Image created with gemini-2.5-flash-image with claude-sonnet-4-5. Image prompt: Cinematic wide shot of a giant silicon wafer floating in deep space like an orbital platform, intricate circuit patterns glowing with cool blue and green light, dramatic rim lighting from a distant star, tactical data streams flowing across the chip surface, Ender’s Game inspired military sci-fi aesthetic, epic scale with cold controlled lighting and deep blacks.
.@satyanadella gave me and @dylan522p an exclusive tour of Fairwater 2, the most powerful AI datacenter in the world. We then chatted through Satya’s vision for Microsoft in a world with AGI. 0:00:00 – Fairwater 2 0:04:15 – Business models for AGI 0:13:42 – Copilot 0:20:56 – https://x.com/dwarkesh_sp/status/1988656226989699138
Anthropic invests $50 billion in American AI infrastructure \ Anthropic https://www.anthropic.com/news/anthropic-invests-50-billion-in-american-ai-infrastructure
OpenAI’s Stargate Project Gets $3 Billion Blue Owl Investment — The Information https://www.theinformation.com/articles/openais-stargate-project-gets-3-billion-blue-owl-investment
AI data center buildouts already rival the Manhattan Project in scale, but there’s little public info about them. So we spent the last few months reading legal permits, staring at satellite images, and scouring news sources. Here’s what you need to know. 🧵 https://x.com/EpochAIResearch/status/1987944116861522227
By the end of the year, AI data centers could collectively see >$300 billion in investment, around 1% of US GDP. That’s bigger than the Apollo Program (0.8%) and Manhattan Project (0.4%) at their peaks.”” / X https://x.com/EpochAIResearch/status/1987944140714447327
How NVIDIA turned GPUs into currency ⬇️ An 18-month-old AI unicorn @nscale_cloud is a perfect example. You’ve probably never heard of it, but you should: • This September, Nscale raised $1.1 billion – the largest Series B in European history • Followed by $433 million https://x.com/TheTuringPost/status/1988002749452349495
How fast can you build a gigawatt-scale data center? Some hyperscalers plan to do it in just 1-2 years from the start of construction. If they succeed, we’ll see the first GW-scale data centers online in 2026, marking one of the fastest infrastructure build-outs in history. 🧵 https://x.com/EpochAIResearch/status/1987938542094610927
The government has played a role in critical infrastructure builds. Our public submission (posted on our blog) shares our thinking and suggests ideas for how the US government can support domestic supply chain/manufacturing. This is very in line with everything we have heard”” / X https://x.com/sama/status/1986917979343495650
Today, we’re announcing a new $40B investment in Texas through 2027 to build Cloud & AI infrastructure and support thousands of new jobs. This includes new data centers in Armstrong and Haskell Counties and a major investment to strengthen energy resilience and abundance. We’re https://x.com/sundarpichai/status/1989468970400055487
Every day, over 1,500 terabytes of open models and datasets are downloaded and uploaded between @huggingface and @googlecloud by millions of AI builders. We suspect it generates over a billion dollars of cloud spend annually already. So we’re excited to announce today a new https://x.com/ClementDelangue/status/1989000335247983049
🤗 @huggingface we’re announcing a closer partnership with @googlecloud to make open model development easier across the Hugging Face ecosystem and Google Cloud! – Deep Learning Containers (DLCs) for streamlined deployment and training – DLCs available via Vertex AI, Cloud Run, https://x.com/alvarobartt/status/1988970441357094984
Private AI Compute advances AI privacy https://blog.google/technology/ai/google-private-ai-compute/
SoftBank’s Nvidia sale rattles market, raises questions | TechCrunch https://techcrunch.com/2025/11/11/softbanks-nvidia-sale-rattles-market-raises-questions/
Nvidia’s Jensen Huang: ‘China is going to win the AI race,’ FT reports https://finance.yahoo.com/news/nvidias-jensen-huang-says-china-211900769.html
OpenAI Stargate Abilene – Frontier Data Centers Satellite Explorer | Epoch AI https://epoch.ai/data/data-centers/satellite-explorer/OpenAIOracleStargateAbileneTexas
AI data centers will be some of the biggest infrastructure projects in history e.g. OpenAI’s Stargate Abilene will need: – As much power as Seattle (1 GW) – >250× the compute of the GPT-4 cluster – 450 soccer fields of land – $32B – Thousands of workers – 2 years to build”” / X https://x.com/EpochAIResearch/status/1987944128903266358
I would like to clarify a few things. First, the obvious one: we do not have or want government guarantees for OpenAI datacenters. We believe that governments should not pick winners or losers, and that taxpayers should not bail out companies that make bad business decisions or”” / X https://x.com/sama/status/1986514377470845007
Ubiquity Partners with AWS to Deliver Custom Generative AI Solutions with AWS Bedrock https://www.ubiquity.com/resources/news/ubiquity-partners-with-aws-to-deliver-custom-generative-ai-solutions-with-aws-bedrock
AMD reveals new roadmap for its Ryzen CPUs, teasing Zen 7 as the true “”next-generation”” leap with 2nm — Lineup confirms 2026 release for Zen 6, coming with expanded AI features | Tom’s Hardware https://www.tomshardware.com/pc-components/cpus/amd-reveals-new-roadmap-for-its-ryzen-cpus-teasing-zen-7-as-the-true-next-generation-leap-with-2nm-lineup-confirms-2026-release-for-zen-6-coming-with-expanded-ai-features
This power runs IT equipment on “server racks” with a small area of 0.5 m^2. But each rack uses enough power for 100 homes! This means a huge amount of heat in a small space. So you can’t cool these chips with fans — you need liquid coolants to efficiently soak up the heat. https://x.com/EpochAIResearch/status/1987944200089084030
A shift from cloud to edge? We took a closer look at “Local LMs” (≤20B active parameters) and found that they are: – Surprisingly capable, with 3.1× improvement since 2023 – Increasingly efficient, with 5.3x improvement since 2023 This suggests a shift from mainframe inference https://x.com/Azaliamirh/status/1988726578264830002
Data centers can be usefully thought of as a type of industrial building Compared to other light industrial sites, they have lower employment & local economic impact, but likely (much?) higher national economic impact. They use less water & often pollute less but use more power”” / X https://x.com/emollick/status/1986662941601468808
We like pushing performance boundaries at @modal, and recently, we’ve started to make LLM engines go brr. In this post: how we worked with @DecagonAI to make SGLang 12% faster at spec dec (and more).”” / X https://x.com/akshat_b/status/1989019570783629366
TUIs are so back. (but, did they ever leave?) Our SDK wizard Dima joined the @thursdai_pod to demo W&B LEET, our new Terminal UI for tracking runs, metrics, and system health right from your TTY. See it in action in this clip! 👇 https://x.com/wandb/status/1989403717305827660
What you need to know about AI data centers | Epoch AI https://epoch.ai/blog/what-you-need-to-know-about-ai-data-centers
– “wen K3?” – “before sam’s trillion-dollar data center is built” 😂 AMA link: https://x.com/Yuchenj_UW/status/1987941323400507850
Only a few countries have enough power to build many >1 GW data centers like Stargate E.g. 30 GW is ~5% of the US’ power, ~2.5% of China’s, but ~90% of the UK’s Other countries can build some frontier data centers and grow their power capacity — but they need more time/money”” / X https://x.com/EpochAIResearch/status/1987944152542441763
Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, https://x.com/satyanadella/status/1988653837461369307
Big Tech and the AI investment boom in underwater cables https://www.cnbc.com/2025/11/08/big-tech-ai-underwater-cables.html
💥HipKittens, from Stanford ML, can 2x performance of AMD kernels vs RoCM’s composable kernels baseline. Kills baseline and up to 2x in some test.”” / X https://x.com/qubitium/status/1988389379984027742
Excited to share this piece from @VentureBeat spotlighting how Baseten is redefining the AI infrastructure game: “Baseten takes on hyperscalers with new AI training platform that lets you own your model weights.” Thanks VentureBeat! Read full article https://x.com/basetenco/status/1987943307532476746
AI has been built on one vendor’s stack for too long. AMD’s GPUs now offer state-of-the-art peak compute and memory bandwidth — but the lack of mature software / the “CUDA moat” keeps that power locked away. Time to break it and ride into our multi-silicon future. 🌊 It’s been a https://x.com/simran_s_arora/status/1988320513052324127
It’s official: SkyPilot integrates natively with @wandb! With SkyPilot + W&B, you can: 🚀 Launch and scale experiments on any cloud or K8s 🔄 Recover from GPU failures and seamlessly keep tracking experiments 🤝 Collaborate with your team – share SSH access and training logs https://x.com/skypilot_org/status/1989377870469501106
Fast Kernels on AMD hardware!! Great work from @simran_s_arora @_williamhu @Drewwad and team on HipKittens ( https://x.com/AnushElangovan/status/1988393252555493739
14 days. 2.2X faster inference. @Modular_AI + AMD Instinct MI355X GPU = state-of-the-art AI performance. A great example of what’s possible when cutting-edge GPUs meet next-gen AI software. 🔗 https://x.com/AMD/status/1987898172484567238
Build times for gigawatt-scale data centers | Epoch AI https://epoch.ai/data-insights/data-centers-buildout-speeds
Baseten used @nvidia Dynamo to double inference speed for long-context code generation and increased throughput by 1.6x. Dynamo simplifies multi-node inference on Kubernetes, helping us scale deployments while reducing costs. Read the full blog post below👇”” / X https://x.com/basetenco/status/1989058852789317717
Scaling Large MoE Models with Wide Expert Parallelism on NVL72 Rack Scale Systems | NVIDIA Technical Blog https://developer.nvidia.com/blog/scaling-large-moe-models-with-wide-expert-parallelism-on-nvl72-rack-scale-systems/
☁️ NVIDIA Dynamo is now available across major cloud providers–including @awscloud, @googlecloud, @Azure, and @OracleCloud–to enable efficient multi-node inference on Kubernetes in the cloud. And It’s already delivering results: @basetenco is seeing faster, more cost-effective https://x.com/NVIDIADC/status/1989005718083518866
Besides A100s still being in use, I actually expect H100s to have a much longer lifespan than even the A100s From V100 -> A100 -> H100 we had pretty dramatic changes in each generation to reflect the realities of LLM training. The B200/300s are really nice and definitely push”” / X https://x.com/code_star/status/1988062247818850421
the NVFP4 kernels on Blackwell competition has started on @GPU_MODE!!! the first problem, NVFP4 GEMV is now out and submissions can be made. good luck to all the participants!”” / X https://x.com/a1zhang/status/1987972190898450922
Amid the global consensus discussion about an AI bubble. H100 spot price to increase by 8% in Q4 2025. H200 spot price to increase by 18% in Q4 2025. $NVDA https://x.com/FundaBottom/status/1987905008541831521
How cool is this! @Siemens is breaking new ground with an open-source-first, self-contained LLM platform, optimized by @vllm_project. Learn how they deployed their sustainable AI stack with flexibility, full control, and cost savings at scale: https://x.com/NVIDIAAIDev/status/1987944094883037559
The 4.5 trillion dollar elephant in the room https://stevenadler.substack.com/p/the-45-trillion-dollar-elephant-in
Intel CEO to oversee its AI efforts after executive departs for OpenAI | Reuters https://www.reuters.com/business/intel-ceo-oversee-companys-ai-efforts-after-departure-exec-openai-2025-11-10/
A couple of tier 1 frontier labs are saying that NVIDIA is not taking seriously the potential perf per TCO advantage of MI450X UALoE72 for inference workloads especially when factoring in that AMD is offering up to 10% of AMD shares to OpenAI https://x.com/SemiAnalysis_/status/1988044940149235844
So power is the core determinant of where AI data centers are built. Other factors like latency matter surprisingly little — it takes >100× longer to generate model responses than transmit data from Texas to Tokyo. Even serving LLMs from the Moon may not be a big latency issue!”” / X https://x.com/EpochAIResearch/status/1987944164286447895
Where does this power come from? Usually a mix of on-site fossil fuel generation and interconnection to the grid. E.g. Stargate Abilene will start off with on-site natural gas, then connect to the grid to access Texas’ abundant renewable power.”” / X https://x.com/EpochAIResearch/status/1987944176089231428
Nvidia presents TiDAR Think in Diffusion, Talk in Autoregression https://x.com/_akhaliq/status/1988963077690438097
🤖 From this week’s issue: A technical blog post explaining how NVIDIA TensorRT-LLM’s Wide Expert Parallelism efficiently scales large Mixture-of-Experts models on GB200 NVL72 systems, achieving significant performance and cost improvements. https://x.com/dl_weekly/status/1987913458654786008
Leave a Reply