Already our Fairwater datacenter in Atlanta has taken over 15 million labor hours to build – even more once it's fully finished. For comparison the Empire State building took 7 million!
AI HARDWARE
-
OpenAI secures record $38B AWS deal for Nvidia GPUs
By
–
Pendant que Google vise les étoiles,
OpenAI signe un contrat record avec AWS : 38 milliards de dollars pour s’assurer assez de GPU Nvidia d’ici 2026. La bataille de la puissance devient le nouveau bras de fer planétaire. -
AI Giants Launch Orbital Datacenters and Space Infrastructure
By
–
L’IA ne se contente plus d’envahir nos réseaux.
Elle regarde désormais vers le ciel. Google veut propulser ses puces d’IA en orbite avec Suncatcher.
Elon Musk prépare ses datacenters stellaires avec Starlink 3.
Et Nvidia finance Starcloud, qui veut lancer des datacenters -
Startup’s Metal Stacks Address AI’s Critical Heat Challenge
By
–
This startup’s metal stacks could help solve AI’s massive heat problem
#AI #AIio #AIInnovation #ML #DataScience #Futureofwork @HaroldSinnott @fogoros @iainljbrown @NandoDF @katecrawford @drhassanrashidi @YuHelenYu -
Intel’s First Single-Chip CPU: 1971 Computing Milestone
By
–
#otd in 1971 Intel unveiled the first single-chip CPU. 4 bits, 16 pins & 740KHz: https://
bit.ly/3YLCqdV -
Unitree G1 Robot Update Shows Remarkable Everyday Task Performance
By
–
La dernière update du robot G1 de unitree, entrainé pour les tâches quotidiennes.
— VISION IA (@vision_ia) 15 novembre 2025
C’est impossible… même s’ils disent “pas téléopéré”. Les mouvements sont tellement fluides et les actions tellement variées.
Si c’est vrai, la Chine a vraiment une longueur d’avance. pic.twitter.com/ghUoLsm1BsLa dernière update du robot G1 de unitree, entrainé pour les tâches quotidiennes. C’est impossible… même s’ils disent “pas téléopéré”. Les mouvements sont tellement fluides et les actions tellement variées. Si c’est vrai, la Chine a vraiment une longueur d’avance.
-
NVIDIA Nominated for Global Semiconductor Awards Recognition
By
–
Honored to be nominated for the 2025 @GlobalSemi Analyst Favorite Semiconductor Company Award and the Most Respected Public Semiconductor Company Award. It’s a privilege to be recognized alongside other remarkable nominees. See the full list: https://
nvda.ws/4nULnvO -
Fairwater AI Atlanta Datacenter Powers AI with NVIDIA Blackwell
By
–
The new Fairwater AI datacenter in Atlanta, powered by hundreds of thousands NVIDIA Blackwell and Blackwell Ultra GPUs leveraging the NVIDIA GB200 and GB300 NVL72 rack-scale systems with Spectrum-X Ethernet, is pushing boundaries for performance, sustainability, and AI… https://t.co/xLjATgdYMg
— NVIDIA (@nvidia) 13 novembre 2025The new Fairwater AI datacenter in Atlanta, powered by hundreds of thousands NVIDIA Blackwell and Blackwell Ultra GPUs leveraging the NVIDIA GB200 and GB300 NVL72 rack-scale systems with Spectrum-X Ethernet, is pushing boundaries for performance, sustainability, and AI
-
NVIDIA Dynamo Cloud Integration Expands Performance Across Major Providers
By
–
Now, with NVIDIA Dynamo integrating with major cloud providers that same breakthrough performance is available to cloud customers—including @AWSCloud
, @Azure
, @GoogleCloud
, and @OracleCloud
. Learn more: https://
nvda.ws/3LBsT6e -
MoE Models Scale Efficiently with NVIDIA Blackwell Hardware
By
–
Mixture-of-Experts (MoE) models like DeepSeek-R1 unlock new levels of capability—but only if they can scale efficiently.
That’s where extreme hardware–software co-design at rack-scale comes in. With NVIDIA Blackwell and NVIDIA Dynamo, AI service providers can transform clusters