Étiquette : artificial intelligence (Page 9 of 32)

A Dormant Nuclear Plant in Iowa Is Poised to Power Google’s Energy Needs

Aerial view of the Cedar River and the Duane Arnold Energy Center near Palo, Iowa.

“A nuclear plant in Iowa—currently out of action—is set to help Google reap reliable, low-carbon energy to feed power-intensive artificial intelligence.
Google said Monday it is partnering with NextEra Energy to reopen the Duane Arnold Energy Center, a nuclear plant that closed five years ago after damage from a windstorm. The plan is to get the plant, which would provide over 600 megawatts of nuclear energy for the regional grid, up and running in early 2029 pending regulatory approval. The Central Iowa Power Cooperative will buy the plant’s energy output that isn’t used by Google.
The move follows a push by the tech industry to galvanize nuclear energy in the U.S., including a deal between Constellation Energy and Microsoft to restart Pennsylvania’s Three Mile Island, the site of the worst nuclear power disaster in the country. And Meta signed a deal this summer with Constellation to secure power from the Clinton Clean Energy Center, an already-operating nuclear plant in Illinois.”

Source : A Dormant Nuclear Plant in Iowa Is Poised to Power Google’s Energy Needs – WSJ

La distorsion systémique de l’information par l’IA est constante dans toutes les langues et tous les territoires

EBU - News Integrity in AI Assistants

  • “45 % de toutes les réponses d’IA présentaient au moins une faille significative.
  • 31 % des réponses présentaient de sérieux problèmes de sourcing – Attributions manquantes, trompeuses ou incorrectes.
  • 20 % présentaient des problèmes d’exactitude majeurs, notamment des détails hallucinants et des informations obsolètes.
  • Gemini a obtenu les pires résultats, avec des problèmes significatifs dans 76 % des réponses, soit plus du double des autres assistants.
  • Les taux de refus ont fortement chuté : les assistants sont désormais plus susceptibles de répondre, même lorsqu’ils ne peuvent pas fournir une réponse fiable.”
  • Belgique (RTBF, VRT) , Canada (CBC-Radio-Canada) , Tchéquie (Radio tchèque) , Finlande (YLE) , France (Radio France) , Géorgie (GPB) , Allemagne (ARD, ZDF, Deutsche Welle) , Italie (Rai), Lituanie (LRT) , Pays-Bas (NOS/NPO) , Norvège (NRK) , Portugal (RTP) , Espagne (RTVE), Suède (SVT), Suisse (SRF), Ukraine (Suspilne), Royaume-Uni (BBC), États-Unis (NPR)

Source : La distorsion systémique de l’information par l’IA est constante dans toutes les langues et tous les territoires, selon une étude internationale menée par les radiodiffuseurs de service public. | EBU

L’IA au bureau, entre perte de temps et perte de sens

“Même sentiment d’exaspération pour Laurent, 39 ans, consultant dans le secteur de la banque à Paris. « Il y a toujours un collègue en réunion qui suggère de demander l’avis de ChatGPT. On reçoit ensuite par e-mail un résumé empli de banalités, sans aucune réelle analyse. Ces messages, c’est comme les pubs sur Internet, on essaie de les ignorer, mais c’est difficile. » Confiant, Laurent espère voir les choses s’améliorer. « Après la phase d’excitation, on entre dans le tunnel de la désillusion. Je pense que les usages vont se rationaliser. Peut-être qu’on arrivera à la conclusion que l’IA ne sert pas à grand-chose ? » Comme Maxime et Laurent, Marie, trentenaire, rédactrice dans une agence de communication nantaise, est, malgré son scepticisme, fortement incitée à utiliser les IA. « Les gens surestiment le temps gagné. Dans la réalité, le gain sur nos missions est peut-être de 5 % si on ne veut pas rogner sur la qualité. L’IA s’apparente surtout à un stagiaire zélé à forte tendance mythomane. »”

Source : L’IA au bureau, entre perte de temps et perte de sens

New Facebook Feature Suggests Edits and Collages to Share

“This new feature enhances your best photos, offers creative edits to make your content stand out, and creates fun collages and videos to help you connect with the friends and family you choose to share with. No design skills required – this feature does the heavy lifting, so you can focus on sharing the fun. All suggestions are private to you and you decide what to share, when to share, and with whom. We don’t use media from your camera roll to improve AI at Meta, unless you choose to edit this media with our AI tools, or share. Head to the Meta Privacy Policy to learn more.”

Source : New Facebook Feature Suggests Edits and Collages to Share

An Opinionated Guide to Using AI Right Now

“The chart at the top of this post shows what people use AI for today. But I’d bet that in two years, that chart looks completely different. And that isn’t just because AI changed what it can do, but also because users figured out what it should do. So, pick a system and start with something that actually matters to you, like a report you need to write, a problem you’re trying to solve, or a project you have been putting off. Then try something ridiculous just to see what happens. The goal isn’t to become an AI expert. It’s to build intuition about what these systems can and can’t do, because that intuition is what will matter as these tools keep evolving.
The future of AI isn’t just about better models. It’s about people figuring out what to do with them.”

Source : An Opinionated Guide to Using AI Right NowEthan Mollick (Wharton School of the University of Pennsylvania)

Les grands modèles de langage à la recherche d’une sanction démocratique

“Le véritable enjeu n’est pas simplement de refléter les souhaits actuels des citoyens, endogènes au système dans lequel ils sont exprimés, mais de permettre à la société tout entière de décider de façon réfléchie ce qu’elle veut devenir à l’ère algorithmique, en s’inscrivant dans un champ de possibles laissé ouvert et incluant des options de sortie – par exemple, le refus de développer ou de déployer plus avant certains genres de modèles ou d’applications. Les mini-publics délibératifs, très en faveur ces derniers temps, pourraient certainement jouer un rôle, à condition toutefois de se faire la caisse de résonance et de jouer un rôle d’éclaireurs d’un débat plus large. Seule l’implication informée du public dans son ensemble dans la discussion des algorithmes et de leurs implications peut soutenir l’effort de régulation de la part des pouvoirs publics nécessaire à réaliser un alignement compatible avec l’autodétermination démocratique.”

Source : Les grands modèles de langage à la recherche d’une sanction démocratique – AOC mediaThéophile Pénigaud – CEVIPOF

Bringing AI to the next generation of fusion energy

https://lh3.googleusercontent.com/BD0Vc3DBLrSfYQXBogn6lEsscuF-49ThPq9JXHUOP_h7RScLy5tlDS3SQ_HZV2Qc10jlD2iBbl0THusUPsjAfQP4m-awS6A3OHi4y17UpNtJ6frqYg=w2144-h1206-n-nu-rw

“Alongside our research, Google has invested in CFS, supporting their work on promising scientific and engineering breakthroughs, and moving their technology toward commercialization. Looking ahead, our vision extends beyond optimizing SPARC operations. We’re building the foundations for AI to become an intelligent, adaptive system at the very heart of future fusion power plants. This is just the beginning of our journey together, and we hope to share more details about our collaboration as we reach new milestones. By uniting the revolutionary potential of AI and fusion, we’re building a cleaner and more sustainable energy future.”

Source : Bringing AI to the next generation of fusion energy – Google DeepMind

California becomes first state to regulate AI companion chatbots

https://no-flux.beaude.net/wp-content/uploads/2025/10/GettyImages-2240040875.jpg

“California Governor Gavin Newsom signed a landmark bill on Monday that regulates AI companion chatbots, making it the first state in the nation to require AI chatbot operators to implement safety protocols for AI companions. The law, SB 243, is designed to protect children and vulnerable users from some of the harms associated with AI companion chatbot use. It holds companies — from the big labs like Meta and OpenAI to more focused companion startups like Character AI and Replika — legally accountable if their chatbots fail to meet the law’s standards.”

Source : California becomes first state to regulate AI companion chatbots | TechCrunch

AI data centers won’t have to disclose water use in California

https://no-flux.beaude.net/wp-content/uploads/2025/10/urlhttps3A2F2Fcalifornia-times-brightspot.s3.amazonaws.com2F352F852F0bb721454beb8d78d4067029218e2F1469296-me-data-centers-needed-for-ai-3.webp

“ Gov. Gavin Newsom vetoed a bill that would have tracked data centers’ growing water footprint in California. He says California is “well positioned” to support the data center boom, and he is reluctant to “impose rigid reporting requirements.” The governor signed other bills on water supply goals, groundwater dispute resolution, and water theft from fire hydrants.”

Source : AI data centers won’t have to disclose water use in California – Los Angeles Times

What billionaire Peter Thiel said in his private ‘Antichrist lectures’

Peter Thiel speaks at the Cambridge Union in 2024. (Nordin Catic/Getty Images)

“In the four, roughly two-hour lectures, which began last month and culminated Monday at the Commonwealth Club in San Francisco, Thiel laid out his religious views to a sold-out audience told to keep the contents “off-the-record,” according to an event listing. He argued that those who propose limits on technology development not only hinder business but also threaten to usher in the destruction of the United States and an era of global totalitarian rule, according to the recordings. “In the 17th, 18th century, the Antichrist would have been a Dr. Strangelove, a scientist who did all this sort of evil crazy science,” Thiel said in his Sept. 15 opening talk, according to the recordings. “In the 21st century, the Antichrist is a Luddite who wants to stop all science. It’s someone like Greta or Eliezer,” he said, referring to Thunberg and Eliezer Yudkowsky, a prominent critic of the tech industry’s approach to AI. Thunberg has criticized global capitalism as a driver of environmental degradation while Yudkowsky advocates for limiting AI research to prevent the technology from surpassing human intelligence.”

Source : What billionaire Peter Thiel said in his private ‘Antichrist lectures’ – The Washington Post

« Older posts Newer posts »

© 2026 no-Flux

Theme by Anders NorenUp ↑