration locale avec des outils comme Gemini CLI, Claude Code, etc. Reciprocal Rank Fusion : un algorithme utile et souvent utilisé pour faire de la recherche hybride, pour mélanger du RAG et des recherches par mots-clé https://glaforge.dev/posts/2026/02/10/advanced-rag-understanding-reciprocal-rank-fusion-in-hybrid-search/ RAG : Qualité LLM dépend de la récupération. Recherche Hybride : Combiner vectoriel et mots-clés (BM25) est optimal. Défi : Fusionner des scores d'échelles différentes. Solution : Reciprocal Rank Fusion (RRF). RRF : Algorithme robuste qui fusionne des listes de résultats en se basant uniquement sur le rang des documents, ignorant les scores. Avantages RRF : Pas de normalisation de scores, scalable, excellente première étape de réorganisation. Architecture RAG fréquente : RRF (large sélection) + Cross-Encoder / modèle de reranking (précision fine). RAG-Fusion : Utilise un LLM pour générer plusieurs variantes de requête, puis RRF agrège tous les résultats pour renforcer le consensus et réduire les hallucinations. Implémentation : LangChain4j utilise RRF par défaut pour agréger les résultats de plusieurs retrievers. Les dernières fonctionnalités de Gemini et Nano Banana supportées dans LangChain4j https://glaforge.dev/posts/2026/02/06/latest-gemini-and-nano-banana-enhancements-in-langchain4j/ Nouveaux modèles d'images Nano Banana (Gemini 2.5/3.0) pour génération et édition (jusqu'à 4K). "Grounding" via Google Search (pour images et texte) et Google Maps (localisation, Gemini 2.5). Outil de contexte URL (Gemini 3.0) pour lecture directe de pages web. Agents multimodaux (AiServices) capables de générer des images. Configuration de la réflexion (profondeur Chain-of-Thought) pour Gemini 3.0. Métadonnées enrichies : usage des tokens et détails des sources de "grounding". Comment configurer Gemini CLI comment agent de code dans IntelliJ grâce au protocole ACP https://glaforge.dev/posts/2026/02/01/how-to-integrate-gemini-cli-with-intellij-idea-using-acp/ But : Intégrer Gemini CLI à IntelliJ IDEA via l'Agent Client Protocol (ACP). Prérequis : IntelliJ IDEA 2025.3+, Node.js (v20+), Gemini CLI. Étapes : Installer Gemini CLI (npm install -g @google/gemini-cli). Localiser l'exécutable gemini. Configurer ~/.jetbrains/acp.json (chemin exécutable, --experimental-acp, use_idea_mcp: true). Redémarrer IDEA, sélectionner "Gemini CLI" dans l'Assistant IA. Usage : Gemini interagit avec le code et exécute des commandes (contexte projet). Important : S'assurer du flag --experimental-acp dans la configuration. Outillage PipeNet, une alternative (open source aussi) à LocalTunnel, mais un plus évoluée https://pipenet.dev/ pipenet: Alternative open-source et moderne à localtunnel (client + serveur). Usages: Développement local (partage, webhooks), intégration SDK, auto-hébergement sécurisé. Fonctionnalités: Client (expose ports locaux, sous-domaines), Serveur (déploiement, domaines personnalisés, optimisé cloud mono-port). Avantages vs localtunnel: Déploiement cloud sur un seul port, support multi-domaines, TypeScript/ESM, maintenance active. Protocoles: HTTP/S, WebSocket, SSE, HTTP Streaming. Intégration: CLI ou SDK JavaScript. JSON-IO — une librairie comme Jackson ou GSON, supportant JSON5, TOON, et qui pourrait être utile pour l'utilisation du "structured output" des LLMs quand ils ne produisent pas du JSON parfait https://github.com/jdereg/json-io json-io : Librairie Java pour la sérialisation et désérialisation JSON/TOON. Gère les graphes d'objets complexes, les références cycliques et les types polymorphes. Support complet JSON5 (lecture et écriture), y compris des fonctionnalités non prises en charge par Jackson/Gson. Format TOON : Notation orientée token, optimisée pour les LLM, réduisant l'utilisation de tokens de 40 à 50% par rapport au JSON. Légère : Aucune dépendance externe (sauf java-util), taille de JAR réduite (~330K). Compatible JDK 1.8 à 24, ainsi qu'avec les environnements JPMS et OSGi. Deux modes de conversion : vers des objets Java typés (toJava()) ou vers des Map (toMaps()). Options de configuration étendues via ReadOptionsBuilder et WriteOptionsBuilder. Optimisée pour les déploiements cloud natifs et les architectures de microservices. Utiliser mailpit et testcontainer pour tester vos envois d'emails https://foojay.io/today/testing-emails-with-testcontainers-and-mailpit/ l'article montre via SpringBoot et sans. Et voici l'extension Quarkus https://quarkus.io/extensions/io.quarkiverse.mailpit/quarkus-mailpit/?tab=docs Tester l'envoi d'emails en développement est complexe car on ne peut pas utiliser de vrais serveurs SMTP Mailpit est un serveur SMTP de test qui capture les emails et propose une interface web pour les consulter Testcontainers permet de démarrer Mailpit dans un conteneur Docker pour les tests d'intégration L'article montre comment configurer une application SpringBoot pour envoyer des emails via JavaMail Un module Testcontainers dédié à Mailpit facilite son intégration dans les tests Le conteneur Mailpit expose un port SMTP (1025) et une API HTTP (8025) pour vérifier les emails reçus Les tests peuvent interroger l'API HTTP de Mailpit pour valider le contenu des emails envoyés Cette approche évite d'utiliser des mocks et teste réellement l'envoi d'emails Mailpit peut aussi servir en développement local pour visualiser les emails sans les envoyer réellement La solution fonctionne avec n'importe quel framework Java supportant JavaMail Architecture Comment scaler un système de 0 à 10 millions d'utilisateurs https://blog.algomaster.io/p/scaling-a-system-from-0-to-10-million-users Philosophie : Scalabilité incrémentale, résoudre les goulots d'étranglement sans sur-ingénierie. 0-100 utilisateurs : Serveur unique (app, DB, jobs). 100-1K : Séparer app et DB (services gérés, pooling). 1K-10K : Équilibreur de charge, multi-serveurs d'app (stateless via sessions partagées). 10K-100K : Caching, réplicas de lecture DB, CDN (réduire charge DB). 100K-500K : Auto-scaling, applications stateless (authentification JWT). 500K-10M : Sharding DB, microservices, files de messages (traitement asynchrone). 10M+ : Déploiement multi-régions, CQRS, persistance polyglotte, infra personnalisée. Principes clés : Simplicité, mesure, stateless essentiel, cache/asynchrone, sharding prudent, compromis (CAP), coût de la complexité. Patterns d'Architecture 2026 - Du Hype à la Réalité du Terrain (Part 1/2) - https://blog.ippon.fr/2026/01/30/patterns-darchitecture-2026-part-1/ L'article présente quatre patterns d'architecture logicielle pour répondre aux enjeux de scalabilité, résilience et agilité business dans les systèmes modernes Il présentent leurs raisons et leurs pièges Un bon rappel L'Event-Driven Architecture permet une communication asynchrone entre systèmes via des événements publiés et consommés, évitant le couplage direct Les bénéfices de l'EDA incluent la scalabilité indépendante des composants, la résilience face aux pannes et l'ajout facile de nouveaux cas d'usage Le pattern API-First associé à un API Gateway centralise la sécurité, le routage et l'observabilité des APIs avec un catalogue unifié Le Backend for Frontend crée des APIs spécifiques par canal (mobile, web, partenaires) pour optimiser l'expérience utilisateur CQRS sépare les modèles de lecture et d'écriture avec des bases optimisées distinctes, tandis que l'Event Sourcing stocke tous les événements plutôt que l'état actuel Le Saga Pattern gère les transactions distribuées via orchestration centralisée ou chorégraphie événementielle pour coordonner plusieurs microservices Les pièges courants incluent l'explosion d'événements granulaires, la complexité du debugging distribué, et la mauvaise gestion de la cohérence finale Les technologies phares sont Kafka pour l'event streaming, Kong pour l'API Gateway, EventStoreDB pour l'Event Sourcing et Temporal pour les Sagas Ces patterns nécessitent une maturité technique et ne sont pas adaptés aux applications CRUD simples ou aux équipes junior Patterns d'architecture 2026 : du hype à la réalité terrain part. 2 - https://blog.ippon.fr/2026/02/04/patterns-darchitecture-2026-part-2/ Deuxième partie d'un guide pratique sur les patterns d'architecture logicielle et système éprouvés pour moderniser et structurer les applications en 2026 Strangler Fig permet de migrer progressivement un système legacy en l'enveloppant petit à petit plutôt que de tout réécrire d'un coup (70% d'échec pour les big bang) Anti-Corruption Layer protège votre nouveau domaine métier des modèles externes et legacy en créant une couche de traduction entre les systèmes Service Mesh gère automatiquement la communication inter-services dans les architectures microservices (sécurité mTLS, observabilité, résilience) Architecture Hexagonale sépare le coeur métier des détails techniques via des ports et adaptateurs pour améliorer la testabilité et l'évolutivité Chaque pattern est illustré par un cas client concret avec résultats mesurables et liste des pièges à éviter lors de l'implémentation Les technologies 2026 mentionnées incluent Istio, Linkerd pour service mesh, LaunchDarkly pour feature flags, NGINX et Kong pour API gateway Tableau comparatif final aide à choisir le bon pattern selon la complexité, le scope et le use case spécifique du projet L'article insiste sur une approche pragmatique : ne pas utiliser un pattern juste parce qu'il est moderne mais parce qu'il résout un problème réel Pour les systèmes simples type CRUD ou avec peu de services, ces patterns peuvent introduire une complexité inutile qu'il faut savoir éviter Méthodologies Le rêve récurrent de remplacer voire supprimer les développeurs https://www.caimito.net/en/blog/2025/12/07/the-recurring-dream-of-replacing-developers.html Depuis 1969, chaque décennie voit une tentative de réduire le besoin de développeurs (de COBOL, UML, visual builders… à IA). Motivation : frustration des dirigeants face aux délais et coûts de développement. La complexité logicielle est intrinsèque et intellectuelle, non pas une question d'outils. Chaque vague technologique apporte de la valeur mais ne supprime pas l'expertise humaine. L'IA assiste les développeurs, améliore l'efficacité, mais ne remplace ni le jugement ni la gestion de la complexité. La demande de logiciels excède l'offre car la contrainte majeure est la réflexion nécessaire pour gérer cette complexité. Pour les dirigeants : les outils rendent-ils nos développeurs plus efficaces sur les problèmes complexes et réduisent-ils les tâches répétitives ? Le "rêve" de remplacer les développeurs, irréalisable, est un moteur d'innovation créant des outils précieux. Comment creuser des sujets à l'ère de l'IA générative. Quid du partage et la curation de ces recherches ? https://glaforge.dev/posts/2026/02/04/researching-topics-in-the-age-of-ai-rock-solid-webhooks-case-study/ Recherche initiale de l'auteur sur les webhooks en 2019, processus long et manuel. L'IA (Deep Research, Gemini, NotebookLM) facilite désormais la recherche approfondie, l'exploration de sujets et le partage des résultats. L'IA a identifié et validé des pratiques clés pour des déploiements de webhooks résilients, en grande partie les mêmes que celles trouvées précédemment par l'auteur. Génération d'artefacts par l'IA : rapport détaillé, résumé concis, illustration sketchnote, et même une présentation (slide deck). Guillaume s'interroge sur le partage public de ces rapports de recherche générés par l'IA, tout en souhaitant éviter le "AI Slop". Loi, société et organisation Le logiciel menacé par le vibe coding https://www.techbuzz.ai/articles/we-built-a-monday-com-clone-in-under-an-hour-with-ai Deux journalistes de CNBC sans expérience de code ont créé un clone fonctionnel de Monday.com en moins de 60 minutes pour 5 à 15 dollars. L'expérience valide les craintes des investisseurs qui ont provoqué une baisse de 30% des actions des entreprises SaaS. L'IA a non seulement reproduit les fonctionnalités de base mais a aussi recherché Monday.com de manière autonome pour identifier et recréer ses fonctionnalités clés. Cette technique appelée "vibe-coding" permet aux non-développeurs de construire des applications via des instructions en anglais courant. Les entreprises les plus vulnérables sont celles offrant des outils "qui se posent sur le travail" comme Atlassian, Adobe, HubSpot, Zendesk et Smartsheet. Les entreprises de cybersécurité comme CrowdStrike et Palo Alto sont considérées plus protégées grâce aux effets de réseau et aux barrières réglementaires. Les systèmes d'enregistrement comme Salesforce restent plus difficiles à répliquer en raison de leur profondeur d'intégration et de données d'entreprise. Le coût de 5 à 15 dollars par construction permet aux entreprises de prototyper plusieurs solutions personnalisées pour moins cher qu'une seule licence Monday.com. L'expérience soulève des questions sur la pérennité du marché de 5 milliards de dollars des outils de gestion de projet face à l'IA générative. Conférences En complément de l'agenda des conférences de Aurélie Vache, il y a également le site https://javaconferences.org/ (fait par Brian Vermeer) avec toutes les conférences Java à venir ! La liste des conférences provenant de Developers Conferences Agenda/List par Aurélie Vache et contributeurs : 12-13 février 2026 : Touraine Tech #26 - Tours (France) 12-13 février 2026 : World Artificial Intelligence Cannes Festival - Cannes (France) 19 février 2026 : ObservabilityCON on the Road - Paris (France) 6 mars 2026 : WordCamp Nice 2026 - Nice (France) 18 mars 2026 : Jupyter Workshops: AI in Jupyter: Building Extensible AI Capabilities for Interactive Computing - Saint-Maur-des-Fossés (France) 18-19 mars 2026 : Agile Niort 2026 - Niort (France) 20 mars 2026 : Atlantique Day 2026 - Nantes (France) 26 mars 2026 : Data Days Lille - Lille (France) 26-27 mars 2026 : SymfonyLive Paris 2026 - Paris (France) 26-27 mars 2026 : REACT PARIS - Paris (France) 27-29 mars 2026 : Shift - Nantes (France) 31 mars 2026 : ParisTestConf - Paris (France) 31 mars 2026-1 avril 2026 : FlowCon France 2026 - Paris (France) 1 avril 2026 : AWS Summit Paris - Paris (France) 2 avril 2026 : Pragma Cannes 2026 - Cannes (France) 2-3 avril 2026 : Xen Spring Meetup 2026 - Grenoble (France) 7 avril 2026 : PyTorch Conference Europe - Paris (France) 9-10 avril 2026 : Android Makers by droidcon 2026 - Paris (France) 9-11 avril 2026 : Drupalcamp Grenoble 2026 - Grenoble (France) 16-17 avril 2026 : MiXiT 2026 - Lyon (France) 17-18 avril 2026 : Faiseuses du Web 5 - Dinan (France) 22-24 avril 2026 : Devoxx France 2026 - Paris (France) 23-25 avril 2026 : Devoxx Greece - Athens (Greece) 6-7 mai 2026 : Devoxx UK 2026 - London (UK) 12 mai 2026 : Lead Innovation Day - Leadership Edition - Paris (France) 19 mai 2026 : La Product Conf Paris 2026 - Paris (France) 21-22 mai 2026 : Flupa UX Days 2026 - Paris (France) 22 mai 2026 : AFUP Day 2026 Lille - Lille (France) 22 mai 2026 : AFUP Day 2026 Paris - Paris (France) 22 mai 2026 : AFUP Day 2026 Bordeaux - Bordeaux (France) 22 mai 2026 : AFUP Day 2026 Lyon - Lyon (France) 28 mai 2026 : DevCon 27 : I.A. & Vibe Coding - Paris (France) 28 mai 2026 : Cloud Toulouse 2026 - Toulouse (France) 29 mai 2026 : NG Baguette Conf 2026 - Paris (France) 29 mai 2026 : Agile Tour Strasbourg 2026 - Strasbourg (France) 2-3 juin 2026 : Agile Tour Rennes 2026 - Rennes (France) 2-3 juin 2026 : OW2Con - Paris-Châtillon (France) 3 juin 2026 : IA–NA - La Rochelle (France) 5 juin 2026 : TechReady - Nantes (France) 5 juin 2026 : Fork it! - Rouen - Rouen (France) 6 juin 2026 : Polycloud - Montpellier (France) 9 juin 2026 : JFTL - Montrouge (France) 9 juin 2026 : C: - Caen (France) 11-12 juin 2026 : DevQuest Niort - Niort (France) 11-12 juin 2026 : DevLille 2026 - Lille (France) 12 juin 2026 : Tech F'Est 2026 - Nancy (France) 16 juin 2026 : Mobilis In Mobile 2026 - Nantes (France) 17-19 juin 2026 : Devoxx Poland - Krakow (Poland) 17-20 juin 2026 : VivaTech - Paris (France) 18 juin 2026 : Tech'Work - Lyon (France) 22-26 juin 2026 : Galaxy Community Conference - Clermont-Ferrand (France) 24-25 juin 2026 : Agi'Lille 2026 - Lille (France) 24-26 juin 2026 : BreizhCamp 2026 - Rennes (France) 2 juillet 2026 : Azur Tech Summer 2026 - Valbonne (France) 2-3 juillet 2026 : Sunny Tech - Montpellier (France) 3 juillet 2026 : Agile Lyon 2026 - Lyon (France) 6-8 juillet 2026 : Riviera Dev - Sophia Antipolis (France) 2 août 2026 : 4th Tech Summit on Artificial Intelligence & Robotics - Paris (France) 20-22 août 2026 : 4th Tech Summit on AI & Robotics - Paris (France) & Online 4 septembre 2026 : JUG Summer Camp 2026 - La Rochelle (France) 17-18 septembre 2026 : API Platform Conference 2026 - Lille (France) 24 septembre 2026 : PlatformCon Live Day Paris 2026 - Paris (France) 1 octobre 2026 : WAX 2026 - Marseille (France) 1-2 octobre 2026 : Volcamp - Clermont-Ferrand (France) 5-9 octobre 2026 : Devoxx Belgium - Antwerp (Belgium) Nous contacter Pour réagir à cet épisode, venez discuter sur le groupe Google https://groups.google.com/group/lescastcodeurs Contactez-nous via X/twitter https://twitter.com/lescastcodeurs ou Bluesky https://bsky.app/profile/lescastcodeurs.com Faire un crowdcast ou une crowdquestion Soutenez Les Cast Codeurs sur Patreon https://www.patreon.com/LesCastCodeurs Tous les épisodes et toutes les infos sur https://lescastcodeurs.com/

El descubrimiento de algunos de los planetas más esquivos del sistema solar, Neptuno, Plutón y quién sabe, una super Tierra en los confines del sistema solar, ha estado marcado por historias de personas que usan las teorías como su única guía y observadores que a veces tienen muchísima paciencia y afortunadamente mucha suerte. En esta conversación hablamos sobre los "planetas X", aquellos cuerpos que en algún momento fueron solo una especulación teórica y o bien fueron posteriormente descubiertos o están pendientes de serlo. Naturalmente, para las personas más curiosas, profundizamos en el que se conoce comúnmente como el Planeta 9 (o el nuevo Planeta X para otras personas).Realizado y presentado por: Antonio Bernal (Observatorio Fabra, Barcelona) y Jorge I. Zuluaga (Universidad de Antioquia, Medellín-Colombia)Producido por: Jorge I. Zuluaga (Universidad de Antioquia)Lista de episodios organizados por tema: ⁠⁠⁠⁠⁠⁠⁠⁠https://tinyurl.com/y58xkvcn⁠⁠⁠⁠⁠⁠⁠⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Lista de reproducción temática: ⁠⁠⁠⁠⁠⁠⁠⁠https://tinyurl.com/89s2fpm8⁠⁠⁠⁠⁠⁠⁠⁠ Enlaces de interés: Artículo sobre la interpretación del planeta 9 como un agujero negro (revista original, repositorio de arXiv)

El Cometa 41P frenó, se detuvo y cambió sentido tras acercarse al Sol en 2017Por Félix Riaño @LocutorCoUn análisis de imágenes del telescopio Hubble reveló que el cometa 41P invirtió su rotación después del perihelio en 2017En 2017, un cometa de apenas medio kilómetro de diámetro hizo algo que parece imposible. El cometa 41P/Tuttle–Giacobini–Kresák, conocido como 41P, redujo su velocidad de giro de forma drástica cuando se acercó al Sol. Pasó de rotar cada 20 horas a hacerlo cada 53 horas en pocas semanas. Luego desapareció de nuestra vista porque quedó demasiado cerca del Sol desde la perspectiva de la Tierra. Meses después, cuando volvió a ser visible, los datos mostraron algo aún más sorprendente: estaba girando en sentido contrario y completaba una vuelta cada 14,4 horas. ¿Cómo puede un objeto espacial frenar hasta detenerse y luego empezar a girar al revés?¿Y si el cometa se frenó hasta cero?Planteamiento descriptivoPara entender esta historia vamos a empezar por lo básico. Un cometa es una mezcla de roca y hielo que quedó como residuo del nacimiento del Sistema Solar hace unos 4.600 millones de años. Muchos de estos cuerpos viajan en órbitas muy alargadas. El cometa 41P pertenece a la llamada “familia de Júpiter”. Eso significa que su órbita está muy influenciada por la gravedad de Júpiter y que tarda menos de 20 años en dar una vuelta alrededor del Sol. En su caso, tarda unos 5,4 años.En abril de 2017, 41P llegó a su perihelio, que es el punto de la órbita más cercano al Sol. Allí recibió más calor. El hielo comenzó a transformarse en gas, un proceso llamado sublimación. Ese gas salió disparado al espacio y formó una nube llamada coma. También aparecieron chorros, como si fueran pequeñas mangueras naturales que empujan el cometa.Aquí aparece el detalle que lo cambia todo. Cuando esos chorros de gas salen de forma desigual, actúan como pequeños motores. Si el gas sale más fuerte de un lado que de otro, el cometa recibe un empujón que altera su rotación. Es parecido a lo que ocurre cuando soplas una rueda con aire de un solo lado.Observaciones del telescopio espacial Swift de la NASA mostraron que entre marzo y mayo de 2017 el periodo de rotación pasó de 20 horas a 46 y luego hasta 53 horas. Normalmente los cambios en cometas se miden en minutos. Aquí hablamos de decenas de horas. Es un cambio enorme para un objeto tan pequeño, de unos 500 metros de diámetro.Después de mayo, el cometa quedó oculto por el brillo solar. No había datos. Cuando regresó en diciembre, el telescopio espacial Hubble tomó nuevas imágenes. El astrónomo David Jewitt, de la Universidad de California en Los Ángeles, analizó esas fotos años después. Encontró que el periodo era ahora de 14,4 horas. Eso significa que estaba girando más rápido que antes y en sentido contrario. Para que eso ocurra, tuvo que frenar completamente, llegar a cero y empezar a girar al revés. Los cálculos indican que ese punto de detención pudo ocurrir en junio de 2017.Este hallazgo ayuda a responder una pregunta más grande. ¿Por qué hay menos cometas pequeños de los que esperamos encontrar? Una posible explicación es que muchos se destruyen por su propia rotación. Si los chorros los aceleran demasiado, la fuerza centrífuga puede romperlos.El estudio también estimó que el núcleo de 41P mide cerca de 500 metros. Es pequeño en términos astronómicos. Su tamaño lo hace más vulnerable a cambios rápidos. Además, la fracción activa de su superficie, es decir, la parte que libera gas, bajó de aproximadamente 2,4 en 2001 a cerca de 0,14 en 2017. Eso sugiere que su superficie ha cambiado con el tiempo, tal vez formando una capa más dura que bloquea parte del hielo.Aquí hay otro dato inquietante. Aunque su órbita podría mantenerse estable durante unos 10.000 años, el análisis indica que podría romperse por inestabilidad rotacional en unos 25 años si continúa este comportamiento. Eso es muy poco tiempo en términos cósmicos.El cometa volverá a acercarse al Sol en 2028. Los astrónomos estarán atentos. Nuevos telescopios como el Observatorio Vera C. Rubin en Chile van a observar miles de objetos pequeños. Vamos a tener muchos más datos para entender si 41P es un caso raro o si este fenómeno es más común de lo que pensábamos.El cometa 41P fue observado por primera vez en 1858 por Horace Tuttle. Más tarde fue redescubierto por Michel Giacobini en 1907 y por Ľubor Kresák en 1951. Por eso lleva tres apellidos. Su órbita lo lleva desde un punto un poco más cercano al Sol que la órbita de la Tierra hasta más allá de Júpiter.En 1973 tuvo un gran estallido de brillo y alcanzó magnitud 4, lo que significa que pudo verse a simple vista en cielos oscuros. En 2017 volvió a llamar la atención por razones muy distintas.El estudio de David Jewitt fue publicado como prepublicación en arXiv y aceptado en la revista The Astronomical Journal. Otros astrónomos, como Dennis Bodewits y Jane Luu, han señalado que aunque se habían propuesto modelos teóricos donde un cometa podía invertir su rotación, nunca se había observado algo así con datos tan claros.Este caso también nos enseña que los cometas no son rocas inertes. Son cuerpos dinámicos. Cambian, pierden masa, se fracturan y evolucionan cada vez que se acercan al Sol. Son como fósiles activos del origen del Sistema Solar.Resumen final y recomendaciónEl cometa 41P redujo su giro, se detuvo y comenzó a rotar al revés en 2017. Chorros de gas actuaron como motores naturales que cambiaron su velocidad. Podría romperse en pocas décadas. Vamos a seguir atentos a su regreso en 2028.Cuéntame qué te parece esta historia y sigue el pódcast en Spotify: Flash DiarioBibliografíaPhys.org The New York Times IFLScience arXiv – David Jewitt (2026) DOI: 10.48550/arxiv.2602.06403Conviértete en un supporter de este podcast: https://www.spreaker.com/podcast/flash-diario-de-el-siglo-21-es-hoy--5835407/support.Apoya el Flash Diario y escúchalo sin publicidad en el Club de Supporters. 

Hey, Alex from W&B here

Author's note: this is somewhat more rushed than ideal, but I think getting this out sooner is pretty important. Ideally, it would be a bit less snarky. Anthropic[1] recently published a new piece of research: The Hot Mess of AI: How Does Misalignment Scale with Model Intelligence and Task Complexity? (arXiv, Twitter thread). I have some complaints about both the paper and the accompanying blog post. tl;dr The paper's abstract says that "in several settings, larger, more capable models are more incoherent than smaller models", but in most settings they are more coherent. This emphasis is even more exaggerated in the blog post and Twitter thread. I think this is pretty misleading. The paper's technical definition of "incoherence" is uninteresting[2] and the framing of the paper, blog post, and Twitter thread equivocate with the more normal English-language definition of the term, which is extremely misleading. Section 5 of the paper (and to a larger extent the blog post and Twitter) attempt to draw conclusions about future alignment difficulties that are unjustified by the experiment results, and would be unjustified even if the experiment results pointed in the other direction. The blog post is substantially LLM-written. I think this [...] ---Outline:(00:39) tl;dr(01:42) Paper(06:25) Blog The original text contained 3 footnotes which were omitted from this narration. --- First published: February 4th, 2026 Source: https://www.lesswrong.com/posts/ceEgAEXcL7cC2Ddiy/anthropic-s-hot-mess-paper-overstates-its-case-and-the-blog --- Narrated by TYPE III AUDIO. ---Images from the article:Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

Evénéments Q2B Santa Clara de décembre 2025 Les vidéos de cette conférence sont maintenant https://www.youtube.com/results?search_query=Q2B+Santa+Clara+2025 CES 2026Une journée organisée sur les technologies quantiques  Quantum beyond the hype »  https://www.youtube.com/watch?v=QRMghWl-h6ohttps://www.youtube.com/watch?v=CJ8VVNaCa6cet une autre sur les capteurs quantiques où intervenait notamment Philippe Bouyer (Quantum Delta aux Pays-Bas). Conférence Qubits2026 de D-WaveElle avait lieu la dernière semaine de janvier https://qubits2026.dwavequantum.com/agenda. A voir le keynote d'Alan Baratz, leur CEO, avec une intervention de Rob Schoelkopf, le cofondateur de QCI https://www.youtube.com/watch?v=AKg11XTaWaw Sophia AntipolisTrois panels à la suite traitant des technologies quantiques, des questions de souveraineté et sociétales.  Du quantique à Davos  Le Président de la République emmenait avec lui des startups deeptechs françaises, dont Quandela et Quobly.Les Suisses organisaient aussi un séminaire quantique qui était assez technique au début https://www.youtube.com/watch?v=jbHO-jgWKso Webinar QEI https://www.talentq.es/es_es/2026/01/27/seminar-olivier-ezratty-yasser-omar/ Yasser Omar et olivier  le 3 février, duo sur l'énergétique du calcul quantique. Et ouverture des inscriptions et propositions de contributions pour le QEI Workshop qui a lieu du 18 au 22 mai à Barcelone.https://quantum-energy-initiative.org/2026/registration-open-for-qei26/ Actus France QuoblySealsq annonce une négociation en cours avec Quobly pour investissement de $200M qui leur permettrait à terme de prendre le contrôle à 100% de la startup. Ils prévoient cela en deux étapes avec une participation d'abord minoritaire, pouvant ensuite devenir majoritaire.  Articles sur D-WaveDeux articles publiés par Philippe Lacomme et Caroline Prodhon, des chercheurs œuvrant dans le domaine de la résolution de problèmes de recherche opérationnelle. Ils analysent la capacité des quantum annealers de D-Wave à résoudre ce genre de problèmes. Assessing Quantum Annealing to Solve the Minimum Vertex Multicut by Ali Abbassi, Yann Dujardin, Eric Gourdin, Philippe Lacomme, and Caroline Prodhon, arXiv, January 2026 (6 pages).Quantum Approaches to the Minimum Edge Multiway Cut Problem by Ali Abbassi, Yann Dujardin, Eric Gourdin, Philippe Lacomme, and Caroline Prodhon, arXiv, January 2026 (10 pages). Article de TotalEnergiesTotalEnergies, Jeremie Messud et Wassil Sennane de Quantsoc, société de conseil en développement quantique créée en mars 2025 publiaient un arXiv sur la manière d'optimiser des circuits de calcul de type QPE.  On the robustness of Quantum Phase Estimation to compute ground properties of many-electron systems by Wassil Sennane, and Jérémie Messud, arXiv, January 2026 (22 pages). ColibriTD Publie un papier sur la résolution d'équation différentielle non linéaire sur IBM Heron. Solving nonlinear differential equations on noisy 156-qubit quantum computers by Karla Baumann, Youcef Modheb, Roman Randrianarisoa, Roland Katz, Aoife Boyle, and Frédéric Holweck, arXiv, January 2026 (12 pages). Actus Internationales D-Wave fait l'acquisition de Quantum Circuits IncLe 7 janvier 2026 l'acquisition de Quantum Circuits Inc, une startup issue de Yale spécialisée dans la conception de processeurs supraconducteurs utilisant la technique du dual-rail. Le tout pour $550M dont $300M d'actions D-Wave et $250M.  IonQ fait l'acquisition de Skywater IonQ continue son marché avec une grosse acquisition à $1.8B. Ils achètent une fab qui fait du silicium (en wafers 200 mm), de la photonique et aussi des composants supraconducteurs comme les puces de D-Wave. https://investors.ionq.com/news/news-details/2026/IonQ-to-Acquire-SkyWater-Technology-Creating-the-Only-Vertically-Integrated-Full-Stack-Quantum-Platform-Company/default.aspx On parle aussi des puces CMOS dont a besoin IonQ / Oxford Ionics, et aussi des composants supraconducteurs d'IDQ. + d'une boite de logiciels dans l'IA.https://investors.ionq.com/news/news-details/2026/IonQ-to-Acquire-AI-Software-and-Technology--RD-Specialist--Seed-Innovations-/default.aspx Google simulation Un nouveau papier d'avantage quantique, toujours sur des simulations physiques sur des machines NISQ Willow.  Evidence for a two-dimensional quantum glass state at high temperatures by Aleksey Lunkin, Nicole S. Ticea et al, arXiv, January 2026 (21 pages). Google. Fidélités d'IBM MiamiLa première incarnation du processeur Night Hawk est en ligne dans le cloud d'IBM et ont peut en consulter les figures de mérite. Microsoft Majorana Un papier des détracteurs de Microsoft emmenés par le fameux Sergei Frolov explique comment Microsoft donne ses résultats. https://scitechdaily.com/scientists-say-a-major-q...

# Astronomy Daily - S05E22## Monday, January 26, 2026Welcome to Astronomy Daily! Join hosts Anna and Avery as they explore the latest developments in space and astronomy, from ambitious plans to terraform Mars to stunning new views of dying stars.### Episode Highlights**Mars Terraforming Gets Serious**Scientists unveil a comprehensive blueprint for transforming Mars into a habitable world. Discover the three-phase plan using Martian resources, engineered nanoparticles, and hardy microorganisms that could warm the Red Planet by 30°C and eventually create breathable air. But should we terraform Mars at all?**Harvesting Water from Mars' Atmosphere**While underground ice remains the primary water source for future Mars missions, researchers reveal how atmospheric moisture could provide a crucial backup. Learn about the innovative technologies that could make Mars settlements more self-sufficient.**Chandra's Cosmic Catalog Milestone**NASA's Chandra X-ray Observatory has now cataloged over 1.3 million X-ray detections across the sky. We explore this treasure trove of data spanning 22 years of observations, including a stunning view of the Galactic Center with over 3,300 sources in just 60 light-years.**Earthquake Sensors Track Space Debris**Ingenious new research shows how seismic monitoring networks can track dangerous falling satellites in near real-time. Discover how scientists reconstructed the trajectory and breakup of China's Shenzhou-15 module using earthquake sensors.**Water Worlds or Lava Planets?**Shocking new findings suggest 98% of planets we thought were ocean-bearing "hycean worlds" might actually be molten rock. Learn about the Solidification Shoreline model that's rewriting our understanding of sub-Neptune exoplanets.**Webb Captures a Dying Star's Beauty**The James Webb Space Telescope reveals the Helix Nebula in unprecedented detail, showing us the eventual fate of our own Sun. Witness stellar recycling in action as a dying star distributes the building blocks of future worlds.### Links & Resources- Research on Mars terraforming strategies- Advances in Space Research journal study on atmospheric water harvesting- Chandra Source Catalog: cxc.cfa.harvard.edu/csc/- Science journal publication on seismic debris tracking- arXiv preprint on sub-Neptune exoplanet composition- Webb Space Telescope Helix Nebula observationsFor more space news and daily episodes, visit astronomydaily.ioFollow us on social media @AstroDailyPod---*Astronomy Daily: Your daily dose of space and astronomy news*Become a supporter of this podcast: https://www.spreaker.com/podcast/astronomy-daily-space-news-updates--5648921/support.Sponsor Details:Ensure your online privacy by using NordVPN. To get our special listener deal and save a lot of money, visit www.bitesz.com/nordvpn. You'll be glad you did!Become a supporter of Astronomy Daily by joining our Supporters Club. Commercial free episodes daily are only a click way... Click HereThis episode includes AI-generated content.

AI maakt ons minder creatief. Dat blijkt uit onderzoek onder 460 innovators aan de Universiteit van Toronto. Hun ideeën? Oppervlakkiger, minder divers. Een broek als vogelverschrikker in plaats van broekzakken als voersysteem. Maar er is een uitweg. In deze aflevering laat ik zien hoe je die trend omdraait — met vijf krachtige prompt-technieken die wél werken. Van beperkingen toevoegen tot extreme cases gebruiken: ik neem je mee in de methode waarmee ik workshops geef bij het Erasmus MC en grote organisaties. Dit is geen theorie. Dit zijn concrete tools die je morgen al kunt inzetten om innovatie te versnellen. (01:00:15) De belofte van artificiële innovatie (01:03:19) De AI-paradox: Een rem op de menselijke geest (01:04:31) Strategie I: De kunst van de open vraag (01:05:39) Strategie II: Heilige huisjes omverwerpen (01:06:46) Strategie III: Innovatie onder hoogspanning (01:07:54) Strategie IV: Van angst naar Barbie-dystopie (01:09:26) De chatbot als architect van de toekomst (01:10:55) De artificiële collega in de praktijk (01:11:16) Context engineering: De delicate dans van data (01:13:03) Toekomstbeelden in de zorg: De Erasmus MC-casus (01:14:07) Het drieluik: Brainstormen, bouwen en creëren (01:17:19) De PROMPT-formule als kompas (01:19:35) Een nieuwe blik op de horizon Meer weten: - Videoversie: https://youtu.be/xp_ujwrVqG4?si=-c9oN3B34nT8ohlu - Artikel: volgt - Nieuwsbrieven: volgt Notities: Kumar, H., Vincentius, J., Jordan, E., & Anderson, A. (2024). Human creativity in the age of LLMs: Randomized experiments on divergent and convergent thinking [Preprint]. arXiv. Boek Peter voor een lezing of workshop: - Boek Peter: peterjoosten.net/boekingsaanvraag Links: - Deze podcast: peterjoosten.net/podcast - Nieuwsbrief: peterjoosten.net/nieuwsbrief Abonneren: - Luister op iOS (Apple): peterjoosten.net/podcast-ios - Abonneer je op Spotify: peterjoosten.net/podcast-spotify - Luister op Google Podcasts: peterjoosten.net/podcast-google Illustraties: Mesa Verde Studio Productie: László Versteeg

El 1 de julio de 2025, el telescopio de sondeo del Sistema ATLAS en Chile descubrió el tercer objeto interestelar conocido: el cometa 3I/ATLAS. Con un brillo extremadamente tenue, en una magnitud aparente de 18, el objeto recién descubierto estaba entrando en el sistema solar interior a una velocidad de 61 km/s (220.000 km/h) con respecto al Sol, situado a una distancia de la Tierra de 3,52 UA (527 millones de kilómetros). Para septiembre de 2025 había aumentado hasta la magnitud 14-15. A medida que 3I/ATLAS se acercó al Sol a finales de octubre, sorprendió a los astrónomos al brillar mucho más rápido de lo esperado: de repente pasó a ser visible para observatorios espaciales con magnitud 9-10. Las observaciones iniciales no dejaron claro si se trataba de un asteroide o de un cometa. Las observaciones realizadas el 2 de julio de 2025 por el Deep Random Survey en Chile, el Lowell Discovery Telescope en Arizona y el Observatorio CFHT (Canadá, Francia, Hawai) en Mauna Kea mostraron una coma marginal y una cola corta de tres segundos de arco de longitud angular, lo que indica que el objeto es un cometa. El 2 de julio de 2025, el MPC (Centro de Planetas Menores, organismo del Observatorio Astrofísico Smithsoniano, que forma parte del Centro de astrofísica Harvard-Smithsonian junto con el Observatorio Harvard College) anunció el descubrimiento del cometa y le otorgó la designación de objeto interestelar «3I», lo que significa que era el tercer objeto interestelar confirmado. El MPC también le dio la designación de cometa no periódico C/2025 N1 (ATLAS).​ Cuando se anunció el proyecto 3I/ATLAS, el Centro de Planetas Menores había recopilado 122 observaciones del cometa desde 31 observatorios diferentes. La controversia ha estado presente en el descubrimiento de este objeto interestelar desde el inicio. El 16 de julio de 2025, el equipo del astrofísico Avi Loeb de la Universidad de Harvard publicó un artículo en arXiv especulando que 3I/ATLAS podría ser una nave espacial extraterrestre​ dadas algunas características «anómalas»,​ como su brillo inusual, no muestra aceleración no gravitacional, la falta de compuestos químicos identificables, su trayectoria precisa a través del Sistema Solar y la ausencia de desgasificación cometaria evidente en sus inicios, argumentando que no se comporta como un cometa. Loeb también ha observado que los chorros del cometa podrían estar disparándose en pulsos, creando un patrón repetitivo a modo de "latido". Cada estallido de gas y polvo hace que la coma se intensifique, se desvanezca y vuelva a intensificarse. Esto aún podría ser un proceso natural —por ejemplo, una zona muy volátil de la superficie que libera gas una vez por cada rotación—. Pero Loeb señala que, si los chorros no apuntan siempre hacia el Sol o si su sincronización resulta demasiado precisa el comportamiento empieza a parecer menos "propio de un cometa" y más... antinatural. En este caso, afirma, podríamos estar viendo indicios de un mecanismo artificial con un sistema de expulsión controlado. Otros astrónomos criticaron inmediatamente la especulación de Loeb.​ Darryl Seligman, que dirigió el primer estudio publicado sobre 3I/ATLAS, afirmó que «se han realizado numerosas observaciones telescópicas que demuestran que muestra señales clásicas de actividad cometaria». Además explicó que las sustancias químicas en el cometa podrían no ser detectables aún, dado que el objeto todavía se encontraba lejos del Sol.​ Desde entonces las observaciones han evidenciado indicios de un cometa clásico: presenta un pequeño núcleo helado, una coma rica en dióxido de carbono, ráfagas de vapor de agua e incluso una rara anticola causada por la geometría de la observación. El 29 de septiembre el cometa sobrevive ileso a una tormenta solar, a diferencia de muchos cometas del Sistema Solar. El 30 de septiembre 3I/ATLAS pasa detrás del Sol, desapareciendo de la vista hasta finales de noviembre. El 29 de octubre 3I/ATLAS sobrevive a su paso cercano al Sol, aumentando de brillo de forma repentina. El 2 de noviembre vuelve a ser visible tras perderse en el resplandor del Sol. El 4 de noviembre parece estar cambiando de color, adquiriendo un tono azulado. El 10 de noviembre el radiotelescopio MeerKAT en Sudáfrica detectó una señal de radio procedente de 3I/ATLAS —una emisión de fondo constante típica del gas y el polvo en el espacio, no una señal modulada que transporte información. Emisiones similares se registraron por primera vez en cometas en la década de 1970. El descubrimiento del cometa 3I/ATLAS es importante porque es un auténtico objeto interestelar, lo que significa que no se formó en nuestro Sistema Solar, sino que proviene de otro sistema estelar. Estos objetos son increíblemente raros, por lo que descubrir uno genera un gran revuelo. Además, las primeras estimaciones indican que su núcleo podría alcanzar los 5 kilómetros de diámetro, lo que lo convierte en uno de los cuerpos interestelares más grandes jamás detectados. Tras su máximo acercamiento al Sol a finales de octubre de 2025, 3I/ATLAS ha comenzado su viaje de regreso al espacio interestelar. Viajando alrededor de 30 kms/s en una trayectoria claramente hiperbólica, se mueve lo bastante rápido como para escapar por completo a la gravedad del Sol. El cometa ha alcanzado el perigeo (máximo acercamiento a la Tierra) el 19 de diciembre de 2025 y cruzará más allá de la órbita de Júpiter para el 16 de marzo de 2026. A comienzos de la década de 2030 habrá abandonado la región planetaria del Sistema Solar, continuando su silencioso viaje por la galaxia, tal como una vez llegó. Kaptep, Daryl Groetsch, Robert Scott Thompson, Ard Bit, 36 & zakè, Terminus Void, Max Corbacho, Infinite Third, David Helpling & Eric "the" Taylor, Pulse Mandala, Applefish, SoLaRiS, Keith Richie, Yawning Portal, Extraworld. 🎧 El playlist detallado: lostfrontier.org/space.html#1070

Hey ya'll, Alex here, and this week I was especially giddy to record the show! Mostly because when a thing clicks for me that hasn't clicked before, I can't wait to tell you all about it! This week, that thing is Agent Skills! The currently best way to customize your AI agents with domain expertise, in a simple, repeatable way that doesn't blow up the context window! We mentioned skills when Anthropic first released them (Oct 16) and when they became an open standard but it didn't really click until last week! So more on that below. Also this week, Anthropic released a research preview of Claude Cowork, an agentic tool for non coders, OpenAI finally let loos GPT 5.2 Codex (in the API, it was previously available only via Codex), Apple announced a deal with Gemini to power Siri, OpenAI and Anthropic both doubled down on healthcare and much more! We had an incredible show, with an expert in Agent Skills, Eleanor Berger and the usual gang on co-hosts, strongly recommend watching the show in addition to the newsletter! Also, I vibe coded skills support for all LLMs to Chorus, and promised folks a link to download it, so look for that in the footer, let's dive in! ThursdAI is where you stay up to date! Subscribe to keep us going! Big Company LLMs + APIs: Cowork, Codex, and a Browser in a WeekAnthropic launches Claude Cowork: Agentic AI for Non‑Coders (research preview)Anthropic announced Claude Cowork, which is basically Claude Code wrapped in a friendly UI for people who don't want to touch a terminal. It's a research preview available on the Max tier, and it gives Claude read/write access to a folder on your Mac so it can do real work without you caring about diffs, git, or command line.The wild bit is that Cowork was built in a week and a half, and according to the Anthropic team it was 100% written using Claude Code. This feels like a “we've crossed a threshold” moment. If you're wondering why this matters, it's because coding agents are general agents. If a model can write code to do tasks, it can do taxes, clean your desktop, or orchestrate workflows, and that means non‑developers can now access the same leverage developers have been enjoying for a year.It also isn't just for files—it comes with a Chrome connector, meaning it can navigate the web to gather info, download receipts, or do research and it uses skills (more on those later)Earlier this week I recorded this first reactions video about Cowork and I've been testing it ever since, it's a very interesting approach of coding agents that “hide the coding” to just... do things. Will this become as big as Claude Code for anthropic (which is reportedly a 1B business for them)? Let's see! There are real security concerns here, especially if you're not in the habit of backing up or using git. Cowork sandboxes a folder, but it can still delete things in that folder, so don't let it loose on your whole drive unless you like chaos.GPT‑5.2 Codex: Long‑Running Agents Are HereOpenAI shipped GPT‑5.2 Codex into the API finally! After being announced as the answer for Opus 4.5 and only being available in Codex. The big headline is SOTA on SWE-Bench and long‑running agentic capability. People describe it as methodical. It takes longer, but it's reliable on extended tasks, especially when you let it run without micromanaging.This model is now integrated into Cursor, GitHub Copilot, VS Code, Factory, and Vercel AI Gateway within hours of launch. It's also state‑of‑the‑art on SWE‑Bench Pro and Terminal‑Bench 2.0, and it has native context compaction. That last part matters because if you've ever run an agent for long sessions, the context gets bloated and the model gets dumber. Compaction is an attempt to keep it coherent by summarizing old context into fresh threads, and we debated whether it really works. I think it helps, but I also agree that the best strategy is still to run smaller, atomic tasks with clean context.Cursor vibe-coded browser with GPT-5.2 and 3M lines of codeThe most mind‑blowing thing we discussed is Cursor letting GPT‑5.2 Codex run for a full week to build a browser called FastRenderer. This is not Chromium‑based. It's a custom HTML parser, CSS cascade, layout engine, text shaping, paint pipeline, and even a JavaScript VM, written in Rust, from scratch. The codebase is open source on GitHub, and the full story is on Cursor's blog It took nearly 30,000 commits and millions of lines of code. The system ran hundreds of concurrent agents with a planner‑worker architecture, and GPT‑5.2 was the best model for staying on task in that long‑running regime. That's the real story, not just “lol a model wrote a browser.” This is a stress test for long‑horizon agentic software development, and it's a preview of how teams will ship in 2026.I said on the show, browsers are REALLY hard, it took two decades for the industry to settle and be able to render websites normally, and there's a reason everyone's using Chromium. This is VERY impressive

From historic medical evacuations to missing galaxies and stunning new images of the Milky Way, today's episode covers the latest breaking news from space exploration and astronomy. Join Anna and Avery as they discuss six fascinating stories from across the cosmos.---## Episode Timestamps**[00:00]** Intro  **[01:15]** Story 1: ISS Medical Evacuation  **[04:45]** Story 2: The Mystery of Missing Tiny Galaxies  **[08:30]** Story 3: NASA's MAVEN Spacecraft in Trouble  **[11:45]** Story 4: Viruses Behave Differently in Microgravity  **[14:30]** Story 5: Two New Exoplanets and Redefining Habitable Zones  **[17:00]** Story 6: Stunning New Radio Image of the Milky Way  **[19:30]** Outro---## Stories Covered### 1. Historic First Medical Evacuation from ISSFour International Space Station crew members successfully completed the first-ever medical evacuation in the ISS's 26-year history, splashing down safely in the Pacific Ocean off San Diego.**Key Points:**- SpaceX Crew-11 returned early after 5 months in space- Crew included US astronauts Mike Fincke and Zena Cardman, Russian cosmonaut Oleg Platonov, and Japanese astronaut Kimiya Yui- Splashdown occurred at 12:41 AM ET on January 15, 2026- Affected crew member remains in stable condition- Three crew members remain aboard ISS to continue operations- Demonstrates importance of medical protocols in long-duration spaceflight**Read More:**- [Phys.org: ISS astronauts splash down on Earth after first-ever medical evacuation](https://phys.org/news/2026-01-iss-astronauts-splash-earth-medical.html)---### 2. The Universe's Missing Tiny GalaxiesNew research using the James Webb Space Telescope suggests there may be far fewer small galaxies in the early universe than predicted by current models, challenging our understanding of cosmic evolution.**Key Points:**- Study led by Xuheng Ma from University of Wisconsin-Madison- Used JWST's UNCOVER program to study galaxies through gravitational lensing- Observed the Epoch of Reionization (12-13 billion years ago)- Discovery of "faint-end suppression" - galaxy numbers drop off at smaller sizes- Suggests intense radiation from early massive stars prevented small galaxies from forming- May require rethinking models of cosmic reionization- Used Abell 2744 galaxy cluster as a natural gravitational lens**Why It Matters:**This finding has major implications for our understanding of how the universe evolved from the "cosmic dark ages" to its current transparent state.**Read More:**- [Space.com: The universe should be packed with tiny galaxies — so where are they?](https://www.space.com/astronomy/galaxies/the-universe-should-be-packed-with-tiny-galaxies-so-where-are-they)- Research paper on arXiv (preprint database)---### 3. NASA Pessimistic About Recovering MAVEN Mars OrbiterNASA officials acknowledge it's "very unlikely" they'll recover the MAVEN spacecraft, which has been silent since December 6, 2025, marking a potential end to a highly productive Mars mission.**Key Points:**- MAVEN (Mars Atmosphere and Volatile Evolution) launched November 2013, entered Mars orbit September 2014- Last communication: December 6, 2025- Telemetry indicates spacecraft is tumbling and orbit may have changed- Solar conjunction (Mars and Earth on opposite sides of Sun) complicated recovery efforts- Attempts to photograph spacecraft with Curiosity rover were unsuccessful- Other orbiters (Mars Reconnaissance Orbiter, Mars Odyssey, ExoMars Trace Gas Orbiter) can maintain communications relay- Spacecraft studied Mars atmospheric loss and recently observed interstellar object 3I/ATLAS**Mission Legacy:**Despite the likely loss, MAVEN has provided over a decade of groundbreaking data about Mars' upper atmosphere and how solar wind strips away the Martian atmosphere.**Read More:**- [SpaceNews: NASA pessimistic about odds of recovering MAVEN](https://spacenews.com/nasa-pessimistic-about-odds-of-recovering-maven/)- [NASA Science: MAVEN Spacecraft Updates](https://science.nasa.gov/blogs/maven/)---### 4. Space Station Study Reveals Unusual Virus-Bacteria DynamicsUniversity of Wisconsin-Madison researchers discovered that viruses infecting bacteria evolve differently in microgravity, potentially opening new avenues for fighting antibiotic-resistant infections on Earth.**Key Points:**- Study used E. coli bacteria and bacteriophage T7- Parallel experiments conducted on ISS and Earth- Virus infection delayed but not blocked in microgravity- Both viruses and bacteria developed unique mutations in space- Space-evolved viruses showed increased activity against drug-resistant E. coli strains- Findings could lead to improved phage therapy for antibiotic-resistant infections- Published in PLOS Biology journal- Demonstrates ISS value as unique research platform**Scientific Significance:**This research shows how the space environment fundamentally alters evolutionary processes, and how these insights can be applied to solve problems on Earth.**Read More:**- [Space Daily: Space station study reveals unusual virus bacteria dynamics in microgravity](https://www.spacedaily.com/reports/Space_station_study_reveals_unusual_virus_bacteria_dynamics_in_microgravity_999.html)- Research paper: "Microgravity reshapes bacteriophage host coevolution aboard the International Space Station" in PLOS Biology---### 5. Two New Exoplanets Challenge Habitable Zone DefinitionsAstronomers have discovered two exoplanets orbiting red dwarf stars that are prompting scientists to expand the definition of potentially habitable worlds through the concept of "temperate zones."**Key Points:**- Research led by Madison Scott (University of Birmingham) and Georgina Dransfield (University of Oxford)- Introduces "temperate zone" concept: broader than traditional habitable zone- Temperate zone defined by insolation flux range: 0.1 < S/S⊕ < 5 (136-6,805 W/m²)- TOI-6716 b: Earth-sized (0.91-1.05 Earth radii), likely rocky- TOI-7384 b: Sub-Neptune (3.35-3.77 Earth radii), rocky core with thick H/He envelope- Both orbit mid to late-type M dwarfs (red dwarf stars)- Part of TEMPOS survey (Temperate M Dwarf Planets With SPECULOOS)- Good candidates for atmospheric studies with JWST- Paper submitted to Monthly Notices of the Royal Astronomical Society**Why Temperate Zones Matter:**As our understanding of habitability evolves, planets in temperate zones may prove more interesting than initially thought, especially for atmospheric characterization studies.**Read More:**- [Universe Today: Two New Exoplanets And The Need For New Habitable Zone Definitions](https://www.universetoday.com/articles/two-new-exoplanets-and-the-need-for-new-habitable-zone-definitions)---### 6. Most Detailed Radio Image of Milky Way Reveals Hidden StructuresAstronomers in Australia have released the most detailed low-frequency radio map of the Milky Way's southern sky, revealing thousands of previously hidden cosmic structures.**Key Points:**- Created by International Centre for Radio Astronomy Research (ICRAR)- Used Murchison Widefield Array (MWA) telescope in Western Australia- Data collected over 141 nights between 2013-2020- Required over 1 million CPU hours to process- GLEAM-X survey: 2x resolution, 10x sensitivity, 2x sky coverage vs. previous efforts- Cataloged over 98,000 radio sources- Shows supernova remnants (red circles) and stellar nurseries (blue regions)- Helps identify hidden supernova remnants and study pulsars- Led by PhD student Silvia Mantovanini (Curtin University)- First complete low-frequency radio image of Southern Galactic Plane**Future Impact:**This image serves as a foundation for the upcoming SKA-Low array, which will provide even more detailed views of the universe when operational.**Read More:**- [Daily Galaxy: New Image of the Milky Way Reveals Massive Hidden Structures](https://dailygalaxy.com/2026/01/new-image-milky-way-massive-structures/)- [ICRAR: GLEAM-X Galactic Plane](https://www.icrar.org/gleam-x-galactic-plane/)---## Key Terms Explained**Habitable Zone:** The range of distances from a star where conditions might allow liquid water to exist on a planet's surface.**Temperate Zone:** A broader classification than habitable zone, encompassing planets that receive moderate levels of stellar radiation.**Insolation Flux:** The amount of solar energy reaching a planet's surface, measured in watts per square meter.**Epoch of Reionization:** A period roughly 12-13 billion years ago when the first stars and galaxies began flooding the universe with ultraviolet light.**Gravitational Lensing:** The bending of light by massive objects due to gravity, which can magnify and brighten distant objects.**Bacteriophage:** A virus that infects and replicates within bacteria.**Solar Conjunction:** When Mars and Earth are on opposite sides of the Sun, disrupting radio communications.**M Dwarf (Red Dwarf):** Small, cool, dim stars that are the most common type of star in the galaxy.**Supernova Remnant:** The expanding cloud of gas and magnetic fields left behind after a star explodes.**Luminosity Function:** A cosmic census tool showing the distribution of galaxies at different brightness levels.---## Resources & Further Reading**Space Agencies:**- [NASA](https://www.nasa.gov)- [European Space Agency (Become a supporter of this podcast: https://www.spreaker.com/podcast/astronomy-daily-space-news-updates--5648921/support.Sponsor Details:Ensure your online privacy by using NordVPN. To get our special listener deal and save a lot of money, visit www.bitesz.com/nordvpn. You'll be glad you did!Become a supporter of Astronomy Daily by joining our Supporters Club. Commercial free episodes daily are only a click way... Click HereThis episode includes AI-generated content.

Send us a textOncoPTs are part of a patient's oncology team during the most vulnerable moments of life, including the end of life. However, many clinicians feel unprepared to hold conversations about death and dying and manage the emotional toll that comes with this work.In this APTA CSM Preview episode, I'm joined by Drs. Adam Matichak and Lori Boright for a grounded, compassionate conversation about what PTs need to know to show up confidently in hospice and palliative care settings.We talk about:The OncoPT's role when the goal is no longer “getting better,” but living fullyHow to navigate difficult conversations about functional decline, goals of care, and quality of lifeWhy “Death is not a dirty word”… and how normalizing these conversations can transform patient careNo matter your previous experience, this episode will help you understand the landscape, communicate more clearly, and care for yourself as you care for others.If you're headed to CSM 2026, add this session to your schedule:Death Is Not a Dirty Word: The What's What of Hospice and Palliative CareAI Research TodaySerious Conversations About Real AI Research; decoding the ArXiv into your headphones. Listen on: Apple Podcasts SpotifyFollow TheOncoPT on Instagram.Follow TheOncoPT on LinkedIn.

In this episode of The New Quantum Era, your host Sebastian Hassinger is joined by Chetan Nayak, Technical Fellow at Microsoft, professor of physics at the University of California Santa Barbara, and driving force behind Microsoft's quantum hardware R&D program. They discuss a modality of qubit that has not been covered on the podcast before, based on Majorana fermonic behaviors, which have the promise of providing topological protection against the errors which are such a challenge to quantum computing. Guest Bio Chetan Nayak is a Technical Fellow at Microsoft and leads the company's topological quantum hardware program, including the Majorana‑1 processor based on Majorana‑zero‑mode qubits.  He is also a professor of physics at UCSB and a leading theorist in topological phases of matter, non‑Abelian anyons, and topological quantum computation.  Chetan co‑founded Microsoft's Station Q  in 2005, building a bridge from theoretical proposals for topological qubits to engineered semiconductor–superconductor devices. What we talk about Chetan's first exposure to quantum computing in Peter Shor's lectures at the Institute for Advanced Study, and how that intersected with his PhD work with Frank Wilczek on non‑Abelian topological phases and Majorana zero modes.  The early days of topological quantum computation: fractional quantum Hall states at , emergent quasiparticles, and the realization that braiding these excitations naturally implements Clifford gates.  How Alexei Kitaev's toric‑code and Majorana‑chain ideas connected abstract topology to concrete condensed‑matter systems, and led to Chetan's collaboration with Michael Freedman and Sankar Das Sarma.  The 2005 proposal for a gallium‑arsenide quantum Hall device realizing a topological qubit, and the founding of Station Q to turn such theoretical blueprints into experimental devices in partnership with academic labs.  Why Microsoft pivoted from quantum Hall platforms to semiconductor–superconductor nanowires: leveraging the Fu–Kane proximity effect, spin–orbit‑coupled semiconductors, and a huge material design space—while wrestling with the challenges of interfaces and integration.  The evolution of the tetron architecture: two parallel topological nanowires with four Majorana zero modes, connected by a trivial superconducting wire and coupled to quantum dots that enable native Z‑ and X‑parity loop measurements.  How topological superconductivity allows a superconducting island to host even or odd total electron parity without a local signature, and why that nonlocal encoding provides hardware‑level protection for the qubit's logical 0 and 1.  Microsoft's roadmap in a 2D “quality vs. complexity” space: improving topological gap, readout signal‑to‑noise, and measurement fidelity while scaling from single tetrons to error‑corrected logical qubits and, ultimately, utility‑scale systems.  Error correction on top of topological qubits: using surface codes and Hastings–Haah Floquet codes with native two‑qubit parity measurements, and targeting hundreds of physical tetrons per logical qubit and thousands of logical qubits for applications like Shor's algorithm and quantum chemistry.  Engineering for scale: digital, on–off control of quantum‑dot couplings; cryogenic CMOS to fan out control lines inside the fridge; and why tetron size and microsecond‑scale operations sit in a sweet spot for both physics and classical feedback.  Where things stand today: the Majorana‑1 chiplet, recent tetron loop‑measurement experiments, DARPA's US2QC program, and how external users—starting with government and academic partners—will begin to access these devices before broader Azure Quantum integration. Papers and resources mentionedThese are representative papers and resources that align with topics and allusions in the conversation; they are good entry points if you want to go deeper.Non‑Abelian Anyons and Topological Quantum Computation – S. Das Sarma, M. Freedman, C. Nayak, Rev. Mod. Phys. 80, 1083 (2008); Early device proposalsSankar Das Sarma, Michael Freedman, and Chetan Nayak, “Topological quantum computation,” Physics Today 59(7), 32–38 (July 2006).Roadmap to fault‑tolerant quantum computation using topological qubits – C. Nayak et al., arXiv:2502.12252. Distinct lifetimes for X and Z loop measurements in a Majorana tetron - C. Nayaak et al., arXiv:2507.08795.Majorana qubit codes that also correct odd-weight errors - S. Kundu and B. Reichardt, arXiv:2311.01779. Microsoft's Majorana 1 chip carves new path for quantum computing, Microsoft blog post 

Hey folks, Alex here from Weights & Biases, with your weekly AI update (and a first live show of this year!) For the first time, we had a co-host of the show also be a guest on the show, Ryan Carson (from Amp) went supernova viral this week with an X article (1.5M views) about Ralph Wiggum (yeah, from Simpsons) and he broke down that agentic coding technique at the end of the show. LDJ and Nisten helped cover NVIDIA's incredible announcements during CES with their Vera Rubin upcoming platform (4-5X improvements) and we all got excited about AI medicine with ChatGPT going into Health officially! Plus, a bunch of Open Source news, let's get into this: ThursdAI - Recaps of the most high signal AI weekly spaces is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.Open Source: The “Small” Models Are WinningWe often talk about the massive frontier models, but this week, Open Source came largely from unexpected places and focused on efficiency, agents, and specific domains.Solar Open 100B: A Data MasterclassUpstage released Solar Open 100B, and it's a beast. It's a 102B parameter Mixture-of-Experts (MoE) model, but thanks to MoE magic, it only uses about 12B active parameters during inference. This means it punches incredibly high but runs fast.What I really appreciated here wasn't just the weights, but the transparency. They released a technical report detailing their “Data Factory” approach. They trained on nearly 20 trillion tokens, with a huge chunk being synthetic. They also used a dynamic curriculum that adjusted the difficulty and the ratio of synthetic data as training progressed. This transparency is what pushes the whole open source community forward.Technically, it hits 88.2 on MMLU and competes with top-tier models, especially in Korean language tasks. You can grab it on Hugging Face.MiroThinker 1.5: The DeepSeek Moment for Agents?We also saw MiroThinker 1.5, a 30B parameter model that is challenging the notion that you need massive scale to be smart. It uses something they call “Interactive Scaling.”Wolfram broke this down for us: this agent forms hypotheses, searches for evidence, and then iteratively revises its answers in a time-sensitive sandbox. It effectively “thinks” before answering. The result? It beats trillion-parameter models on search benchmarks like BrowseComp. It's significantly cheaper to run, too. This feels like the year where smaller models + clever harnesses (harnesses are the software wrapping the model) will outperform raw scale.Liquid AI LFM 2.5: Running on Toasters (Almost)We love Liquid AI and they are great friends of the show. They announced LFM 2.5 at CES with AMD, and these are tiny ~1B parameter models designed to run on-device. We're talking about running capable AI on your laptop, your phone, or edge devices (or the Reachy Mini bot that I showed off during the show! I gotta try and run LFM on him!)Probably the coolest part is the audio model. Usually, talking to an AI involves a pipeline: Speech-to-Text (ASR) -> LLM -> Text-to-Speech (TTS). Liquid's model is end-to-end. It hears audio and speaks audio directly. We watched a demo from Maxime Labonne where the model was doing real-time interaction, interleaving text and audio. It's incredibly fast and efficient. While it might not write a symphony for you, for on-device tasks like summarization or quick interactions, this is the future.NousCoder-14B and Zhipu AI IPOA quick shoutout to our friends at Nous Research who released NousCoder-14B, an open-source competitive programming model that achieved a 7% jump on LiveCodeBench accuracy in just four days of RL training on 48 NVIDIA B200 GPUs. The model was trained on 24,000 verifiable problems, and the lead researcher Joe Li noted it achieved in 4 days what took him 2 years as a teenager competing in programming contests. The full RL stack is open-sourced on GitHub and Nous published a great WandB results page as well! And in historic news, Zhipu AI (Z.ai)—the folks behind the GLM series—became the world's first major LLM company to IPO, raising $558 million on the Hong Kong Stock Exchange. Their GLM-4.7 currently ranks #1 among open-source and domestic models on both Artificial Analysis and LM Arena. Congrats to them!Big Companies & APIsNVIDIA CES: Vera Rubin Changes EverythingLDJ brought the heat on this one covering Jensen's CES keynote that unveiled the Vera Rubin platform, and the numbers are almost hard to believe. We're talking about a complete redesign of six chips: the Rubin GPU delivering 50 petaFLOPS of AI inference (5x Blackwell), the Vera CPU with 88 custom Olympus ARM cores, NVLink 6, ConnectX-9 SuperNIC, BlueField-4 DPU, and Spectrum-6 Ethernet.Let me put this in perspective using LDJ's breakdown: if you look at FP8 performance, the jump from Hopper to Blackwell was about 5x. The jump from Blackwell to Vera Rubin is over 3x again—but here's the kicker—while only adding about 200 watts of power draw. That's insane efficiency improvement.The real-world implications Jensen shared: training a 10 trillion parameter mixture-of-experts model now requires 75% fewer GPUs compared to Blackwell. Inference token costs drop roughly 10x—a 1MW cluster goes from 1 million to 10 million tokens per second at the same power. HBM4 memory delivers 22 TB/s bandwidth with 288GB capacity, exceeding NVIDIA's own 2024 projections by nearly 70%.As Ryan noted, when people say there's an AI bubble, this is why it's hilarious. Jensen keeps saying the need for inference is unbelievable and only going up exponentially. We all see this. I can't get enough inference—I want to spin up 10 Ralphs running concurrently! The NVL72 rack-scale system achieves 3.6 exaFLOPS inference with 20.7TB total HBM, and it's already shipping. Runway 4.5 is already running on the new platform, having ported their model from Hopper to Vera Rubin NVL72 in a single day.NVIDIA also recently acqui-hidred Groq (with a Q) in a ~$20 billion deal, bringing the inference chip expertise from the guy who created Google's TPUs in-house.Nemotron Speech ASR & The Speed of Voice (X, HF, Blog)NVIDIA also dropped Nemotron Speech ASR. This is a 600M parameter model that offers streaming transcription with 24ms latency.We showed a demo from our friend Kwindla Kramer at Daily. He was talking to an AI, and the response was virtually instant. The pipeline is: Nemotron (hearing) -> Llama/Nemotron Nano (thinking) -> Magpie TTS (speaking). The total latency is under 500ms. It feels like magic. Instant voice agents are going to be everywhere this year.XAI Raises $20B While Grok Causes Problems (Again)So here's the thing about covering anything Elon-related: it's impossible to separate signal from noise because there's an army of fans who hype everything and an army of critics who hate everything. But let me try to be objective here.XAI raised another massive Round E of $20 billion! at a $230 billion valuation, with NVIDIA and Cisco as strategic investors. The speed of their infrastructure buildout is genuinely incredible. Grok's voice mode is impressive. I use Grok for research and it's really good, notable for it's unprecedented access to X !But. This raise happened in the middle of a controversy where Grok's image model was being used to “put bikinis” on anyone in reply threads, including—and this is where I draw a hard line—minors. As Nisten pointed out on the show, it's not even hard to implement guardrails. You just put a 2B VL model in front and ask “is there a minor in this picture?” But people tested it, asked Grok not to use the feature, and it did it anyway. And yeah, putting Bikini on Claude is funny, but basic moderation is lacking! The response of “we'll prosecute illegal users” is stupid when there's no moderation built into the product. There's an enormous difference between Photoshop technically being able to do something after hours of work, and a feature that generates edited images in one second as the first comment to a celebrity, then gets amplified by the platform's algorithm to millions of people. One is a tool. The other is a product with amplification mechanics. Products need guardrails. I don't often link to CNN (in fact this is the first time) but they have a great writeup about the whole incident here which apparently includes the quitting of a few trust and safety folks and Elon's pushback on guardrails. CrazyThat said, Grok 5 is in training and XAI continues to ship impressive technology. I just wish they'd put the same engineering effort into safety as they do into capabilities!OpenAI Launches GPT HealthThis one's exciting. OpenAI CEO Fidji Simo announced ChatGPT Health, a privacy-first space for personalized health conversations that can connect to electronic health records, Apple Health, Function Health, Peloton, and MyFitnessPal.Here's why this matters: health already represents about 5% of all ChatGPT messages globally and touches 25% of weekly active users—often outside clinic hours or in underserved areas. People are already using these models for health advice constantly.Nisten, who has worked on AI doctors since the GPT-3 days and even published papers on on-device medical AI, gave us some perspective: the models have been fantastic for health stuff for two years now. The key insight is that medical data seems like a lot, but there are really only about 2,000 prescription drugs and 2,000 diseases (10,000 if you count rare ones). That's nothing for an LLM. The models excel at pattern recognition across this relatively contained dataset.The integration with Function Health is particularly interesting to me. Function does 160+ lab tests, but many doctors won't interpret them because they didn't order them. ChatGPT could help bridge that gap, telling you “hey, this biomarker looks off, you should discuss this with your doctor.” The bad news is, this is just a waitlist and you can add yourself to the waitlist here, we'll keep monitoring the situation and let you know when it opens upDoctronic: AI Prescribing Without Physician OversightSpeaking of healthcare, Doctronic launched a pilot in Utah where AI can autonomously renew prescriptions for chronic conditions without any physician in the loop. The system covers about 190 routine medications (excluding controlled substances) at just $4 per renewal. Trial data showed 99.2% concordance with physician treatment plans, and they've secured pioneering malpractice insurance that treats the AI like a clinician.Nisten made the case that it's ethically wrong to delay this kind of automation when ER wait times keep increasing and doctors are overworked. The open source models are already excellent at medical tasks. Governments should be buying GPUs rather than creating administrative roadblocks. Strong strong agree here! Google Brings Gmail into the Gemini Era (X)Breaking news from the day of our show: Google announced Gmail's biggest AI transformation since its 2004 launch, powered by Gemini 3. This brings AI Overviews that summarize email threads, natural language queries (”Who gave me a plumber quote last year?”), Help Me Write, contextual Suggested Replies matching your writing style, and the upcoming AI Inbox that filters noise to surface VIPs and urgent items.For 3 billion Gmail users, this is huge. I'm very excited to test it—though not live on the show because I don't want you reading my emails.This weeks buzz - covering Weights & Biases updatesNot covered on the show, but a great update on stuff from WandB, Chris Van Pelt (@vanpelt), one of the 3 co-founders released a great project I wanted to tell you about! For coders, this is an app that allows you to run multiple Claude Codes on free Github sandboxes, so you can code (or Ralph) and control everything away from home! GitHub gives personal users 120 free Codespaces hours/month, and Catnip automatically shuts down inactive instances so you can code for quite a while with Catnip! It's fully open source on Github and you can download the app hereInterview: Ryan Carson - What the hell is Ralph Wiggum?Okay, let's talk about the character everyone is seeing on their timeline: Ralph Wiggum. My co-host Ryan Carson went viral this week with an article about this technique, and I had to have him break it down.Ralph isn't a new model; it's a technique for running agents in a loop to perform autonomous coding. The core idea is deceptively simple: Ralph is a bash script that loops an AI coding agent. In a loop, until it a certain condition is met. But why is it blowing up? Normally when you use a coding agent like Cursor, Claude Code, or AMP, you need to be in the loop. You approve changes, look at code, fix things when the agent hits walls or runs out of context. Ralph solves this by letting the agent run autonomously while you sleep.Here's how it works: First, you write a Product Requirements Doc (PRD) by talking to your agent for a few minutes about what you want to build. Then you convert that PRD into a JSON file containing atomic user stories with clear acceptance criteria. Each user story is small enough for the agent to complete in one focused thread.The Ralph script then loops: it picks the first incomplete user story, the agent writes code to implement it, tests against the acceptance criteria, commits the changes, marks the story as complete, writes what it learned to a shared “agents.md” file, and loops to the next story. That compound learning step is crucial—without it, the agent would keep making the same mistakes.What makes this work is the pre-work. As Ryan put it, “no real work is done one-shot.” This is how software engineering has always worked—you break big problems into smaller problems into user stories and solve them incrementally. The innovation is letting AI agents work through that queue autonomously while you sleep! Ryan's excellent (and viral) X article is here! Vision & VideoLTX-2 Goes Fully Open Source (HF, Paper)Lightricks finally open-sourced LTX-2, marking a major milestone as the first fully open audio-video generation model. This isn't just “we released the weights” open—it's complete model weights (13B and 2B variants), distilled versions, controllable LoRAs, a full multimodal trainer, benchmarks, and evaluation scripts. For a video model that is aiming to be the open source SORA, supports audio and lipsyncThe model generates synchronized audio and video in a single DiT-based architecture—motion, dialogue, ambience, and music flow simultaneously. Native 4K at up to 50 FPS with audio up to 10 seconds. And there's also a distilled version (Thanks Pruna AI!) hosted on ReplicateComfyUI provided day-0 native support, and community testing shows an A6000 generating 1280x720 at 120 frames in 50 seconds. This is near Sora-level quality that you can fine-tune on your own data for custom styles and voices in about an hour.What a way to start 2026. From chips that are 5x faster to AI doctors prescribing meds in Utah, the pace is only accelerating. If anyone tells you we're in an AI bubble, just show them what we covered today. Even if the models stopped improving tomorrow, the techniques like “Ralph” prove we have years of work ahead of us just figuring out how to use the intelligence we already have.Thank you for being a ThursdAI subscriber. See you next week!As always, here's the show notes and TL;DR links: * Hosts & Guests* Alex Volkov - AI Evangelist & Weights & Biases (@altryne)* Co-Hosts - @WolframRvnwlf, @nisten, @ldjconfirmed* Special Guest - Ryan Carson (@ryancarson) breaking down the Ralph Wiggum technique.* Open Source LLMs* Solar Open 100B - Upstage's 102B MoE model. Trained on 19.7T tokens with a heavy focus on “data factory” synthetic data and high-performance Korean reasoning (X, HF, Tech Report).* MiroThinker 1.5 - A 30B parameter search agent that uses “Interactive Scaling” to beat trillion-parameter models on search benchmarks like BrowseComp (X, HF, GitHub).* Liquid AI LFM 2.5 - A family of 1B models designed for edge devices. Features a revolutionary end-to-end audio model that skips the ASR-LLM-TTS pipeline (X, HF).* NousCoder-14B - competitive coding model from Nous Research that saw a 7% LiveCodeBench accuracy jump in just 4 days of RL (X, WandB Dashboard).* Zhipu AI IPO - The makers of GLM became the first major LLM firm to go public on the HKEX, raising $558M (Announcement).* Big Co LLMs & APIs* NVIDIA Vera Rubin - Jensen Huang's CES reveal of the next-gen platform. Delivers 5x Blackwell inference performance and 75% fewer GPUs needed for MoE training (Blog).* OpenAI ChatGPT Health - A privacy-first vertical for EHR and fitness data integration (Waitlist).* Google Gmail Era - Gemini 3 integration into Gmail for 3 billion users, featuring AI Overviews and natural language inbox search (Blog).* XAI $20B Raise - Elon's XAI raises Series E at a $230B valuation, even as Grok faces heat over bikini-gate and safety guardrails (CNN Report).* Doctronic - The first US pilot in Utah for autonomous AI prescription renewals without a physician in the loop (Web).* Alexa+ Web - Amazon brings the “Smart Alexa” experience to browser-based chat (Announcement).* Autonomous Coding & Tools* Ralph Wiggum - The agentic loop technique for autonomous coding using small, atomic user stories. Ryan Carson's breakdown of why this is the death of “vibe coding” (Viral X Article).* Catnip by W&B - Chris Van Pelt's open-source iOS app to run Claude Code anywhere via GitHub Codespaces (App Store, GitHub).* Vision & Video* LTX-2 - Lightricks open-sources the first truly open audio-video generation model with synchronized output and full training code (GitHub, Replicate Demo).* Avatar Forcing - KAIST's framework for real-time interactive talking heads with ~500ms latency (Arxiv).* Qwen Edit 2512 - Optimized by PrunaAI to generate high-res realistic images in under 7 seconds (Replicate).* Voice & Audio* Nemotron Speech ASR - NVIDIA's 600M parameter streaming model with sub-100ms stable latency for massive-scale voice agents (HF). This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sub.thursdai.news/subscribe

Evénements QUEST-IS chez EDF La conférence dédiée à l'ingénierie quantique organisée par la SEE avec le soutien du SGPI et de l'AID durait trois jours pendant la première semaine de décembre, chez EDF à Palaiseau.https://www.oezratty.net/Files/Conferences/Olivier%20Ezratty%20QUEST-IS%20Quantum%20Engineering%20Dec2025.pdfhttps://conference-questis.org/quest-is-2025/program/proceddings/Conférence organisée par le Fermilab faisait aussi le point sur le lien entre calcul quantique et HPC, notamment dans le cadre de simulations dans la high energy physics (HEP). https://indico.fnal.gov/event/71571/. Q2B Santa Clara du 9 au 11 décembre. La conférence rassemblait des intervenants tels que Scott Aaronson (University of Texas), John Preskill (Caltech), Ryan Babbush (Google), et plein de CTOs et CEO de startups US et internationales. Dans les startups étrangères, il y avait notamment les interventions de Photonic (Stephanie Simmons), Diraq (Andrew Dzurak), Quantum Machines (Yonathan Cohen), Q-CTRL (Michael Biercuk), Classiq, Quemix (Japon, dans le logiciel), etc. Et aussi Joe Altepeter de la DARPA, qui présentait l'état des lieu du programme Quantum Benchmark Initiative.https://www.oezratty.net/Files/Conferences/Olivier%20Ezratty%20Q2B%20SV%20FTQC%20Energetics%20Dec2025.pdfhttps://www.oezratty.net/Files/Conferences/Olivier%20Ezratty%20Q2B%20SV%20Case%20Studies%20Dec2025.pdfhttps://www.oezratty.net/wordpress/2025/back-from-the-q2b-santa-clara-2025/Les vidéos de la conférence à Munich qui avait lieu en octobre 2025.https://www.cda.cit.tum.de/research/quantum/mqsf/ Nobel lectures des trois lauréats du prix Nobel de physique 2025, John Clarke, Michel Devoret et John Martinis.https://www.youtube.com/watch?v=jTtT2jTF4Xc A noter également les interventions de Michel Devoret et Alain Aspect chez Google à Paris le 18 décembre. Le duo était suivi d'un panel avec eux en compagnie de Pierre Rouchon et Théau Péronnin d'Alice&Bob. Conférence inaugurale de Pascale Senellart au Collège de France https://www.college-de-france.fr/fr/agenda/lecon-inaugurale/les-debuts-une-seconde-revolution-quantique/les-debuts-une-seconde-revolution-quantiqueBell prize coattribué à Antoine Browayes, Mikhael Lukin et Mark Safman. Le trio mondial des atomes froids. https://cqiqc.physics.utoronto.ca/bell-prize/bell-prize-winners/browaeys-lukin-and-saffman-awarded-the-9th-bell-prize/https://en.wikipedia.org/wiki/John_Stewart_Bell_Prize Evénements à venir ·       CES 2026 & Quantum World Congress https://www.quantumworldcongress.com/ces-2026.       Un séminaire d'une journée sur les technologies quantiques organisé à l'école navale près de Brest le 5 février 2025. https://www.ecole-navale.fr/entreprise/nos-rendez-vous/journees-sciences-navales/·       Une matinée organisée pour le MEDEF à Lyon le 27 février janvier avec Alain Aspect, Olivier Hess (Eviden) et Andréa Le Vot (Crédit Agricole). Organisée par la maison du quantique de Grenoble.·       APS March Meeting à Denver des 15 au 20 mars 2026.  France Pasqal Annonce de 324 qubits en décembre en mode analogique.  https://www.linkedin.com/posts/pasqal_quantumcomputing-quantumadvantage-innovation-activity-7407389213739991040-He_T?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAAVgdUB3BosVI5xO_hehYGjYCuwBq1TXaA En décembre 2025, Pasqal annonçait aussi avoir cumulé 145M€ de financements qui rassemble de l'investissement en capital, des promesses d'investissements à venir, des aides publiques et des commandes clients.https://www.linkedin.com/posts/pasqal_quantumcomputing-innovation-deeptech-activity-7399718793712914433-2bW5 Alice&Bob Alice & Bob communiquait sur l'énergie qui sera consommée par Graphene, leur futur ordinateur quantique supportant 100 qubits logiques et un million d'opérations en 2030. Cela fera 160 kW. https://www.linkedin.com/posts/alice-bob_sc25-quantumcomputing-hpc-activity-7399453247091929089-xaEe/ Qubit Pharmaceuticals Papier sur la simulation du placement de molécules d'eau dans des protéines, réalisée sur un ordinateur quantique d'IBM avec jusqu'à 123 qubits. L'algorithme utilise une fonction de coût formulée en QUBO et un solver provenant de Q-CTRL.  Practical protein-pocket hydration-site prediction for drug discovery on a quantum computer by Daniele Loco, Kisa Barkemeyer, Andre R. R. Carvalho, and Jean-Philip Piquemal, arXiv, December 2025 (20 pages). Analyse géopolitique d'Axel FerrazziniQuantum, Diplomacy, and Geopolitics by Axel Ferrazzini, arXiv, December 2025 (20 pages). International IBM Leur nouveau processeur NightHawk serait en ligne depuis décembre pour certains clients. Il s'appelle

This is a recap of the top 10 posts on Hacker News on December 31, 2025. This podcast was generated by wondercraft.ai (00:30): Stardew Valley developer made a $125k donation to the FOSS C# framework MonoGameOriginal post: https://news.ycombinator.com/item?id=46445068&utm_source=wondercraft_ai(01:50): Warren Buffett steps down as Berkshire Hathaway CEO after six decadesOriginal post: https://news.ycombinator.com/item?id=46448705&utm_source=wondercraft_ai(03:10): I canceled my book dealOriginal post: https://news.ycombinator.com/item?id=46446815&utm_source=wondercraft_ai(04:31): Show HN: Use Claude Code to Query 600 GB Indexes over Hacker News, ArXiv, etc.Original post: https://news.ycombinator.com/item?id=46442245&utm_source=wondercraft_ai(05:51): 2025: The Year in LLMsOriginal post: https://news.ycombinator.com/item?id=46449643&utm_source=wondercraft_ai(07:12): Tell HN: Happy New YearOriginal post: https://news.ycombinator.com/item?id=46443744&utm_source=wondercraft_ai(08:32): Akin's Laws of Spacecraft Design (2011) [pdf]Original post: https://news.ycombinator.com/item?id=46442903&utm_source=wondercraft_ai(09:52): Efficient method to capture carbon dioxide from the atmosphereOriginal post: https://news.ycombinator.com/item?id=46444076&utm_source=wondercraft_ai(11:13): The rise of industrial softwareOriginal post: https://news.ycombinator.com/item?id=46442597&utm_source=wondercraft_ai(12:33): Meta created 'playbook' to fend off pressure to crack down on scammersOriginal post: https://news.ycombinator.com/item?id=46446838&utm_source=wondercraft_aiThis is a third-party project, independent from HN and YC. Text and audio generated using AI, by wondercraft.ai. Create your own studio quality podcast with text as the only input in seconds at app.wondercraft.ai. Issues or feedback? We'd love to hear from you: team@wondercraft.ai

L'intelligence artificielle est-elle réellement prête à remplacer les humains au travail ? La question agite les entreprises, entre fantasmes de productivité totale et scepticisme assumé. Pour dépasser les discours, des chercheurs de Carnegie Mellon University ont tenté une expérience originale : simuler une entreprise presque entièrement peuplée d'agents d'intelligence artificielle. Les résultats, publiés en prépublication sur arXiv, sont loin d'annoncer la fin du salariat humain. Les scientifiques ont confié le fonctionnement de cette entreprise virtuelle à des agents issus des modèles les plus en vue du moment : Claude d'Anthropic, GPT-4o d'OpenAI, Gemini de Google, Nova d'Amazon, Llama de Meta ou encore Qwen d'Alibaba. Chaque agent s'est vu attribuer un poste classique : analyste financier, chef de projet, ingénieur logiciel. En parallèle, d'autres agents jouaient le rôle de collègues ou de services internes, comme les ressources humaines.Sur le papier, tout était en place. Dans les faits, la performance est restée très limitée. Le meilleur élève, Claude 3.5 Sonnet, n'a réussi à mener à terme que 24 % des tâches confiées. En incluant les missions partiellement accomplies, son score plafonne à 34,4 %. Gemini 2.0 Flash arrive en deuxième position, avec à peine 11,4 % de tâches finalisées. Aucun autre modèle ne dépasse la barre des 10 %. Un contraste saisissant avec les promesses d'autonomie souvent associées à ces systèmes. Les chercheurs identifient plusieurs causes à ces échecs. Les agents peinent à comprendre les implicites humains : demander un fichier en « .docx » ne leur permet pas toujours de déduire qu'il s'agit d'un document Word. Ils manquent aussi de compétences sociales élémentaires et se retrouvent rapidement bloqués lorsqu'ils doivent naviguer sur le Web, gérer des fenêtres surgissantes ou interpréter des interfaces complexes. Plus préoccupant encore, certains agents estiment avoir réussi une tâche après en avoir simplement contourné les étapes difficiles.En clair, cette expérience rappelle une réalité souvent oubliée : si les IA excellent sur des tâches ciblées et bien définies, elles restent très loin de pouvoir gérer seules un environnement de travail réel, avec ses imprévus, ses règles implicites et ses interactions humaines. Le remplacement généralisé des salariés, lui, peut encore attendre. Hébergé par Acast. Visitez acast.com/privacy pour plus d'informations.

We welcome back to the podcast our friend Kevin Knuth - of viking extraction ("Knud") … . Kevin is a Professor of Physics at the University at Albany. He is a former NASA research scientist designing artificial intelligence algorithms for astrophysical data analysis. His current research interests include the foundations of physics, quantum information, autonomous robotics, the search for and characterization of extrasolar planets, and unidentified aerial phenomena (UAP). Put on your seatbelts: angel hair, engineer-animals and chemist-plants, "curiosity band-width", extra terrestrial gardeners, lizards and spoons, 1954 incident above a soccer field in Florence, studying refractory time, remote viewing in the family, UAPx, levitation, caterpillar silk from barbell-shaped ufos, believing and seeing, physics we haven't studied yet, what Kevin figured out finally and a donation to Kevin's university and endowment to help him and his team study it all. Stickshifts and Safetybelts. https://youtu.be/qW9mZLnQ7Yc?si=-0od8M2LGZOLoLX3   Episodes mentioned: Kevin Day (ep 171), Stephan Schwartz (ep 155,156), Diana Pasulka (ep 142), Jeffrey Krypal (ep 118, 189)   Kevin, UAP paper: ·       The new science of Unidentified Aerospace-Undersea Phenomena (UAP): http://knuthlab.org/papers/knuth+etal---2025--new-science-UAP.pdf ·       arXiv: https://arxiv.org/abs/2502.06794 New Endowment Fund at University of Albany: https://www.albany.edu/news-center/news/2025-ualbany-receives-major-gift-advance-scientific-research-unidentified-aerial A Tear in the Sky: https://www.imdb.com/title/tt15120486/ UAPx: https://www.uapexpedition.org/about 1954 incident in Florence: https://www.bbc.com/news/magazine-29342407

TL;DR LLMs pretrained on data about misaligned AIs themselves become less aligned. Luckily, pretraining LLMs with synthetic data about good AIs helps them become more aligned. These alignment priors persist through post-training, providing alignment-in-depth. We recommend labs pretrain for alignment, just as they do for capabilities. Website: alignmentpretraining.ai Us: geodesicresearch.org | x.com/geodesresearch Note: We are currently garnering feedback here before submitting to ICML. Any suggestions here or on our Google Doc are welcome! We will be releasing a revision on arXiv in the coming days. Folks who leave feedback will be added to the Acknowledgment section. Thank you! Abstract We pretrained a suite of 6.9B-parameter LLMs, varying only the content related to AI systems, and evaluated them for misalignment. When filtering the vast majority of the content related to AI, we see significant decreases in misalignment rates. The opposite was also true - synthetic positive AI data led to self-fulfilling alignment. While post-training decreased the effect size, benign fine-tuning[1] degrades the effects of post-training, models revert toward their midtraining misalignment rates. Models pretrained on realistic or artificial upsampled negative AI discourse become more misaligned with benign fine-tuning, while models pretrained on only positive AI discourse become more aligned. This [...] ---Outline:(00:15) TL;DR(01:10) Abstract(02:52) Background and Motivation(04:38) Methodology(04:41) Misalignment Evaluations(06:39) Synthetic AI Discourse Generation(07:57) Data Filtering(08:27) Training Setup(09:06) Post-Training(09:37) Results(09:41) Base Models: AI Discourse Causally Affects Alignment(10:50) Post-Training: Effects Persist(12:14) Tampering: Pretraining Provides Alignment-In-Depth(14:10) Additional Results(15:23) Discussion(15:26) Pretraining as Creating Good Alignment Priors(16:09) Curation Outperforms Naive Filtering(17:07) Alignment Pretraining(17:28) Limitations(18:16) Next Steps and Call for Feedback(19:18) Acknowledgements The original text contained 1 footnote which was omitted from this narration. --- First published: December 20th, 2025 Source: https://www.lesswrong.com/posts/TcfyGD2aKdZ7Rt3hk/alignment-pretraining-ai-discourse-causes-self-fulfilling --- Narrated by TYPE III AUDIO. ---Images from the article:

An interstellar object just cruised into our cosmic neighborhood and—naturally—humanity immediately responded with calm, measured curiosity… by screaming “ALIENS!” into the void.This week on Hysteria 51, we're diving into 3I/ATLAS, the latest confirmed visitor from beyond our solar system. Is it just a dirty, fast-moving comet minding its own business? Or is it something a little more… engineered? We break down what scientists actually know so far, why this thing has the internet acting like it found a Wi-Fi signal in the Kuiper Belt, and what the “completely natural” crowd says (spoiler: they have math and spectra and other buzzkills).Then we do what we do best: lean into the fun. If 3I/ATLAS is under intelligent control—even hypothetically—what does that mean for humanity? First contact? Surveillance? A cosmic prank? Or just a high-speed reminder that we are absolutely not the main character of the universe.Strap in, stare into the sky suspiciously, and remember: “closest approach” doesn't mean “close,” but it does mean the conspiracy machine is about to hit maximum RPM this week on Hysteria 51!Special thanks to this week's research sources:YouTube (video): 3I/ATLAS Is Causing Scientists To Panic. Here's Why. YouTube. YouTubeNASA Science: NASA. (n.d.). Comet 3I/ATLAS (mission/observation timeline page). NASA Science. NASA ScienceMinor Planet Center (MPEC): Minor Planet Center. (2025, Jul 2). MPEC 2025-N12: 3I/ATLAS = C/2025 N1 (ATLAS). Minor Planet CenterarXiv (early discovery & characterization): Discovery and Preliminary Characterization of a Third Interstellar Object: 3I/ATLAS. arXiv:2507.02757. (2025). arXivMNRAS (SOAR photometry): Frincke, T. T., et al. (2026). Near-discovery SOAR photometry of the third interstellar object: 3I/ATLAS. Monthly Notices of the Royal Astronomical Society (Advance Article). OUP AcademicMNRAS Letters (spectrum paper): Opitom, C., Snodgrass, C., Jehin, E., et al. (2025, Nov). Snapshot of a new interstellar comet: 3I/ATLAS has a red and featureless spectrum. MNRAS: Letters, 544(1), L31–L36. OUP AcademicA&A (context vs other comets/interstellars): de la Fuente Marcos, R., et al. (2025). Assessing interstellar comet 3I/ATLAS with the 10.4 m Gran Telescopio Canarias… Astronomy & Astrophysics. A&A Scientific JournalarXiv (technological hypothesis paper): Hibberd, A., Crowl, A., & Loeb, A. (2025, Jul 16). Is the Interstellar Object 3I/ATLAS Alien Technology? arXiv:2507.12213. arXivJason Wright (rebuttal / anomaly breakdown): Wright, J. (2025, Nov 9). Loeb's 3I/ATLAS “Anomalies” Explained. AstroWright (Penn State). Penn State SitesSecondary roundup referencing rebuttal: ScienceAlert. (2025, Nov 13). Don't Panic! 3I/ATLAS Isn't an Alien Death Probe, But It Is Wildly Unusual. ScienceAlertEmail us your favorite WEIRD news stories:weird@hysteria51.comSupport the ShowGet exclusive content & perks as well as an ad and sponsor free experience at https://www.patreon.com/Hysteria51 from just $1ShopBe the Best Dressed at your Cult Meeting!https://www.teepublic.com/stores/hysteria51?ref_id=9022See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

What is Spaceborne Lunar? This week, Technology Now explore how, and why, you would put a supercomputer on the moon. We ask why anyone would want to put a supercomputer on the moon, we discover how one would go about doing such a thing, and we explore the benefits that this sort of extreme edge computing could bring. Norm Follett, Senior Director, HPE Global Technical Marketing, Space Technologies & Solutions, tells us more.This is Technology Now, a weekly show from Hewlett Packard Enterprise. Every week, hosts Michael Bird and Sam Jarrell look at a story that's been making headlines, take a look at the technology behind it, and explain why it matters to organizations. This episode is available in both video and audio formats.About Norm: https://www.linkedin.com/in/normfollett/Sources:https://www.hpe.com/us/en/newsroom/accelerating-space-exploration-with-the-spaceborne-computer.htmlhttps://spectrum.ieee.org/software-as-hardware-apollos-rope-memoryhttps://www.bcs.org/articles-opinion-and-research/the-first-computers-on-the-moon/https://www.sciencefocus.com/space/what-tech-would-the-apollo-11-mission-have-todayhttps://www.bbc.co.uk/future/article/20230516-apollo-how-moon-missions-changed-the-modern-worldAverill C., 2022, a Brief Analysis of the Apollo Guidance Computer, https://doi.org/10.48550/arXiv.2201.08230

The Women in AI Healthcare event series - hosted by Real Chemistry in collaboration with Pharma Brands – brings together dynamic female leaders to discuss the transformative role of artificial intelligence in life sciences. It is also a call to action: to ensure women are not only present, but pivotal in shaping the future of AI in healthcare. In a new pharmaphorum podcast focused on the important and timely subject of women in AI, web editor Nicole Raleigh spoke with: Kate Eversole, event director at Pharma Brands; Celine Parmentier, EVP, head of global med comms at Real Chemistry; and Emma Slade, head of applied AI at Tangram Therapeutics. The guests discuss their own work with AI, the risk of training AI models predominantly on male data, and how, within life sciences, women are already shaping, challenging, and advocating for AI. The conversation also touches upon the possible next greatest impacts of AI in the sector, and the need to keep the ‘human in the loop', as well as the possible negative impacts if AI is relied upon too much. You can listen to episode 232 of the pharmaphorum podcast in the player below, download the episode to your computer, or find it - and subscribe to the rest of the series – on Apple Podcasts, Spotify, Overcast, Pocket Casts, Podbean, and pretty much wherever else you download your other podcasts from.   Resources mentioned within the conversation are as below: Kotek, H., Dockum, R., & Sun, C. (2023). Gender Bias and Stereotypes in Large Language Models. arXiv:2304.02485. UN Women & UNESCO (2020). I'd Blush If I Could: Closing Gender Divides in Digital Skills Through Education. Tatman, R. (2017). Gender and dialect bias in YouTube's automatic captions. Criado-Perez, C. (2019). Invisible Women: Data Bias in a World Designed for Men. King, M. (2020). The Fix: Overcome the Invisible Barriers That Are Holding Women Back at Work.   You can register to be a part of the women in AI community here: https://www.pharmabrands.ca/womeninai   Information on the survey being run by Dr Michelle Penelope King, on AI and workplace motivation, can be found here: https://lnkd.in/eCg87_7w

A inteligência artificial, em seus múltiplos sentidos, tem dominado a agenda pública e até mesmo o direcionamento do capital das grandes empresas de tecnologia. Mas você já parou para pensar na infraestrutura gigantesca que dê conta de sustentar o crescimento acelerado das IAs? O futuro e o presente da inteligência artificial passa pela existência dos datacenters. E agora é mais urgente que nunca a gente discutir esse assunto. Estamos vendo um movimento se concretizar, que parece mais uma forma de colonialismo digital: com a crescente resistência à construção de datacenters nos países no norte global, empresas e governos parecem estar convencidos a trazer essas infraestruturas imensas com todos os seus impactos negativos ao sul global. Nesse episódio Yama Chiodi e Damny Laya conversam com pesquisadores, ativistas e atingidos para tentar aprofundar o debate sobre a infraestrutura material das IAs. A gente conversa sobre o que são datacenters e como eles impactam e irão impactar nossas vidas. No segundo episódio, recuperamos movimentos de resistência a sua instalação no Brasil e como nosso país se insere no debate, seguindo a perspectiva de ativistas e de pesquisadores da área que estão buscando uma regulação mais justa para esses grandes empreendimentos.  ______________________________________________________________________________________________ ROTEIRO [ vinheta da série ] [ Começa bio-unit ] YAMA: A inteligência artificial, em seus múltiplos sentidos, tem dominado a agenda pública e até mesmo o direcionamento do capital das grandes empresas de tecnologia. Mas você já parou para pensar na infraestrutura gigantesca que dê conta de sustentar o crescimento acelerado das IA? DAMNY: O futuro e o presente da inteligência artificial passa pela existência dos data centers. E agora é mais urgente que nunca a gente discutir esse assunto. Estamos vendo um movimento se concretizar, que parece mais uma forma de colonialismo digital: com a crescente resistência à construção de datacenters nos países no norte global, empresas e governos parecem estar convencidos a trazer os datacenters com todos os seus impactos negativos ao sul global. YAMA: Nós conversamos com pesquisadores, ativistas e atingidos e em dois episódios nós vamos tentar aprofundar o debate sobre a infraestrutura material das IAs. No primeiro, a gente conversa sobre o que são datacenters e como eles impactam e irão impactar nossas vidas. DAMNY: No segundo, recuperamos movimentos de resistência a sua instalação no Brasil e como nosso país se insere no debate, seguindo a perspectiva de ativistas e de pesquisadores da área que estão buscando uma regulação mais justa para esses grandes empreendimentos. [ tom baixo ] YAMA: Eu sou o Yama Chiodi, jornalista de ciência e pesquisador do campo das mudanças climáticas. Se você já é ouvinte do oxigênio pode ter me ouvido aqui na série cidade de ferro ou no episódio sobre antropoceno. Ao longo dos últimos meses investiguei os impactos ambientais das inteligências artificiais para um projeto comum entre o LABMEM, o laboratório de mudança tecnológica, energia e meio ambiente, e o oxigênio. Em setembro passado, o Damny se juntou a mim pra gente construir esses episódios juntos. E não por acaso. O Damny publicou em outubro passado um relatório sobre os impactos socioambientais dos data centers no Brasil, intitulado “Não somos quintal de data center”. O link para o relatório completo se encontra disponível na descrição do episódio. Bem-vindo ao Oxigênio, Dam. DAMNY: Oi Yama. Obrigado pelo convite pra construir junto esses episódios. YAMA: É um prazer, meu amigo. DAMNY: Eu também atuo como jornalista de ciência e sou pesquisador de governança da internet já há algum tempo. Estou agora trabalhando como jornalista e pesquisador aqui no LABJOR, mas quando escrevi o relatório eu tava trabalhando como pesquisador-consultor na ONG IDEC, Instituto de Defesa de Consumidores. YAMA: A gente começa depois da vinheta. [ Termina Bio Unit] [ Vinheta Oxigênio ] [ Começa Documentary] YAMA: Você já deve ter ouvido na cobertura midiática sobre datacenters a formulação que te diz quantos litros de água cada pergunta ao chatGPT gasta. Mas a gente aqui não gosta muito dessa abordagem. Entre outros motivos, porque ela reduz o problema dos impactos socioambientais das IA a uma questão de consumo individual. E isso é um erro tanto político como factual. Calcular quanta água gasta cada pergunta feita ao ChatGPT tira a responsabilidade das empresas e a transfere aos usuários, escondendo a verdadeira escala do problema. Mesmo que o consumo individual cresça de modo acelerado e explosivo, ele sempre vai ser uma pequena fração do problema. Data centers operam em escala industrial, computando quantidades incríveis de dados para treinar modelos e outros serviços corporativos. Um único empreendimento pode consumir em um dia mais energia do que as cidades que os abrigam consomem ao longo de um mês. DAMNY: Nos habituamos a imaginar a inteligência artificial como uma “nuvem” etérea, mas, na verdade, ela só existe a partir de data centers monstruosos que consomem quantidades absurdas de recursos naturais. Os impactos sociais e ambientais são severos. Data centers são máquinas de consumo de energia, água e terra, e criam poluição do ar e sonora, num modelo que reforça velhos padrões de racismo ambiental. O desenvolvimento dessas infraestruturas frequentemente acontece à margem das comunidades afetadas, refazendo a cartilha global da injustiça ambiental. Ao seguir suas redes, perceberemos seus impactos em rios, no solo, no ar, em territórios indígenas e no crescente aumento da demanda por minerais críticos e, por consequência, de práticas minerárias profundamente destrutivas. YAMA: De acordo com a pesquisadora Tamara Kneese, diretora do programa de Clima, Tecnologia e Justiça do instituto de pesquisa Data & Society, com quem conversamos, essa infraestrutura está criando uma nova forma de colonialismo tecnológico. Os danos ambientais são frequentemente direcionados para as comunidades mais vulneráveis, de zonas rurais às periferias dos grandes centros urbanos, que se tornam zonas de sacrifício para o progresso dessa indústria. DAMNY: Além disso, a crescente insatisfação das comunidades do Norte Global com os data centers tem provocado o efeito colonial de uma terceirização dessas estruturas para o Sul Global. E o Brasil não apenas não é exceção como parece ser um destino preferencial por sua alta oferta de energia limpa. [pausa] E com o aval do governo federal, que acaba de publicar uma medida provisória chamada REDATA, cujo objetivo é atrair data centers ao Brasil com isenção fiscal e pouquíssimas responsabilidades. [ Termina Documentary] [tom baixo ] VOICE OVER: BLOCO 1 – O QUE SÃO DATA CENTERS? YAMA: Pra entender o que são data centers, a gente precisa antes de tudo de entender que a inteligência artificial não é meramente uma nuvem etérea que só existe virtualmente. Foi assim que a gente começou nossa conversa com a pesquisadora estadunidense Tamara Kneese. Ela é diretora do programa de Clima, Tecnologia e Justiça do instituto de pesquisa Data & Society. TAMARA: PT – BR [ Eu acho que o problema da nossa relação com a computação é que a maioria parte do tempo a gente não pensa muito sobre a materialidade dos sistemas informacionais e na cadeia de suprimentos que permitem que eles existam. Tudo que a gente faz online não depende só dos nossos aparelhos, ou dos serviços de nuvem que a gente contrata, mas de uma cadeia muito maior. De onde ver o hardware que a gente usa? Que práticas de trabalho são empregadas nessa cadeia? E então, voltando à cadeia de suprimentos, pensar sobre os materiais brutos e os minerais críticos e outras formas de extração, abusos de direitos humanos e trabalhistas que estão diretamente relacionados à produção dos materiais que precisamos pra computação em geral. ] So I think, you know, the problem with our relationship to computing is that, most of the time, we don’t really think that much about the materiality of the computing system and the larger supply chain. You know, thinking about the fact that, of course, everything we do relies not just on our own device, or the particular cloud services that we subscribe to, but also on a much larger supply chain. So, where does the hardware come from, that we are using, and what kind of labor practices are going into that? And then be, you know, further back in the supply chain, thinking about raw materials and critical minerals and other forms of extraction, and human rights abuses and labor abuses that also go into the production of the raw materials that we need for computing in general. DAMNY: A Tamara já escreveu bastante sobre como a metáfora da nuvem nos engana, porque ela dificulta que a gente enxergue a cadeia completa que envolve o processamento de tantos dados. E isso se tornou uma questão muito maior com a criação dos chatbots e das IAs generativas. YAMA: Se a pandemia já representou uma virada no aumento da necessidade de processamento de dados, quando passamos a ir à escola e ao trabalho pelo computador, o boom das IA generativas criou um aumento sem precedentes da necessidade de expandir essas cadeias. DAMNY: E na ponta da infraestrutura de todas as nuvens estão os data centers. Mais do que gerar enormes impactos sócio-ambientais, eles são as melhores formas de enxergar que o ritmo atual da expansão das IAs não poderá continuar por muito tempo, por limitações físicas. Não há terra nem recursos naturais que deem conta disso. YAMA: A gente conversou com a Cynthia Picolo, que é Diretora Executiva do LAPIN, o Laboratório de Políticas Públicas e Internet. O LAPIN tem atuado muito contra a violação de direitos na implementação de data centers no Brasil e a gente ainda vai conversar mais sobre isso. DAMNY: Uma das coisas que a Cynthia nos ajudou a entender é como não podemos dissociar as IAs dos data centers. CYNTHIA: Existe uma materialidade por trás. Existe uma infraestrutura física, que são os data centers. Então os data centers são essas grandes estruturas que são capazes de armazenar, processar e transferir esses dados, que são os dados que são os processamentos que vão fazer com que a inteligência artificial possa acontecer, possa se desenvolver, então não existe sem o outro. Então falar de IA é falar de Datacenter. Então não tem como desassociar. YAMA: Mas como é um datacenter? A Tamara descreve o que podemos ver em fotos e vídeos na internet. TAMARA: [ Sim, de modo geral, podemos dizer que os data centers são galpões gigantes de chips, servidores, sistemas em redes e quando você olha pra eles, são todos muitos parecidos, prédios quadrados sem nada muito interessante. Talvez você nem saiba que é um data center se não observar as luzes e perceber que é uma estrutura enorme sem pessoas, sem trabalhadores. ] Yeah, so, you know, essentially, they’re like giant warehouses of chips, of servers, of networked systems, and, you know, they look like basically nondescript square buildings, very similar. And you wouldn’t really know that it’s a data center unless you look at the lighting, and you kind of realize that something… like, it’s not inhabited by people or workers, really. DAMNY: No próximo bloco a gente tenta resumir os principais problemas socioambientais que os data centers já causam e irão causar com muita mais intensidade no futuro. [tom baixo ] VOICE OVER: BLOCO 2 – A ENORME LISTA DE PROBLEMAS YAMA: O consumo de energia é provavelmente o problema mais conhecido dos data centers e das IAs. Segundo dados da Agência Internacional de Energia, a IEA, organização internacional da qual o Brasil faz parte, a estimativa para o ano de 2024 é que os data centers consumiram cerca de 415 TWh. A cargo de comparação, segundo a Empresa de Pesquisa Energética, instituto de pesquisa público associado ao Ministério das Minas e Energia, o Brasil consumiu no ano de 2024 cerca de 600 TWh. DAMNY: Segundo o mesmo relatório da Agência Internacional de Energia, a estimativa é que o consumo de energia elétrica por datacenters em 2030 vai ser de pelo menos 945 TWh, o que representaria 3% de todo consumo global projetado. Quando a gente olha pras estimativas de outras fontes, contudo, podemos dizer que essas são projeções até conservadoras. Especialmente considerando o impacto da popularização das chamadas LLM, ou grandes modelos de linguagem – aqueles YAMA: Ou seja, mesmo com projeções conservadoras, os data centers do mundo consumiriam em 2030, daqui a menos de cinco anos, cerca de 50% a mais de energia que o Brasil inteiro consome hoje. Segundo a IEA, em 2030 o consumo global de energia elétrica por data centers deve ser equivalente ao consumo da Índia, o país mais populoso do mundo. E há situações locais ainda mais precárias. DAMNY: É o caso da Irlanda. Segundo reportagem do New York Times publicada em outubro passado, espera-se que o consumo de energia elétrica por data centers por lá represente pelo menos 30% do consumo total do país nos próximos anos. Mas porquê os datacenters consomem tanta energia? TAMARA: [ Então, particularmente com o tipo de IA que as empresas estão investindo agora, há uma necessidade de chips e GPUs muito mais poderosos, de modo que os data centers também são sobre prover energia o suficiente pra todo esse poder computacional que demandam o treinamento e uso de grandes modelos de linguagem. Os data centers são estruturas incrivelmente demandantes de energia e água. A água em geral serve para resfriar os servidores, então tem um número considerável de sistemas de cooling que usam água. Além disso tudo, você também precisa de fontes alternativas de energia, porque algumas vezes, uma infraestrutura tão demandante de energia precisa recorrer a geradores para garantir que o data center continue funcionando caso haja algum problema na rede elétrica. ] So, you know, particularly with the kinds of AI that companies are investing in right now, there’s a need for more powerful chips, GPUs, and so Data centers are also about providing enough energy and computational power for these powerful language models to be trained and then used. And so the data center also, you know, in part because it does require so much energy, and it’s just this incredibly energy-intensive thing, you also need water. And the water comes from having to cool the servers, and so… So there are a number of different cooling systems that use water. And then on top of that, you also need backup energy sources, so sometimes, because there’s such a draw on the power grid, you have to have backup generators to make sure that the data center can keep going if something happens with the grid. YAMA: E aqui a gente começa a entender o tamanho do problema. Os data centers são muitas vezes construídos em lugares que já sofrem com infraestruturas precárias de eletricidade e com a falta de água potável. Então eles criam problemas de escassez onde não havia e aprofundam essa escassez em locais onde isso já era uma grande questão – como a região metropolitana de Fortaleza sobre a qual falaremos no próximo episódio, que está em vias de receber um enorme data center do Tiktok. DAMNY: É o que também relatam os moradores de Querétaro, no México, que vivem na região dos data centers da Microsoft. A operação dos data centers da Microsoft gerou uma crise sem precedentes, com quedas frequentes de energia e o interrompimento do abastecimento de água que muitas vezes duram semanas. Os data-centers impactaram de tal forma as comunidades que escolas cancelaram aulas e, indiretamente, foram responsáveis por uma crise de gastroenterite entre crianças. YAMA: E isso nos leva pro segundo ponto. O consumo de água, minerais críticos e outros recursos naturais. TAMARA: [O problema da energia tem recebido mais atenção, porque é uma fonte de ansiedade também. Pensar sobre o aumento da demanda de energia em tempos em que supostamente estaríamos transicionando para deixar de usar energias fósseis, o que obviamente pode ter efeitos devastadores. Mas eu acredito que num nível mais local, o consumo de água é mais relevante. Nós temos grandes empresas indo às áreas rurais do México, por exemplo, e usando toda a água disponível e basicamente deixando as pessoas sem água. E isso é incrivelmente problemático. Então isso acontece em áreas que já tem problemas de abastecimento de água, onde as pessoas já não tem muito poder de negociação com as empresas. Não têm poder político pra isso. São lugares tratados como zonas de sacrifício, algo que já vimos muitas vezes no mundo, especialmente em territórios indígenas. Então as consequências são na verdade muito maiores do que só problemas relacionados à energia. ] I think the energy problem has probably gotten the most attention, just because it is a source of anxiety, too, so thinking about, you know, energy demand at a time when we’re supposed to be transitioning away from fossil fuels. And clearly, the effects that that can have will be devastating. But I think on a local level, things like the water consumption can matter more. So, you know, if we have tech companies moving into rural areas in Mexico and, you know, using up all of their water and basically preventing people in the town from having access to water. That is incredibly problematic. So I think, you know, in water-stressed areas and areas where the people living in a place don’t have as much negotiating power with the company. Don’t have as much political power, and especially if places are basically already treated as sacrifice zones, which we’ve seen repeatedly many places in the world, with Indigenous land in particular, you know, I think the consequences may go far beyond just thinking about, you know, the immediate kind of energy-related problems. YAMA: Existem pelo menos quatro fins que tornam os data centers máquinas de consumir água. O mais direto e local é a água utilizada na refrigeração de todo equipamento que ganha temperatura nas atividades de computação, o processo conhecido como cooling. Essa prática frequentemente utiliza água potável. Apesar de já ser extremamente relevante do ponto de vista de consumo, essa é apenas uma das formas de consumo abundante de água. DAMNY: Indiretamente, os data centers também consomem a água relacionada ao seu alto consumo de energia, em especial na geração de energia elétrica em usinas hidrelétricas e termelétricas. Também atrelada ao consumo energético, está o uso nas estações de tratamento de água, que visam tratar a água com resíduos gerada pelo data center para tentar reduzir a quantidade de água limpa utilizada. YAMA: Por fim, a cadeia de suprimentos de chips e servidores que compõem os data centers requer água ultrapura e gera resíduos químicos. Ainda que se saiba que esse fator gera gastos de água e emissões de carbono relevantes, os dados são super obscuros, entre outros motivos, porque a maioria dos dados que temos sobre o consumo de água em data centers são fornecidos pelas próprias empresas. CYNTHIA: A água e os minérios são componentes também basilares para as estruturas de datacenter, que são basilares para o funcionamento da inteligência artificial. (…). E tem toda uma questão, como eu disse muitas vezes, captura um volume gigante de água doce. E essa água que é retornada para o ecossistema, muitas vezes não é compensada da água que foi capturada. Só que as empresas também têm uma promessa em alguns relatórios, você vai ver que elas têm uma promessa até de chegar em algum ponto para devolver cento e vinte por cento da água. Então a empresa está se comprometendo a devolver mais água do que ela capturou. Só que a realidade é o quê? É outra. Então, a Google, por exemplo, nos últimos cinco anos, reportou um aumento de cento e setenta e sete por cento do uso de água. A Microsoft mais trinta e oito e a Amazon sequer reporta o volume de consumo de água. Então uma lacuna tremenda para uma empresa desse porte, considerando todo o setor de Data centers. Mas tem toda essa questão da água, que é muito preocupante, não só por capturar e o tratamento dela e como ela volta para o meio ambiente, mas porque há essa disputa também com territórios que têm uma subsistência muito específica de recursos naturais, então existe uma disputa aí por esse recurso natural entre comunidade e empreendimento. DAMNY: Nessa fala da Cynthia a gente observa duas coisas importantes: a primeira é que não existe data center sem água para resfriamento, de modo que o impacto local da instalação de um empreendimento desses é uma certeza irrefutável. E é um dano contínuo. Enquanto ele estiver em operação ele precisará da água. É como se uma cidade de grande porte chegasse de repente, demandando uma quantidade de água e energia que o local simplesmente não tem para oferecer. E na hora de escolher entre as pessoas e empreendimentos multimilionários, adivinha quem fica sem água e com a energia mais cara? YAMA: A segunda coisa importante que a Cynthia fala é quando ela nos chama a atenção sobre a demanda por recursos naturais. Nós sabemos que recursos naturais são escassos. Mais do que isso, recursos naturais advindos da mineração têm a sua própria forma de impactos sociais e ambientais, o que vemos frequentemente na Amazônia brasileira. O que acontecerá com os data centers quando os recursos naturais locais já não forem suficientes para seu melhor funcionamento? Diante de uma computação que passa por constante renovação pela velocidade da obsolescência, o que acontece com o grande volume de lixo eletrônico gerado por data centers? Perguntas que não têm resposta. DAMNY: A crise geopolítica em torno dos minerais conhecidos como terra-rara mostra a complexidade política e ambiental do futuro das IA do ponto de vista material e das suas cadeias de suprimento. No estudo feito pelo LAPIN, a Cynthia nos disse que considera que esse ponto do aumento da demanda por minerais críticos que as IA causam é um dos pontos mais opacos nas comunicações das grandes empresas de tecnologia sobre o impacto de seus data centers. CYNTHIA: E outro ponto de muita, muita lacuna, que eu acho que do nosso mapeamento, desses termos mais de recursos naturais. A cadeia de extração mineral foi o que mais foi opaco, porque, basicamente, as empresas não reportam nada sobre essa extração mineral e é muito crítico, porque a gente sabe que muitos minérios vêm também de zonas de conflito. Então as grandes empresas, pelo menos as três que a gente mapeou, elas têm ali um trechinho sobre uma prestação de contas da cadeia mineral. Tudo que elas fazem é falar que elas seguem um framework específico da OCDE sobre responsabilização. YAMA: Quando as empresas falam de usar energias limpas e de reciclar a água utilizada, eles estão se desvencilhando das responsabilidades sobre seus datacenters. Energia limpa não quer dizer ausência de impacto ambiental. Pras grandes empresas, as fontes de energia limpa servem para gerar excedente e não para substituir de fato energias fósseis. Você pode ter um data center usando majoritariamente energia solar no futuro, mas isso não muda o fato de que ele precisa funcionar 24/7 e as baterias e os geradores a diesel estarão sempre lá. Além disso, usinas de reciclagem de água, fazendas de energia solar e usinas eólicas também têm impactos socioambientais importantes. O uso de recursos verdes complexifica o problema de identificar os impactos locais e responsabilidades dos data centers, mas não resolve de nenhuma forma os problemas de infraestrutura e de fornecimento de água e energia causados pelos empreendimentos. DAMNY: É por isso que a gente alerta pra não comprar tão facilmente a história de que cada pergunta pro chatGPT gasta x litros de água. Se você não perguntar nada pro chatGPT hoje, ou se fizer 1000 perguntas, não vai mudar em absolutamente nada o alto consumo de água e os impactos locais destrutivos dos data centers que estão sendo instalados a todo vapor em toda a América Latina. A quantidade de dados e de computação que uma big tech usa para treinar seus modelos, por exemplo, jamais poderá ser equiparada ao consumo individual de chatbots. É como comparar as campanhas que te pedem pra fechar a torneira ao escovar os dentes, enquanto o agro gasta em minutos água que você não vai gastar na sua vida inteira. Em resumo, empresas como Google, Microsoft, Meta e Amazon só se responsabilizam pelos impactos diretamente causados por seus data centers e, mesmo assim, é uma responsabilização muito entre aspas, à base de greenwashing. Você já ouviu falar de greenwashing? CYNTHIA: Essa expressão em inglês nada mais é do que a tradução literal, que é o discurso verde. (…)É justamente o que a gente está conversando. É justamente quando uma empresa finge se preocupar com o meio ambiente para parecer sustentável, mas, na prática, as ações delas não trazem esses benefícios reais e, pelo contrário, às vezes trazem até danos para o meio ambiente. Então, na verdade, é uma forma até de manipular, ou até mesmo enganar as pessoas, os usuários daqueles sistemas ou serviços com discursos e campanhas com esses selos verdes, mas sem comprovar na prática. YAMA: Nesse contexto, se torna primordial que a gente tenha mais consciência de toda a infraestrutura material que está por trás da inteligência artificial. Como nos resumiu bem a Tamara: TAMARA: [ Eu acredito que ter noção da infraestrutura completa que envolve a cadeia da IA realmente ajuda a entender a situação. Mesmo que você esteja usando, supostamente, energia renovável para construir e operar um data center, você ainda vai precisar de muitos outros materiais, chips, minerais e outras coisas com suas próprias cadeias de suprimento. Ou seja, independente da forma de energia utilizada, você ainda vai causar dano às comunidades e destruição ambiental. ] But that… I think that is why having a sense of the entire AI supply chain is really helpful, just in terms of thinking about, you know, even if you’re, in theory, using renewable energy to build a data center, you still are relying on a lot of other materials, including chips, including minerals, and other things that. (…) We’re still, you know, possibly going to be harming communities and causing environmental disruption. [ tom baixo ] YAMA: Antes de a gente seguir pro último bloco, eu queria só dizer que a entrevista completa com a Dra. Tamara Kneese foi bem mais longa e publicada na íntegra no blog do GEICT. O link para a entrevista tá na descrição do episódio, mas se você preferir pode ir direto no bloco do GEICT. [ tom baixo ] VOICE OVER: BLOCO 3 – PROBLEMAS GLOBAIS, PROBLEMAS LOCAIS YAMA: Mesmo conhecendo as cadeias, as estratégias de greenwashing trazem um grande problema à tona, que é uma espécie de terceirização das responsabilidades. As empresas trazem medidas compensatórias que não diminuem em nada o impacto local dos seus data centers. Então tem uma classe de impactos que são globais, como as emissões de carbono e o aumento da demanda por minerais críticos, por exemplo. E globais no sentido de que eles são parte relevante dos impactos dos data centers, mas não estão impactando exatamente nos locais onde foram construídos. CYNTHIA: Google, por exemplo, nesse recorte que a gente fez da pesquisa dos últimos cinco anos, ela simplesmente reportou um aumento de emissão de carbono em setenta e três por cento. Não é pouca coisa. A Microsoft aumentou no escopo dois, que são as emissões indiretas, muito por conta de data centers, porque tem uma diferenciação por escopo, quando a gente fala de emissão de gases, a Microsoft, nesse período de cinco anos, ela quadruplicou o tanto que ela tem emitido. A Amazon aumentou mais de trinta por cento. Então a prática está mostrando que essas promessas estão muito longe de serem atingidas. Só que aí entra um contexto mais de narrativa. Por que elas têm falado e prometido a neutralidade de carbono? Porque há um mecanismo de compensação. (…) Então elas falam que estão correndo, correndo para atingir essa meta de neutralidade de carbono, mas muito por conta dos instrumentos de compensação, compensação ou de crédito de carbono ou, enfim, para uso de energias renováveis. Então se compra esse certificado, se fazem esses contratos, mas, na verdade, não está tendo uma redução de emissão. Está tendo uma compensação. (…) Essa compensação é um mecanismo financeiro, no final do dia. Porque, quando você, enquanto empresa, trabalha na compensação dos seus impactos ambientais e instrumentos contratuais, você está ignorando o impacto local. Então, se eu estou emitindo impactando aqui o Brasil, e estou comprando crédito de carbono em projetos em outra área, o impacto local do meu empreendimento está sendo ignorado. YAMA: E os impactos materiais locais continuam extremamente relevantes. Além do impacto nas infraestruturas locais de energia e de água sobre as quais a gente já falou, há muitas reclamações sobre a poluição do ar gerada pelos geradores, as luzes que nunca desligam e até mesmo a poluição sonora. A Tamara nos contou de um caso curioso de um surto de distúrbios de sono e de enxaqueca que tomou regiões de data centers nos Estados Unidos. TAMARA: [ Uma outra coisa que vale ser lembrada: as pessoas que vivem perto dos data centers tem nos contado que eles são super barulhentos, eles também relatam a poluição visual causada pelas luzes e a poluição sonora. Foi interessante ouvir de comunidades próximas a data centers de mineração de criptomoedas, por exemplo, que os moradores começaram a ter enxaquecas e distúrbios de sono por viverem próximos das instalações. E além de tudo isso, ainda tem a questão da poluição do ar, que é visível a olho nu. Há muitas partículas no ar onde há geradores movidos a diesel para garantir que a energia esteja sempre disponível. ] And the other thing is, you know, for people who live near them, they’re very loud, and so if you talk to people who live near data centers, they will talk about the light pollution, the noise pollution. And it’s been interesting, too, to hear from communities that are near crypto mining facilities, because they will complain of things like migraine headaches and sleep deprivation from living near the facilities. And, you know, the other thing is that the air pollution is quite noticeable. So there’s a lot of particulate matter, particularly in the case of using diesel-fueled backup generators as an energy stopgap. DAMNY: E do ponto de vista dos impactos locais, há um fator importantíssimo que não pode ser esquecido: território. Data centers podem ser gigantes, mas ocupam muito mais espaço que meramente seus prédios, porque sua cadeia de suprimentos demanda isso. Como a água e a energia chegarão até os prédios? Mesmo que sejam usados fontes renováveis de energia, onde serão instaladas as fazendas de energia solar ou as usinas de energia eólica e de tratamento de água? Onde a água contaminada e/ou tratada será descartada? Quem vai fiscalizar? YAMA: E essa demanda sem fim por território esbarra justamente nas questões de racismo ambiental. Porque os territórios que são sacrificados para que os empreendimentos possam funcionar, muito frequentemente, são onde vivem povos originários e populações marginalizadas. Aqui percebemos que a resistência local contra a instalação de data centers é, antes de qualquer coisa, uma questão de justiça ambiental. É o caso de South Memphis nos Estados Unidos, por exemplo. TAMARA: [ Pensando particularmente sobre os tipos de danos causados pelos data centers, não é somente a questão da conta de energia ficar mais cara, ou quantificar a quantidade de energia e água gasta por data centers específicos. A verdadeira questão, na minha opinião, é a relação que existe entre esses danos socioambientais, danos algorítmicos e o racismo ambiental e outras formas de impacto às comunidades que lidam com isso a nível local. Especialmente nos Estados Unidos, com todo esse histórico de supremacia branca e a falta de direitos civis, não é coincidência que locais onde estão comunidades negras, por exemplo, sejam escolhidos como zonas de sacrifício. As comunidades negras foram historicamente preferenciais para todo tipo de empreendimento que demanda sacrificar território, como estradas interestaduais, galpões da Amazon… quer dizer, os data centers são apenas a continuação dessa política histórica de racismo ambiental. E tudo isso se soma aos péssimos acordos feitos a nível local, onde um prefeito e outras lideranças governamentais pensam que estão recebendo algo de grande valor econômico. Em South Memphis, por exemplo, o data center é da xAI. Então você para pra refletir como essa plataforma incrivelmente racista ainda tem a audácia de poluir terras de comunidades negras ainda mais ] I think, the way of framing particular kinds of harm, so, you know, it’s not just about, you know, people’s energy bills going up, or, thinking about how we quantify the energy use or the water use of particular data centers, but really thinking about the relationship between a lot of those social harms and algorithmic harms and the environmental racism and other forms of embodied harms that communities are dealing with on that hyper-local level. And, you know, in this country, with its history of white supremacy and just general lack of civil rights, you know, a lot of the places where Black communities have traditionally been, tend to be, you know, the ones sacrificed for various types of development, like, you know, putting up interstates, putting up warehouses for Amazon and data centers are just a continuation of the what was already happening. And then you have a lot of crooked deals on the local level, where, you know, maybe a mayor and other local officials think that they’re getting something economically of value. In South Memphis, the data center is connected to x AI. And so thinking about this platform that is so racist and so incredibly harmful to Black communities, you know, anyway, and then has the audacity to actually pollute their land even more. DAMNY: Entrando na questão do racismo ambiental a gente se encaminha para o nosso segundo episódio, onde vamos tentar entender como o Brasil se insere na questão dos data centers e como diferentes setores da população estão se organizando para resistir. Antes de encerrar esse episódio, contudo, a gente traz brevemente pra conversa dois personagens que vão ser centrais no próximo episódio. YAMA: Eles nos ajudam a compreender como precisamos considerar a questão dos territórios ao avaliar os impactos. Uma dessas pessoas é a Andrea Camurça, do Instituto Terramar, que está lutando junto ao povo Anacé pelo direito de serem consultados sobre a construção de um data center do TIKTOK em seus territórios. Eu trago agora um trechinho dela falando sobre como mesmo medidas supostamente renováveis se tornam violações territoriais num contexto de racismo ambiental. ANDREA: A gente recebeu notícias agora, recentemente, inclusive ontem, que está previsto um mega empreendimento solar que vai ocupar isso mais para a região do Jaguaribe, que vai ocupar, em média, de equivalente a seiscentos campos de futebol. Então, o que isso representa é a perda de terra. É a perda de água. É a perda do território. É uma diversidade de danos aos povos e comunidades tradicionais que não são reconhecidos, são invisibilizados. Então é vendido como território sem gente, sendo que essas energias chegam dessa forma. Então, assim a gente precisa discutir sobre energias renováveis. A gente precisa discutir sobre soberania energética. A gente precisa discutir sobre soberania digital, sim, mas construída a partir da necessidade do local da soberania dessas populações. DAMNY: A outra pessoa que eu mencionei é uma liderança Indígena, o cacique Roberto Anacé. Fazendo uma ótima conexão que nos ajuda a perceber como os impactos globais e locais dos data centers estão conectados, ele observa como parecemos entrar num novo momento do colonialismo, onde a soberania digital e ambiental do Brasil volta a estar em risco, indo de encontro à violação de terras indígenas. CACIQUE ROBERTO: Há um risco para a questão da biodiversidade, da própria natureza da retirada da água, do aumento de energia, mas também não somente para o território da Serra, mas para todos que fazem uso dos dados. Ou quem expõe esses dados. Ninguém sabe da mão de quem vai ficar, quem vai controlar quem vai ordenar? E para que querem essa colonização? Eu chamo assim que é a forma que a gente tem essa colonização de dados. Acredito eu que a invasão do Brasil em mil e quinhentos foi de uma forma. Agora nós temos a invasão de nossas vidas, não somente para os indígenas, mas de todos, muitas vezes que fala muito bem, mas não sabe o que vai acontecer depois que esses dados estão guardados. Depois que esses dados vão ser utilizados, para que vão ser utilizados, então esses agravos. Ele é para além do território indígena na série. [ tom baixo ] [ Começa Bio Unit ] YAMA: A pesquisa, entrevistas e apresentação desse episódio foi feita pelo Damny Laya e por mim, Yama Chiodi. Eu também fiz o roteiro e a produção. Quem narrou a tradução das falas da Tamara foi Mayra Trinca. O Oxigênio é um podcast produzido pelos alunos do Laboratório de Estudos Avançados em Jornalismo da Unicamp e colaboradores externos. Tem parceria com a Secretaria Executiva de Comunicação da Unicamp e apoio do Serviço de Auxílio ao Estudante, da Unicamp. Além disso, contamos com o apoio da FAPESP, que financia bolsas como a que nos apoia neste projeto de divulgação científica. DAMNY: A lista completa de créditos para os sons e músicas utilizados você encontra na descrição do episódio. Você encontra todos os episódios no site oxigenio.comciencia.br e na sua plataforma preferida. No Instagram e no Facebook você nos encontra como Oxigênio Podcast. Segue lá pra não perder nenhum episódio! Aproveite para deixar um comentário. [ Termina Bio Unit ] [ Vinheta Oxigênio ] Créditos: Aerial foi composta por Bio Unit; Documentary por Coma-Media. Ambas sob licença Creative Commons. Os sons de rolha e os loops de baixo são da biblioteca de loops do Garage Band. Roteiro, produção: Yama Chiodi Pesquisa: Yama Chiodi, Damny Laya Narração: Yama Chiodi, Danny Laya, Mayra Trinca Entrevistados: Tamara Kneese, Cynthia Picolo, Andrea Camurça e Cacique Roberto Anacé __________ Descendo a toca do coelho da IA: Data Centers e os Impactos Materiais da “Nuvem” – Uma entrevista com Tamara Kneese: https://www.blogs.unicamp.br/geict/2025/11/06/descendo-a-toca-do-coelho-da-ia-data-centers-e-os-impactos-materiais-da-nuvem-uma-entrevista-com-tamara-kneese/ Não somos quintal de data centers: Um estudo sobre os impactos socioambientais e climáticos dos data centers na América Latina: https://idec.org.br/publicacao/nao-somos-quintal-de-data-centers Outras referências e fontes consultadas: Relatórios técnicos e dados oficiais: IEA (2025), Energy and AI, IEA, Paris https://www.iea.org/reports/energy-and-ai, Licence: CC BY 4.0 “Inteligência Artificial e Data Centers: A Expansão Corporativa em Tensão com a Justiça Socioambiental”. Lapin. https://lapin.org.br/2025/08/11/confira-o-relatorio-inteligencia-artificial-e-data-centers-a-expansao-corporativa-em-tensao-com-a-justica-socioambiental/ Estudo de mercado sobre Power & Cooling de Data Centers. DCD – DATA CENTER DYNAMICS.https://media.datacenterdynamics.com/media/documents/Report_Power__Cooling_2025_PT.pdf Pílulas – Impactos ambientais da Inteligência Artificial. IPREC. https://ip.rec.br/publicacoes/pilulas-impactos-ambientais-da-inteligencia-artificial/ Policy Brief: IA, data centers e os impactos ambientais. IPREC https://ip.rec.br/wp-content/uploads/2025/05/Policy-Paper-IA-e-Data-Centers.pdf MEDIDA PROVISÓRIA Nº 1.318, DE 17 DE SETEMBRO DE 2025 https://www.in.gov.br/en/web/dou/-/medida-provisoria-n-1.318-de-17-de-setembro-de-2025-656851861 Infográfico sobre minerais críticos usados em Data Centers do Serviço de Geologia do Governo dos EUA https://www.usgs.gov/media/images/key-minerals-data-centers-infographic Notícias e reportagens: From Mexico to Ireland, Fury Mounts Over a Global A.I. Frenzy. Paul Mozur, Adam Satariano e Emiliano Rodríguez Mega. The New York Times, 20/10/2025. https://www.nytimes.com/2025/10/20/technology/ai-data-center-backlash-mexico-ireland.html Movimentos pedem ao MP fim de licença de data center no CE. Maristela Crispim, EcoNordeste. 25/08/2025. https://agenciaeconordeste.com.br/sustentabilidade/movimentos-pedem-ao-mp-fim-de-licenca-de-data-center-no-ce/#:~:text=’N%C3%A3o%20somos%20contra%20o%20progresso’&text=Para%20o%20cacique%20Roberto%20Anac%C3%A9,ao%20meio%20ambiente%E2%80%9D%2C%20finaliza. ChatGPT Is Everywhere — Why Aren’t We Talking About Its Environmental Costs? Lex McMenamin. Teen Vogue. https://www.teenvogue.com/story/chatgpt-is-everywhere-environmental-costs-oped Data centers no Nordeste, minérios na África, lucros no Vale do Silício. Le Monde Diplomatique, 11 jun. 2025. Accioly Filho. https://diplomatique.org.br/data-centers-no-nordeste-minerios-na-africa-lucros-no-vale-do-silicio/. The environmental footprint of data centers in the United States. Md Abu Bakar Siddik et al 2021 Environ. Res. Lett. 16064017: https://iopscience.iop.org/article/10.1088/1748-9326/abfba1 Tecnología en el desierto – El debate por los data centers y la crisis hídrica en Uruguay. MUTA, 30 nov. Soledad Acunã https://mutamag.com/cyberpunk/tecnologia-en-el-desierto/. Acesso em: 17 set. 2025. Las zonas oscuras de la evaluación ambiental que autorizó “a ciegas” el megaproyecto de Google en Cerrillos. CIPER Chile, 25 maio 2020. https://www.ciperchile.cl/2020/05/25/las-zonas-oscuras-de-la-evaluacion-ambiental-que-autorizo-aciegas-el-megaproyecto-de-google-en-cerrillos/. Acesso em: 17 set. 2025. Thirsty data centres spring up in water-poor Mexican town. Context, 6 set. 2024. https://www.context.news/ai/thirsty-data-centres-spring-up-in-water-poor-mexican-town BNDES lança linha de R$ 2 bilhões para data centers no Brasil. https://agenciadenoticias.bndes.gov.br/industria/BNDES-lanca-linha-de-R$-2-bilhoes-para-data-centersno-Brasil/. Los centros de datos y sus costos ocultos en México, Chile, EE UU, Países Bajos y Sudáfrica. WIRED, 29 maio 2025. Anna Lagos https://es.wired.com/articulos/los-costos-ocultos-del-desarrollo-de-centros-de-datos-en-mexico-chile-ee-uu-paises-bajos-y-sudafrica Big Tech's data centres will take water from world's driest areas. Eleanor Gunn. SourceMaterial, 9 abr. 2025. https://www.source-material.org/amazon-microsoft-google-trump-data-centres-water-use/ Indígenas pedem que MP atue para derrubar licenciamento ambiental de data center do TikTok. Folha de S.Paulo, 26 ago. 2025. https://www1.folha.uol.com.br/mercado/2025/08/indigenas-pedem-que-mp-atue-para-derrubar-licenciamento-ambiental-de-data-center-do-tiktok.shtml The data center boom in the desert. MIT Technology Review https://www.technologyreview.com/2025/05/20/1116287/ai-data-centers-nevada-water-reno-computing-environmental-impact/ Conferências, artigos acadêmicos e jornalísticos: Why are Tech Oligarchs So Obsessed with Energy and What Does That Mean for Democracy? Tamara Kneese. Tech Policy Press. https://www.techpolicy.press/why-are-tech-oligarchs-so-obsessed-with-energy-and-what-does-that-mean-for-democracy/ Data Center Boom Risks Health of Already Vulnerable Communities. Cecilia Marrinan. Tech Policy Press. https://www.techpolicy.press/data-center-boom-risks-health-of-already-vulnerable-communities/ RARE/EARTH: The Geopolitics of Critical Minerals and the AI Supply Chain. https://www.youtube.com/watch?v=GxVM3cAxHfg Understanding AI with Data & Society / The Environmental Costs of AI Are Surging – What Now? https://www.youtube.com/watch?v=W4hQFR8Z7k0 IA e data centers: expansão corporativa em tensão com justiça socioambiental. Camila Cristina da Silva, Cynthia Picolo G. de Azevedo. https://www.jota.info/opiniao-e-analise/colunas/ia-regulacao-democracia/ia-e-data-centers-expansao-corporativa-em-tensao-com-justica-socioambiental LI, P.; YANG, J.; ISLAM, M. A.; REN, S. Making AI Less “Thirsty”: Uncovering and Addressing the Secret Water Footprint of AI Models. arXiv, 2304.03271, 26 mar. 2025. Disponível em: https://doi.org/10.48550/arXiv.2304.03271 LIU, Y.; WEI, X.; XIAO, J.; LIU, Z.;XU, Y.; TIAN, Y. Energy consumption and emission mitigation prediction based on data center traffic and PUE for global data centers. Global Energy Interconnection, v. 3, n.3, p. 272-282, 3 jun. 2020. https://doi.org/10.1016/j.gloei.2020.07.008 SIDDIK, M. A. B.; SHEHABI, A.; MARSTON, L. The environmental footprint of data centers in the United States. Environmental Research Letters, v. 16, n. 6, 21 maio 2021. https://doi.org/10.1088/1748-9326/abfba1 Las Mentiras de Microsoft en Chile: Una Empresa No tan Verde. Por Rodrigo Vallejos de Resistencia Socioambiental de Quilicura. Revista De Frente, 18 mar. 2022. https://www.revistadefrente.cl/las-mentiras-de-microsoft-en-chile-una-empresa-no-tan-verde-porrodrigo-vallejos-de-resistencia-socioambiental-de-quilicura/. Acesso em: 17 set. 2025.

AI is becoming ubiquitous in our lives. It shapes how we work, play, interact, create, and even manage our health—and this is only the beginning. To understand where we are and where we might go, we first need to understand how we got here. By tracing the evolving nature of machine intelligence, we can appreciate how today's AI differs from its past and how it is likely to evolve. With that in mind, we can begin to ask the big questions: When should we trust AI over human judgment? How should we govern its development? How will it change what it means to be human? And what roles will humans play in the future of work?To help us through this journey, I'm delighted to welcome back to TRIUM Connects Professor Vasant Dhar, the Robert A. Miller Professor at NYU's Stern School of Business and Professor of Data Science at NYU. Vasant is one of the world's leading thinkers on the impact of AI on society. He was present at the birth of AI and has been involved in every step of its evolution—both as an entrepreneur and as a scholar. He also hosts the acclaimed podcast Brave New World, which explores how machines are transforming humanity in the post-COVID era.In this episode, we discuss his newest book, Thinking With Machines: The Brave New World of AI. It's a remarkable hybrid: part autobiography, tracing how his professional life has intertwined with the development of AI; part user's guide, offering a lucid framework for deciding when to trust machines over human control; and part deep dive into the philosophical and policy implications of creating an alien intelligence.It was a real pleasure to welcome Vasant back onto the show. I learned a lot during our conversation, and I hope you will enjoy it as much as I did.CitationsDawid A, LeCun Y. Introduction to Latent Variable Energy-Based Models: A Path Towards Autonomous Machine Intelligence. arXiv. June 5, 2023.Dennett DC. Intentional systems. J Philos. 1971;68(4):87-106.Dhar V. Thinking With Machines: The Brave New World of AI. Galloway S, foreword. Hoboken, NJ: Wiley; 2025.Frank, R. H., & Cook, P. J. The winner-take-all society: Why the few at the top get so much more than the rest of us. Penguin Books; 1995.Ganguli D, Askell A, Henighan T, et al. Alignment faking in large language models. arXiv. December 20, 2024.Harari YN. Nexus: A Brief History of Information Networks from the Stone Age to AI. New York, NY: Random House; 2024.Kauffmann J, Dippel J, Ruff L, et al. Explainable AI reveals Clever Hans effects in unsupervised learning models. Nat Mach Intell. 2025;7:1–10.Pearl J, Mackenzie D. The Book of Why: The New Science of Cause and Effect. New York, NY: Basic Books; 2018.Pfungst O. Clever Hans (The Horse of Mr. Von Osten): A Contribution to Experimental Animal and Human Psychology. Rahn H, trans. New York: Henry Holt; 1911.Popper KR. The Logic of Scientific Discovery. London, UK: Hutchinson; 1959Suleyman M, Bhaskar M. The Coming Wave: Technology, Power, and the Twenty-First Century's Greatest Dilemma. New York, NY: Crown; 2023.Yudkowsky E, Soares N. If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All. New York, NY: Little, Brown and Company; 2025. Hosted on Acast. See acast.com/privacy for more information.

English Edition: WHAM, BOOM, CLAC, WHOOSH - sound can represent many things, so why not experimental data? James Trayford and Chris Harrison want to show that you can, with their project called Audio Universe. And we're going to hear some of the sound samples during my conversation with them.Links:https://www.audiouniverse.org the project's home pagehttps://github.com/james-trayford/straus The Sonification software STRAUSShttps://joss.theoj.org/papers/10.21105/joss.07875 The article on STRAUSS in the Journal of Open Source Software (JOSS)https://arxiv.org/abs/2311.16847 the ArXiv articlehttps://strauss.readthedocs.io/en/latest/ documentationSome sound/video samples:https://www.youtube.com/watch?v=RjkAoqgJvYg&t=572s Caribbean Audio Universe Tour of the Solar System, stars appearinghttps://www.youtube.com/watch?v=4jH1WNpDi10&t=1632s tour of the solar system, planet orbit sequencehttps://www.youtube.com/watch?v=Z0rsZiIqcbc Spectral Data Cubehttps://www.youtube.com/watch?v=DKZkPPhaty0 light curves with different sound designshttps://www.youtube.com/watch?v=78S0lkSc7so warming stripeshttps://soundcloud.com/audio-universe-685767042/sonification-predatorprey-cycles-goats-and-wolvesGet in touchThank you for listening! Merci de votre écoute! Vielen Dank für´s Zuhören! Contact Details/ Coordonnées / Kontakt: Email mailto:peter@code4thought.org UK RSE Slack (ukrse.slack.com): @code4thought or @piddie Bluesky: https://bsky.app/profile/code4thought.bsky.social LinkedIn: https://www.linkedin.com/in/pweschmidt/ (personal Profile)LinkedIn: https://www.linkedin.com/company/codeforthought/ (Code for Thought Profile) This podcast is licensed under the Creative Commons Licence: https://creativecommons.org/licenses/by-sa/4.0/

 Evénements Séminaire SFGP- 5 novembre : séminaire organisé par la Société Française du Génie des Procédés : Génie des procédés & Machines quantiques. https://www.sfgp.asso.fr/events/genie-des-procedes-machines-quantiques/Etat de l'art du calcul quantique par Olivier https://www.oezratty.net/Files/Conferences/Olivier%20Ezratty%20Informatique%20Quantique%20SFGP%20Nov2025.pdf  Journée Minalogic sur la photonique quantique-  Minalogic Innovation Talks – Photonique pour le quantique avait lieu à Lyon le 18 novembre et regroupait de nombreux acteurs de la photonique couvrant les capteurs quantiques, les communications quantiques et le calcul quantique. https://www.minalogic.com/evenements/minalogic-innovation-talks-photonique-pour-le-quantique/OVHcloud SummitAnnonce de la Quantum Platform avec accès à la QPU Pasqal Orion beta en pay as you go à la seconde et de Nvidia CUDA-Q au format notebook https://summit.ovhcloud.com/en/ Adopt AIAnnonce de la disponibilité de Merlin de Quandela chez OVHcloudhttps://adoptai.artefact.com/session/f83102e1-2abb-f011-8194-6045bd90aa7b/hybrid-ai-quantum-computing-the-next-cloud-revolution Symposium France Singapourchez Bpifrance à Paris les 25 et 26 novembre https://french-singaporean-quantum-symposium.com/. 2 journées coorganisées par le laboratoire CNRS Majulab de Singapour et Quantonation. A venir ·       QUEST-IS début du 1ier au 4 décembre 2025 (inscriptions). Avec les keynotes d'Alain Aspect, Pierre Rouchon, Djeylan Aktas, etc.https://conference-questis.org/·       TQCI Teratechttps://conference-questis.org/quest-is-2025/program/tqci-seminar/·       Q2B Santa Clara la seconde semaine de décembre. https://q2b.qcware.com/conference/2025-silicon-valleyActus France Avancements sur la QKD à Nice avec des travaux intéressants de l'équipe de Sébastien Tanzilli du laboratoire InPhyNi qui a opéré un réseau de QKD sur 100 km et jusqu'à une station de base permettant par la suite d'établir des communications satellite.Deployed quantum key distribution network: further, longer and more users by Nathan Lecaron, Yoann Pelet, Sébastien Tanzilli, and Olivier Alibart, arXiv, November 2025 (8 pages). Acquisition de Qperfect par BTQLe 11 novembre 2025, la société canado-taïwanaise BTQ faisait l'acquisition complète de la startup française QPerfect. https://thequantuminsider.com/2025/11/10/btq-exercises-option-to-acquire-qperfect/  Quandela annonçaient un partenariat avec Nvidia sur la simulation de leurs qubits et inauguraient Lucy au TGCC, en compagnie de celle de l'ordinateur quantique de Pasqal. Lucy est un ordinateur quantique supportant 12 qubits.https://www.quandela.com/resources/blog/quandela-and-nvidia-accelerate-spin-photon-simulation/ Partenariat SEALSQ et Quoblyhttps://quobly.io/news/quobly-and-sealsq-announce-a-collaboration-to-advance-secure-and-scalable-quantum-technologies/ Quantum Foundations  Une émission intéressante sur les quantum foundations sur France Culture avec Shane Manfield et Hypolyte Dourdent. Et puis une intervention de Philippe Grangier qui présentait sa vision dans le cycle de conférences organisé par Hervé Zwirn à l'Académie des Science, autour de l'ontologie CSM (contexte-systèmes-modalités).https://www.radiofrance.fr/franceculture/podcasts/la-science-cqfd/metaphysique-la-quantique-change-sa-physique-d-epaule-4657891 Actus à l'International Quantinuum lançait Helios, un ordinateur quantique de 98 qubits à base d'ions piégés. Helios: A 98-qubit trapped-ion quantum computer by Anthony Ransford, M.S. Allman et al, arXiv, November 2025 (25 pages).https://arxiv.org/abs/2511.05465Quantinuum Helios advances by Olivier Ezratty, novembre 2025.https://www.oezratty.net/wordpress/2025/quantinuum-helios-advances/ IBMLors de leur conférence développeur du 12 novembre 2025, IBM a fait quelques annonces confirmant celles de juin 2025, notamment autour du processeur Night Hawk qui va succéder à Heron, avec 120 qubits et une connectivité améliorée de 4+2 : 4 pour les qubits voisins et 2 pour des qubits distants, cette connectique servant à la mise en œuvre des codes de correction d'erreur non locaux de type qLDPC (gross code).https://www.ibm.com/quantum/blog/networked-quantum-computershttps://www.linkedin.com/posts/jay-gambetta-a274753a_qdc25-activity-7394551922256486400-wj7q Lancement de la Quantum Scaling AllianceQolab et HPE

Episode overviewJohn Martinis, Nobel laureate and former head of Google's quantum hardware effort, joins Sebastian Hassinger on The New Quantum Era to trace the arc of superconducting quantum circuits—from the first demonstrations of macroscopic quantum tunneling in the 1980s to today's push for wafer-scale, manufacturable qubit processors. The episode weaves together the physics of “synthetic atoms” built from Josephson junctions, the engineering mindset needed to turn them into reliable computers, and what it will take for fabrication to unlock true large-scale quantum systems.Guest bioJohn M. Martinis is a physicist whose experiments on superconducting circuits with John Clarke and Michel Devoret at UC Berkeley established that a macroscopic electrical circuit can exhibit quantum tunneling and discrete energy levels, work recognized by the 2025 Nobel Prize in Physics “for the discovery of macroscopic quantum mechanical tunnelling and energy quantisation in an electric circuit.” He went on to lead the superconducting quantum computing effort at Google, where his team demonstrated large-scale, programmable transmon-based processors, and now heads Qolab (also referred to in the episode as CoLab), a startup focused on advanced fabrication and wafer-scale integration of superconducting qubits.Martinis's career sits at the intersection of precision instrumentation and systems engineering, drawing on a scientific “family tree” that runs from Cambridge through John Clarke's group at Berkeley, with strong theoretical influence from Michel Devoret and deep exposure to ion-trap work by Dave Wineland and Chris Monroe at NIST. Today his work emphasizes solving the hardest fabrication and wiring challenges—pursuing high-yield, monolithic, wafer-scale quantum processors that can ultimately host tens of thousands of reproducible qubits on a single 300 mm wafer.Key topicsMacroscopic quantum tunneling on a chip: How Clarke, Devoret, and Martinis used a current-biased Josephson junction to show that a macroscopic circuit variable obeys quantum mechanics, with microwave control revealing discrete energy levels and tunneling between states—laying the groundwork for superconducting qubits. The episode connects this early work directly to the Nobel committee's citation and to today's use of Josephson circuits as “synthetic atoms” for quantum computing.From DC devices to microwave qubits: Why early Josephson devices were treated as low-frequency, DC elements, and how failed experiments pushed Martinis and collaborators to re-engineer their setups with careful microwave filtering, impedance control, and dilution refrigerators—turning noisy circuits into clean, quantized systems suitable for qubits. This shift to microwave control and readout becomes the through-line from macroscopic tunneling experiments to modern transmon qubits and multi-qubit gates.Synthetic atoms vs natural atoms: The contrast between macroscopic “synthetic atoms” built from capacitors, inductors, and Josephson junctions and natural atomic systems used in ion-trap and neutral-atom experiments by groups such as Wineland and Monroe at NIST, where single-atom control made the quantum nature more obvious. The conversation highlights how both approaches converged on single-particle control, but with very different technological paths and community cultures.Ten-year learning curve for devices: How roughly a decade of experiments on quantum noise, energy levels, and escape rates in superconducting devices built confidence that these circuits were “clean enough” to support serious qubit experiments, just as early demonstrations such as Yasunobu Nakamura's single-Cooper-pair box showed clear two-level behavior. This foundational work set the stage for the modern era of superconducting quantum computing across academia and industry.Surface code and systems thinking: Why Martinis immersed himself in the surface code, co-authoring a widely cited tutorial-style paper “Surface codes: Towards practical large-scale quantum computation” (Austin G. Fowler, Matteo Mariantoni, John M. Martinis, Andrew N. Cleland, Phys. Rev. A 86, 032324, 2012; arXiv:1208.0928), to translate error-correction theory into something experimentalists could build. He describes this as a turning point that reframed his work at UC Santa Barbara and Google around full-system design rather than isolated device physics.Fabrication as the new frontier: Martinis argues that the physics of decent transmon-style qubits is now well understood and that the real bottleneck is industrial-grade fabrication and wiring, not inventing ever more qubit variants. His company's roadmap targets wafer-scale integration—e.g., ~100-qubit test chips scaling toward ~20,000 qubits on a 300 mm wafer—with a focus on yield, junction reproducibility, and integrated escape wiring rather than current approaches that tile many 100-qubit dies into larger systems.From lab racks of cables to true integrated circuits: The episode contrasts today's dilution-refrigerator setups—dominated by bulky wiring and discrete microwave components—with the vision of a highly integrated superconducting “IC” where most of that wiring is brought on-chip. Martinis likens the current state to pre-IC TTL logic full of hand-wired boards and sees monolithic quantum chips as the necessary analog of CMOS integration for classical computing.Venture timelines vs physics timelines: A candid discussion of the mismatch between typical three-to-five-year venture capital expectations and the multi-decade arc of foundational technologies like CMOS and, now, quantum computing. Martinis suggests that the most transformative work—such as radically improved junction fabrication—looks slow and uncompetitive in the short term but can yield step-change advantages once it matures.Physics vs systems-engineering mindsets: How Martinis's “instrumentation family tree” and exposure to both American “build first, then understand” and French “analyze first, then build” traditions shaped his approach, and how system engineering often pushes him to challenge ideas that don't scale. He frames this dual mindset as both a superpower and a source of tension when working in large organizations used to more incremental science-driven projects.Collaboration, competition, and pre-competitive science: Reflections on the early years when groups at Berkeley, Saclay, UCSB, NIST, and elsewhere shared results openly, pushing the field forward without cut-throat scooping, before activity moved into more corporate settings around 2010. Martinis emphasizes that many of the hardest scaling problems—especially in materials and fabrication—would benefit from deeper cross-organization collaboration, even as current business constraints limit what can be shared.Papers and research discussed“Energy-Level Quantization in the Zero-Voltage State of a Current-Biased Josephson Junction” – John M. Martinis, Michel H. Devoret, John Clarke, Physical Review Letters 55, 1543 (1985). First clear observation of quantized energy levels and macroscopic quantum tunneling in a Josephson circuit, forming a core part of the work recognized by the 2025 Nobel Prize in Physics. Link: https://link.aps.org/doi/10.1103/PhysRevLett.55.1543“Quantum Mechanics of a Macroscopic Variable: The Phase Difference of a Josephson Junction” – J. Clarke et al., Science 239, 992 (1988). Further development of macroscopic quantum tunneling and wave-packet dynamics in current-biased Josephson junctions, demonstrating that a circuit-scale degree of freedom behaves as a quantum variable. Link (PDF via Cleland group):

Een beetje een populair nieuwsitem, maar goed. Voor het eerst is een CME, een Coronal Mass Ejection, waargenomen bij een andere ster dan onze zon. Wellicht biedt deze aflevering nog wat extra context, die u in andere nieuwsbronnen moest missen.First confirmed sighting of giant explosion on nearby star:https://www.eurekalert.org/news-releases/1105091Radio burst from a stellar coronal mass ejection (Arxiv):https://arxiv.org/pdf/2511.09289Radio burst from a stellar coronal mass ejection (Nature):https://www.nature.com/articles/s41586-025-09715-3Steun deze podcast en koop het album met muziek:https://henszimmermanaudio.bandcamp.com/album/really-slow-tapesAfbeelding credit: Olena Shmahalo/Callingham et al.De Zimmerman en Space podcast is gelicenseerd onder een Creative Commons CC0 1.0 licentie.http://creativecommons.org/publicdomain/zero/1.0

Une étude surprenante montre que les requêtes les plus impolies adressées à ChatGPT produisent des réponses plus précises. Une découverte contre-intuitive sur le pouvoir du ton et du langage face à l'IA.Quand l'impolitesse devient un atoutSelon une étude de l'Université de Pennsylvanie, parler sèchement à ChatGPT améliorerait ses performances. Les chercheurs Om Dobaria et Akhil Kumar ont testé cinquante questions sur le modèle GPT-4o, reformulées du ton le plus courtois au plus brutal. Résultat : les requêtes les plus rudes obtiennent 84,8 % de bonnes réponses, contre 80,8 % pour les plus polies. Une différence modeste mais réelle, détaillée dans le rapport disponible sur arXiv.Une question d'attention et de clartéLes chercheurs avancent plusieurs explications. Les formulations polies dilueraient l'attention du modèle : trop de mots périphériques comme « s'il te plaît » ou « pourrais-tu » dispersent le sens. À l'inverse, un prompt direct va droit au but : « Résous ce problème » plutôt que « Peux-tu m'aider à résoudre ce problème ? ». Cette concision renforce la focalisation de l'IA sur sa tâche.Défier l'IA, ça marcheAutre hypothèse : la rudesse agirait comme un défi cognitif. Face à une tournure provocante – « Si tu n'es pas complètement à côté de la plaque, réponds à ceci » – le modèle basculerait dans un mode plus compétitif, optimisant sa recherche de précision. Une attitude qui rappelle, ironiquement, le comportement humain face à la pression.N'insultez pas pour autant les IA !Les chercheurs insistent : inutile de devenir désagréable avec vos robots. L'expérience portait sur un échantillon limité et un seul modèle. Surtout, la clé ne serait pas la méchanceté, mais la clarté syntaxique et la concision. Et pour la santé mentale collective, mieux vaut préserver un climat bienveillant, même à l'ère des machines.-----------♥️ Soutien : https://mondenumerique.info/don

Welcome to Nerd Alert, a series of special episodes bridging the gap between marketing academia and practitioners. We're breaking down highly involved, complex research into plain language and takeaways any marketer can use.In this episode, Elena and Rob explore how consumer response to advertising behaves more like a phase transition than a smooth curve. They discuss saturation points, tipping points, and why some channels scale differently than others.Topics covered:   [01:00] "Symmetry Scaling Laws and Phase Transitions in Consumer Advertising Response"[03:00] Three parameters: marketing effectiveness, response sensitivity, and behavioral sensitivity[04:00] Which parameter matters most for TV?[05:00] Testing against traditional models[06:00] Network effects and audience clustering  To learn more, visit marketingarchitects.com/podcast or subscribe to our newsletter at marketingarchitects.com/newsletter.  Resources: Marín, J. (2024). Symmetries, scaling laws and phase transitions in consumer advertising response [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2404.02175   Get more research-backed marketing strategies by subscribing to The Marketing Architects on Apple Podcasts, Spotify, or wherever you listen to podcasts. 

In this week's round-up of the latest news in online speech, content moderation and internet regulation, Mike and Ben cover:It's Cool to Have No Followers Now (New Yorker)Introducing Safe for Work? — all about T&S jobs (Everything in Moderation*)Kids Turn Podcast Comments Into Secret Chat Rooms, Because Of Course They Do (Techdirt)Reddit and Kick added to child social media ban (ABC News)X boss explains why ‘horrific' video viewed by Axel Rudakubana wasn't removed (The Independent)Southport Inquiry (YouTube)How Elon Musk is Boosting The British Right (Sky News)arXiv Changes Rules After Getting Spammed With AI-Generated 'Research' Papers (404 Media)TikTok Investigated in France Over Content That Promotes Suicide (Bloomberg) France Moves to Block the Shein Website Over a Sex Doll Scandal (New York Times) EU leaders paper over splits on US tech reliance (Politico) This episode is brought to you by our sponsor Safer by Thorn, a purpose-built CSAM and CSE solution. Powered by trusted data and Thorn's issue expertise, Safer helps trust and safety teams proactively detect CSAM and child sexual exploitation messages. Ctrl-Alt-Speech is a weekly podcast from Techdirt and Everything in Moderation. Send us your feedback at podcast@ctrlaltspeech.com and sponsorship enquiries to sponsorship@ctrlaltspeech.com. Thanks for listening.

Das ist das KI-Update vom 05.11.2025 unter anderem mit diesen Themen: KI-Rechenzentren im Orbit Die Huxley Gödel Maschine AI Slop bei Arxiv Chatbots und das Thema Bewusstsein === Anzeige / Sponsorenhinweis === Dieser Podcast wird von einem Sponsor unterstützt. Alle Infos zu unseren Werbepartnern findet ihr hier. https://wonderl.ink/%40heise-podcasts === Anzeige / Sponsorenhinweis Ende === Links zu allen Themen der heutigen Folge findet Ihr im Begleitartikel auf heise online: https://heise.de/-11056835 heise Themenabend „Crashkurs in Online-Sicherheit“ https://aktionen.heise.de/heise-themenabend Weitere Links zu diesem Podast: https://www.heise.de/thema/KI-Update https://pro.heise.de/ki/ https://www.heise.de/newsletter/anmeldung.html?id=ki-update https://www.heise.de/thema/Kuenstliche-Intelligenz https://the-decoder.de/ https://www.heiseplus.de/podcast https://www.ct.de/ki Eine neue Folge gibt es montags, mittwochs und freitags ab 15 Uhr.

Die Themen in den Wissensnachrichten: +++ Attraktive Leute verdienen etwas mehr +++ Wie eigennützig KI-Modelle handeln +++ Warum Golfbälle aus dem Loch wieder herausspringen +++ **********Weiterführende Quellen zu dieser Folge:Spontaneous Giving and Calculated Greed in Language Models, arXiv, 29.10.2025Gendered beauty inequalities? A multiverse analysis of physical attractiveness, occupational gender-typicality and earnings in the German labour market, European Sociological Review, 04.05.2025Increasing extreme winds challenge offshore wind energy resilience, Nature Communications, 04.11.2025Multifunctional Fluidic Units for Emergent, Responsive Robotic Behaviors, Advanced Materials,Mechanics of the golf lip out, Royal Society Open Science, 05.11.2025Alle Quellen findet ihr hier.**********Ihr könnt uns auch auf diesen Kanälen folgen: TikTok und Instagram .

Para entender la importancia de este hallazgo, hace falta darle un repaso a un concepto fundamental en la astrofísica: la metalicidad. Y aquí es donde entra en escena nuestra protagonista: SDSS J0715-7334. Porque esta estrella ha pulverizado todos los récords conocidos de pureza. De hecho, es aproximadamente 20.000 veces más pobre en metales que nuestro Sol. Los datos preliminares de su espectro, presentados en el servidor arXiv a la espera de revisión por pares, sugieren que su metalicidad total es tan baja que supera incluso a las galaxias más distantes y supuestamente "vírgenes" que hemos podido observar con el telescopio espacial James Webb en el borde mismo del Universo observable. Para hacernos una idea, nuestra estrella protagonista es diez veces más pura que esas remotísimas galaxias. Es, en esencia, la estrella más cercana a lo que llamamos una estrella prístina, o libre de metales, que hemos encontrado hasta ahora.

Ceasefires are good; Trump's personal stake in Middle East diplomacy; the implausibility of the AI apocalypse; the incremental ceding of human control over autonomous systems; plausible mechanisms for AI control of nuclear arsenals; misalignment, poisoning, and the manipulation of AI models; and Marcus discusses the meaning of consciousnessThe opinions expressed on this podcast are solely our own and do not reflect the policies or positions of William & Mary.Please subscribe to Cheap Talk on Apple Podcasts, Spotify, or your podcast player of choice to be notified when new episodes are posted.Check out our online store at https://cheaptalk.shop.Further Reading:Alexandra Souly et al. 2025. “Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples.” Arxiv 2510.07192.“AI: What Could Go Wrong? With Geoffrey Hinton.” 2025. The Weekly Show with Jon Stewart.See all Cheap Talk episodes

David Deutsch argues that Einstein would struggle to secure modern research grants, exposing how funding systems favor incremental work over bold, fundamental ideas. He connects this bias to quantum computing, constructor theory, free will, and the role of creativity in scientific breakthroughs.- 00:00 - Einstein's Grant Application- 07:00 - Funding People, Not Projects- 12:35 - Is Physics Stagnant?- 17:34 - The "Checkbox" Problem- 26:05 - Physics vs. Math Departments- 32:42 - Fundamental vs. Foundational- 40:08 - Physicists and Philosophy- 45:44 - Why Academics Are Silent- 51:20 - The Problem of Quantum Gravity- 58:31 - Qubit Field Theory- 1:03:18 - Problem-Solving in Physics- 1:17:14 - Deutsch's "Impossible" List- 1:24:23 - Meeting Hugh Everett- 1:35:01 - Susskind's MWI Objections- 1:46:44 - Everett and Quantum Computing- 1:56:20 - Constructor Theory- 2:03:01 - Free Will and Knowledge- 2:09:08 - Follow The FunSPONSORS:- The Economist: 20% off - https://www.economist.com/toe- Claude: 50% off Claude Pro - http://claude.ai/theoriesofeverythingRESOURCES:- Beginning Of Infinity [Book]: https://www.amazon.com/Beginning-Infinity-Explanations-Transform-World/dp/0143121359- How To Reverse Academia's Stagnation [YouTube]: https://youtu.be/Em-85baHx0A- Qubit Field Theory [Paper]: https://arxiv.org/pdf/quant-ph/0401024- Quantum Theory, The Church–Turing Principle And The Universal Quantum Computer [Paper]: https://royalsocietypublishing.org/doi/10.1098/rspa.1985.0070- ArXiv: https://arxiv.org/- Scott Aaronson [TOE]: https://youtu.be/1ZpGCQoL2Rk- Wayne Myrvold [TOE]: https://youtu.be/HIoviZe14pY- Neil Turok [TOE]: https://youtu.be/zNZCa1pVE20- String Theory Iceberg [TOE]: https://youtu.be/X4PdPnQuwjY- Alex Honnold [TOE]: https://youtu.be/D4oXvxqzSyA- Michael Levin Λ Anna Ciaunica: https://youtu.be/2aLhkm6QUgA- Stephen Wolfram [TOE]: https://youtu.be/FkYer0xP37E- The Heisenberg Picture: https://chem.libretexts.org/Bookshelves/Physical_and_Theoretical_Chemistry_Textbook_Maps/Advanced_Statistical_Mechanics_(Tuckerman)/09%3A_Review_of_the_basic_postulates_of_quantum_mechanics/9.04%3A_The_Heisenberg_Picture- Jacob Barandes Λ Emily Adlam: https://youtu.be/rw1ewLJUgOg- Everett's Letter To DeWitt: https://www.pbs.org/wgbh/nova/manyworlds/orig-02.html- The Many-Worlds Interpretation Of Quantum Mechanics [Book]: https://www.amazon.com/Interpretation-Quantum-Mechanics-Princeton-Library/dp/069161895X- Leonard Susskind [TOE]: https://youtu.be/2p_Hlm6aCok- Sean Carroll [TOE]: https://youtu.be/9AoRxtYZrZo- David Wallace [TOE]: https://youtu.be/4MjNuJK5RzM- Chiara Marletto [TOE]: https://youtu.be/40CB12cj_aM- Roger Penrose [TOE]: https://youtu.be/sGm505TFMbU- Robert Sapolsky [TOE]: https://youtu.be/z0IqA1hYKY8- Yang-Hui He [TOE]: https://youtu.be/spIquD_mBFk- Maria Violaris [TOE]: https://youtu.be/Iya6tYN37ow Learn more about your ad choices. Visit megaphone.fm/adchoices

Roderick Hyde discusses his recent paper on using high-altitude hoses for solar geoengineering. While most proposals focus on aircraft delivery, Hyde revisits an older but largely dismissed concept. He describes suspending a 20 km hose by balloons to continuously pump sulfur-bearing fluids into the stratosphere, and argues that advances in modern materials and engineering may overcome past barriers.The conversation covers the technical hurdles such as wind dynamics, hose stability, extreme pressures, and material stress, as well as design variations for pumping H₂S as liquid or gas. Hyde explains how streamlining, intermediate pumps, and lightweight aero-shrouds could make the system viable.The discussion also highlights the potential advantages of this approach, including affordability, continuous operation, and scalability. While a single hose could not halt global warming, Hyde suggests that a distributed network of ~20 installations could offset warming from CO₂, offering a near-term, low-cost option to buy time while longer-term climate solutions take effect.Paper: Hyde, R. A. (2025). A Planetary Cooling Hose. arXiv preprint arXiv:2509.07985. https://doi.org/10.48550/arXiv.2509.07985

Julie M. Liss, Ph.D., CCC-SLP, is Senior Associate Dean in the College of Health Solutions and Professor of Speech and Hearing Science at Arizona State University. A certified speech-language pathologist, she is internationally recognized for advancing the use of artificial intelligence in speech-language pathology. Her work has focused on transforming clinical speech science into digital tools that can detect and track neurological disease, expand access to care, and support more precise clinical decision-making. As co-founder of Aural Analytics, Dr. Liss has helped pioneer speech-based biomarkers and AI-driven assessment platforms now in use around the world. She is also a thought leader in promoting the ethical and responsible application of AI in healthcare and scientific publishing. Beyond her research and innovation, Dr. Liss has served in key leadership roles with ASHA, including as Editor-in-Chief and now Chair of the ASHA Journals Board, where she is helping shape policy around emerging technologies in scholarly communication. In recognition of her impact on the profession and her leadership at the intersection of speech science and technology, Dr. Liss is receiving Honors of the Association from ASHA in 2025. Visar Berisha, Ph.D., is a Professor at Arizona State University with a joint appointment in the College of Engineering and the College of Health Solutions and Associate Dean for Research Commercialization in the College of Engineering. His main research interests reside at the intersection of AI and the human voice. He has developed and commercialized new speech AI models for healthcare. This work is primarily funded by the National Institutes of Health, the Department of Defense, and the National Science Foundation. This work has led to many academic publications, several patents, and two companies. Dr. Berisha's work has been featured in the New York Times, on ESPN, National Public Radio, the Wall Street Journal, and a number of other international media outlets. He was the 2023-2024 ISCA Distinguished Lecturer. References: Berisha, V., & Liss, J. M. (2024). Responsible development of clinical speech AI: Bridging the gap between clinical research and technology. npj Digital Medicine, 7, Article 208. https://doi.org/10.1038/s41746-024-01199-1 Liss, J., & Berisha, V. (2024). Operationalizing clinical speech analytics: Moving from features to measures for real-world clinical impact. Journal of Speech, Language, and Hearing Research, 67(11), 4226-4232. Liss, J., & Berisha, V. (2020, August). How will artificial intelligence reshape speech-language pathology services and practice in the future? ASHA Journals Academy. https://academy.pubs.asha.org/2020/08/how-will-artificial-intelligence-reshape-speech-language-pathology-services-and-practice-in-the-future/ Xu, L., Chen, K., Mueller, K. D., Liss, J., & Berisha, V. (2025). Articulatory precision from connected speech as a marker of cognitive decline in Alzheimer's disease risk-enriched cohorts. Journal of Alzheimer's Disease, 103(2), 476-486. Yeo, E., Liss, J., Berisha, V., & Mortensen, D. (2025). Applications of Artificial Intelligence for Cross-language Intelligibility Assessment of Dysarthric Speech. arXiv. https://arxiv.org/pdf/2501.15858

Welcome to Nerd Alert, a series of special episodes bridging the gap between marketing academia and practitioners. We're breaking down highly involved, complex research into plain language and takeaways any marketer can use.In this episode, Elena and Rob explore how the Mixture of Experts model uses AI to uncover four distinct consumer segments that traditional econometric models miss entirely. They reveal why assuming all consumers behave the same way leads to missed opportunities and how dynamic segmentation can transform marketing strategy.Topics covered:   [01:00] "How Do Consumers Really Choose? Exposing Hidden Preferences With the Mixture of Experts Model"[03:00] Four hidden consumer segments uncovered by AI[05:00] Why traditional models oversimplify consumer behavior[06:00] The power of discount thresholds in promotion-driven buying[07:00] How AI creates flexible, realistic customer pictures[08:00] Why consumers are more like a zoo than a herd of cows  To learn more, visit marketingarchitects.com/podcast or subscribe to our newsletter at marketingarchitects.com/newsletter.  Resources: Vallarino, D. (2025). How Do Consumers Really Choose? Exposing Hidden Preferences with the Mixture of Experts Model. arXiv preprint arXiv:2503.05800.   Get more research-backed marketing strategies by subscribing to The Marketing Architects on Apple Podcasts, Spotify, or wherever you listen to podcasts. 

Let's go, Grue Jays!New kinds of birds are not usually discovered while browsing Facebook, but an ornithologist spotted something he'd never seen before in a photo, and tracked down the strange bird. Brian Stokes, a PhD candidate at the University of Texas at Austin, discovered it was actually a previously unknown hybrid of the familiar blue jay and a green jay, better known from southern parts of North America. Climate change likely played a part in bringing the two species together. Their research was published in the journal Ecology and Evolution. Chimpanzees' taste for ripe fruit is equivalent to two drinks a dayChimpanzees in the wild can eat about 10 per cent of their body weight worth of fruit each day, and all of that fruit contains small amounts of alcohol. A team of scientists, including Aleksey Maro from the University of California Berkeley, wanted to understand just how much alcohol the chimps were getting from all this fruit. Three different methods of analysis over three years revealed the chimps were consuming the equivalent of two standard drinks a day. This suggests an evolutionary explanation for the human taste for ethanol. The research was published in the journal Science Advances.Sea life says make homes, not bombsAfter the defeat of Germany in 1945, an estimated 1.6 million tons of munitions were dumped into the Baltic sea off the German coast. A team of researchers, including marine biologist Andrey Vedenin from the Senckenberg Research Institute, wanted to understand how this potentially toxic legacy had affected sea life. They were stunned to discover thousands of animals surviving on the abandoned weapons despite the toxic burden they carried. The research was published in the journal Nature Communications Earth & Environment.Structure of social media sites 'inherently lead to something problematic'Our experience of social media sites is that they often descend into extremism, divisiveness and conflict, but this may be a feature, not a bug. In a pre-print study on arXiv, scientists simulated social media interactions between AI-generated participants to test various interventions to see how they'd impact the problems that emerge, such as the rise of echo chambers, the concentration of influence and the amplification of polarized voices. Petter Törnberg, a University of Amsterdam computational social scientist, said he was disappointed to learn that none of the interventions worked.Your brain's two halves hand off perception like a baton in a relay raceWhen something passes from one side of your visual field to the other, something amazing happens, according to new research published in the Journal of Neuroscience. Matthew Broschart, a postdoctoral fellow at MIT, tracked how the visual parts of each half of the brain, connected to each eye, do a coordinated dance to create a unified visual perception in primates. The bear necessities of tracking B.C. grizzlies with machine learning softwareScientists and guardians from five First Nations of the Nanwakolas Council are working together to track individual grizzlies across the southern Great Bear Rainforest in B.C.. Using camera traps and machine learning techniques, they've developed an automated system through the BearID Project to identify individual bears and track them over the landscape. We spoke with conservation scientist and director of the BearID Project, Melanie Clapham, and Tashina James-Matilpi, from the Tlowitsis First Nation, the project's guardian logistics coordinator for the Nanwakolas Council.

Most people prefer AI art until they know it's AI-generated.  Today on Nudge, Professor of Consumer Psychology Matt Johnson explains why.  ---  Watch the bonus episode: https://nudge.kit.com/5fa3398dfb   More on Matt's books, work and newsletters: https://www.neuroscienceof.com/ Sign up for my newsletter: https://www.nudgepodcast.com/mailing-list  Connect on LinkedIn: https://www.linkedin.com/in/phill-agnew-22213187/  Watch Nudge on YouTube: https://www.youtube.com/@nudgepodcast/  --- Sources:  Bellaiche, L., Shahi, R., Turpin, M. H., Ragnhildstveit, A., Sprockett, S., Barr, N., & Seli, P. (2023). Humans versus AI: Whether and why we prefer human-created compared to AI-created artwork. Cognitive Research: Principles and Implications, 8(1), 42. Glenn, J., & Walker, R. (Eds.). (2012). Significant objects: 100 extraordinary stories about ordinary things. Fantagraphics. Lecamwasam, K., & Ray Chaudhuri, T. (2025). Exploring listeners' perceptions of AI-generated and human-composed music for functional emotional applications. arXiv preprint arXiv:2506.02856. Oasis. (2002, April 15). Oasis – The Hindu Times (Official Video) [Video]. YouTube. https://youtu.be/Wp5zZ5cdu98 Oasis – Lost In The Clouds (AI Song). (2024, approximate). Oasis – Lost In The Clouds (AI Song) [Video]. YouTube. https://youtu.be/ZmC9RYRitLs

ESA's Solar Orbiter camera probe begins raising its orbit towards the sun's poles, whilst Betelgeuse's elusive buddy continues to sneak past our best telescopes.Earlier this year, Solar Orbiter started to stretch its orbit over greater latitudes – effectively standing on cosmic tiptoes to catch a glimpse of the Sun's poles. This week, we have seen the first ever pictures of them, and as solar scientist Steph Yardley tells us, the views will only get better.Meanwhile, Andrea Dupree of the Harvard Smithsonian Center for Astrophysics and colleagues have had time to study new Hubble and Chandra telescope observations of the iconic star Betelgeuse searching for signs of its hypothesised binary companion – dubbed “Betelbuddy”. The papers that appeared on the Arxiv pre-print server have not yet been fully peer-reviewed, but it seems astronomers will have to keep looking.Humans use machines to read gene sequences as best they can, but it takes time and is not perfect because we do not know what all of it means. Of course nature has its own genome reader – the ribosome. It is this that interprets the genetic instructions contained in our DNA and translates them into actual proteins. Viruses, of course, use it too when a cell gets infected. Shira Weingarten-Gabbay has this week demonstrated how scientists can make use of ribosomes too. Working somewhat in reverse, her team have identified many thousands of proteins previously unknown, that could for example provide targets for future vaccines or antivirals should the need arise.Finally, Nanshu Lu and team in the University of Texas at Austin have been working for some years on two-dimensional wearable electronic “E-Tattoos” to monitor health non-invasively through our skin. Their latest work, describes “A wireless forehead e-tattoo for mental workload estimation”.Presenter: Roland Pease Producer: Alex Mansfield Production co-ordinator: Jasmine Cerys GeorgePhoto Credit: ESA & NASA/Solar Orbiter/SPICE Team, M. Janvier (ESA) & J. Plowman (SwRI)