Ultimate PocketIA
IA
OpenAI just wants to win
Two mathematicians at the heart of recent controversies tell The Verge how AI giants’ ‘childish’ rivalries are upending the field. Two mathematicians at the heart of recent controversies tell The Verge how AI giants’ ‘childish’ rivalries are upending the field. OpenAI has spent the last few years planting flags across the increasingly difficult terrain in mathematics. This week, it claimed one of its biggest prizes yet: a solution to a legendary Millennium Prize problem.…
Lire l’article complet
IA
‘I’m Really Terrified’: A Mathematician Grapples With AI’s Recent Breakthroughs
Mathematician and author Steven Strogatz starts to cry when he talks about the artificial-intelligence-driven breakthroughs in his field over the past week. “The science is thrilling,” the Cornell University professor says, “but there's a lot of human unpleasantness going with it.” On Tuesday, OpenAI said it had used tens of thousands of agents to solve a 90-year-old math problem, which had a $1 million prize attached. The solution—which still needs to be independently verified—builds…
Lire l’article complet
IA
Apple researchers unveil SimpleDesign, a new AI model for protein design
In a new study, Apple researchers detail SimpleDesign, a streamlined AI model that can jointly generate protein sequences and structures. Here are the details. Last September, Apple researchers published a study titled “SimpleFold: Folding Proteins is Simpler than You Think,” which detailed a streamlined approach to predicting a protein’s 3D structure from its amino acid sequence. In a nutshell, SimpleFold uses a flow-matching model to generate a protein’s 3D structure directly from its amino…
Lire l’article complet
IA
Mecka AI nears $500M valuation in Sequoia-led deal amid rush for robot training data
Mecka AI, a startup that collects and analyzes human motion data to train humanoid robots and other robotics, is nearing a new round led by Sequoia Capital at a valuation of about $500 million, according to two people with knowledge of the deal. The new financing comes just three months after Mecka announced that it raised $60 million in a round led by Framework Ventures that included participation from Menlo Ventures, SV Angel, and…
Lire l’article complet
IA
Y Combinator’s Garry Tan wants US open-weight AI labs to ‘distill’ frontier models, too
When it comes to Chinese AI labs using distillation techniques to extract knowledge from frontier model makers, Y Combinator CEO Garry Tan is hoping regulators stay out of it. In fact, he thinks U.S. AI labs should perhaps play the same game. “I would do nothing,” he told CNBC in an interview earlier this week. “We could argue that there should be an American distillation regime.” He elaborated to TechCrunch that this means he…
Lire l’article complet
IA
OpenAI’s feud with mathematicians is only escalating
Twenty-five leading mathematicians signed an open letter arguing that AI labs are threatening their intellectual work as they seek to one-up each other with solutions to famous math problems. Each signatory has been awarded the Fields Medal, considered the most prestigious prize in mathematics. This week, NYU professor Tristan Buckmaster accused OpenAI of pressuring him not to credit a collaborator who works for Anthropic for solving an important math problem, and wondered if the…
Lire l’article completBuilding websites with ChatGPT just got faster and more collaborative
TL;DR ChatGPT Sites’ new Build Together feature lets multiple people edit, save, and publish the same Site. Sites can now stay private and be shared with specific people, and deployment is roughly twice as fast. ChatGPT can now inspect a Site’s database, and Sites also support custom domains. ChatGPT seems to get a new feature every other day. This time, OpenAI is turning its attention to ChatGPT Sites, which launched about three months ago…
Lire l’article complet
IA
Lawyer fined $5K over AI-hallucinated witnesses in a murder case
A justice asked, ‘Counsel, do you watch the news?’ when a lawyer admitted to relying on ChatGPT without understanding the risks. A justice asked, ‘Counsel, do you watch the news?’ when a lawyer admitted to relying on ChatGPT without understanding the risks. New Mexico’s Supreme Court is punishing a lawyer for including AI-fabricated witnesses and fake police testimony in an appeal for his client’s murder conviction, according to a report from Reuters. In a…
Lire l’article complet
IA
Roundtables: AI’s apocalypse crisis
Employees at the world's leading AI labs are saying there's a real possibility that advanced AI could destroy humanity. Are they right? Or is this more scaremongering and hype? Join MIT Technology Review executive editor Niall Firth for a conversation with senior AI editor Will Douglas Heaven and AI reporter Grace Huckins unpacking AI extinction fears: where they come from, whether they hold any water, and, if so, what we should do. Going live…
Lire l’article complet
IA
Kimi-maker Moonshot AI targets $2B in annual revenue
One of China’s most prominent AI labs, Moonshot AI, believes it can turn its popular open-weight model into serious sales growth. On Friday, Bloomberg reported that the lab is targeting $2 billion in annualized revenue by the end of the year, double the company’s reported revenue run rate for August. It’s an aggressive goal that reflects the success of the company’s K3 model since its release this summer. While K3’s usage figures have declined…
Lire l’article complet
IA
ChatGPT-using lawyer punished for citing fake testimony from made-up witnesses
The New Mexico Supreme Court held a ChatGPT-using lawyer in direct contempt of court for submitting a brief with "false testimony from wholly fabricated witnesses," including fake police testimony and other mistakes. The state's top court referred the lawyer to a disciplinary board for further proceedings and concluded that he "demonstrated a lack of remorse and a lack of concern for his client." Attorney Stephen Aarons "admitted to the Court that he did not…
Lire l’article complet
IA
Google Find Hub rolls out new ‘Remembered’ tab with Gemini [U]
Following last week’s Android Drop, Google is now rolling out an app update that brings the “Remembered” tab to Find Hub. Update 9/11: Find Hub 3.1.732-02 has now widely rolled out via the Play Store with the new “Remembered” capability. Original 9/8: After updating, you’ll see a new tab join Devices and People. Remembered uses a bookmark icon badged by the Gemini spark. You can “Tell Gemini to remember where you put everyday belongings,”…
Lire l’article complet
IA
En Europe, l’usage public des lunettes Meta avec l’IA jugé illégal par des experts
Harcèlement de rue augmenté par IA Dans un rapport très fourni, l’autorité de protection des données hambourgeoise donne sa position sur l’utilisation légale des lunettes connectées de Meta, que ce soit dans leur utilisation « simple » pour prendre des photos, l’utilisation avec Meta AI ou celle en connexion avec les réseaux sociaux. Elle martèle clairement que la reconnaissance de personnes en dehors d’un cercle proche restreint via ces lunettes sera interdite. Les experts…
Lire l’article complet
IA
Microsoft Copilot joins CarPlay, leaving Gemini as the biggest holdout
Apple CarPlay now works with six major third-party AI chatbot apps. Microsoft Copilot is the latest iPhone app to add CarPlay support. This leaves Google Gemini as the last major holdout for now. Starting Monday, CarPlay will also include a dedicated Siri AI app. That’s when iOS 27 will be officially released and out of beta. Siri AI itself will still be in beta and rolled out in waves. Meanwhile, Microsoft has added CarPlay…
Lire l’article complet
IA
Vos tableaux Google Sheets se lisent enfin facilement sur smartphone grâce à Gemini
Lire une feuille de calcul depuis un smartphone devient vite un casse-tête. Pour faciliter cette tâche, Google intègre désormais Gemini à l’application Google Sheets sur Android. L’assistant permet d’interpréter les données sans obliger l’utilisateur à tout examiner lui-même. Depuis plusieurs mois, l’intelligence artificielle de Google prend une place croissante dans les services mobiles du groupe. De plus en plus d’applications et de paramètres du système accueillent désormais son assistant. Les ajouts se succèdent chaque…
Lire l’article complet
IA
Un essaim d’agents IA s’acharne sur ce logiciel et pirate 395 organisations dans 48 pays
Recevez tous les soirs un résumé de l’actu importante avec Le Récap’ PaperCut, voilà la cible sur laquelle se sont concentrés les attaquants. Ce logiciel de gestion d’impression très répandu permet aux organisations de suivre et facturer leurs travaux d’impression, copie et numérisation. Problème : auto-hébergés, ces logiciels tournent régulièrement avec des privilèges système sous Windows, et sont généralement connectés à un domaine Active Directory. Autrement dit, une brèche dans PaperCut peut ouvrir la…
Lire l’article complet
IA
ZDNET Morning 11/09/2026 : GPT-6 Astra pour la finance, Test iPhone Duo, Quand Amazon et Anthropic sacrifient des livres rares,…
Le ZDNET Morning le brief de l'actu tech pour les pros tous les matins à 9h00. Transformation numérique, IA, matériel, logiciels,... ne passez pas à côté de ce qui fait la Une du secteur. Le ZDNET Morning c'est LE condensé de l'actu tech pour les pros tous les matins à 9h00 sur le site. Transformation numérique, IA, matériel, logiciels,... ne passez pas à côté de ce qui fait la Une du secteur. 🤖 GPT-6…
Lire l’article complet
IA
Web pratique : pour naviguer avec l’IA utilisez la Smart Window de Firefox
Mozilla a enfin lancé la version bêta de sa « fenêtre intelligente » en France et en français, son déploiement venant juste démarrer. Voici ci-dessous quelques atouts qui pourraient bien vous faire revenir à son navigateur au cas où vous vous en seriez éloigné. La "fenêtre intelligente" Smart Window de Firefox, Ph. Jack Wallen/ZDNET. Points essentiels à retenir de Smart Window de Firefox Pendant la majeure partie de la dernière décennie, Firefox a été…
Lire l’article complet
IA
OpenAI s'attaque aux opérations financières automatisées avec son nouveau GPT-6 Astra
OpenAI lance ChatGPT for Financial Services, une déclinaison de son assistant conçue pour les banques et institutions financières. Recherche, modélisation financière, analyse de données et présentations commerciales sont au cœur de cette nouvelle offre. OpenAI veut pousser l'IA générative plus loin dans les salles de marché et les banques d'investissement. Avec ChatGPT for Financial Services, l'entreprise propose un service destiné aux professionnels de la finance, avec l'objectif d'automatiser une partie des tâches de recherche…
Lire l’article complet
IA
ZD Tech : Voici pourquoi Amazon et Anthropic détruisent des livres rares pour alimenter leurs modèles d'IA
Face à la pénurie de données en ligne, les géants de la tech traquent désormais les savoirs jamais numérisés. En découpant et scannant des ouvrages d'occasion rares avant de les jeter, Amazon et Anthropic exploitent une subtilité juridique particulièrement rentable. Acheter des livres rares d'occasion, arracher leurs reliures, passer leurs pages à toute vitesse au scanner puis tout jeter à la poubelle. Non, ce n'est pas une scène absurde, mais la méthode choc employée…
Lire l’article complet
IA
Oracle accélère sa bascule vers l’IA avec un chiffre d’affaires cloud en hausse de 121 %
Pour financer cette expansion sans alourdir excessivement son propre bilan, Oracle s’appuie de plus en plus sur les capitaux et le matériel de ses clients, une stratégie qui lui permet de maintenir le rythme de ses investissements en infrastructure IA tout en limitant la pression sur sa trésorerie. L’infrastructure dédiée à l’IA porte désormais la croissance d’Oracle. Lors du premier trimestre de l’exercice fiscal 2027, le chiffre d’affaires de l’infrastructure cloud (OCI) a atteint…
Lire l’article complet
IA
CA governor signs 'landmark' laws on youth use of social media and AI chatbots
They’re aimed at protecting minors, but the rules are not universally supported. California Governor Gavin Newsom has signed multiple measures aimed at protecting the state's youth. These new laws include restrictions for potentially addictive social media features and limitations on how young users can interact with AI chatbots. The state is enacting restrictions on how teens can interact with chatbots from AI companies, including time limits. Chatbot makers will be required to implement mental…
Lire l’article complet
IA
Anthropic caught scientists using Claude to further biological weapon research
The company has produced an extensive collection of case studies covering the ways its current AI models have been misused. Anthropic has flagged and stopped multiple scientists who were using its AI models to create potential biological weapons, according to a new report the company shared on Thursday. "Biological misuse" is just one flavor of troubling AI usage the report documents, but it's of growing importance because newer models like Fable 5 are increasingly…
Lire l’article complet
IA
OpenAI puts Pro subscriptions on hold due to Astra demand
Demand for OpenAI’s newest and most powerful model, Astra, has led the company to temporarily pause subscriptions for its $200-per-month Pro plan, citing strain on its infrastructure. The move was announced on X by OpenAI’s product leader, Thibault (Tibo) Sottiaux, who leads core products like Codex and ChatGPT at the AI lab. He said that the Pro plan puts the most strain on its systems, which is why sign-ups for this service tier are…
Lire l’article complet