Infrastructure & Hardware
Infrastructure and hardware comprise the physical compute systems, specialized chips, and data center components that power artificial intelligence workloads. Truncated covers chip architectures, supply chains, energy requirements, and the hardware deployments that enable large scale computing.
Baidu's ERNIE Evolve Hid Its Scorer, Then Posted 6.2x Gain on Kunlun Chips
techtimes.com — Baidu ERNIE Evolve optimized Kunlun P800 chip inference by 623.6 percent, but the same report shows far smaller GPU gains, and outside analysts say the Kunlun baseline started much less tuned. The AI agent making the changes never saw its own scoring code, a design meant to stop the kind of
Caterpillar Stocks Jump 6.61% as Anthropic Takes Over $3.4 Billion Generator Order
finance.yahoo.com — Anthropic secured 1.7 gigawatts of Caterpillar gas generators originally ordered by OnePWR, as AI labs race to lock in their own power.
Advanced Micro Devices (AMD) Lands EPYC Role In New AI Security System
finance.yahoo.com — Advanced Micro Devices (NasdaqGS:AMD) has been selected by Google Cloud to supply EPYC CPUs for the new TPU 8tx system. Google is shifting the host layer of its flagship TPU 8tx platform back to x86 CPUs, after previously relying on in house chips. The TPU 8tx infrastructure is designed with hardware enforced confidential computing to protect sensitive AI workloads running on Google Cloud. The choice of AMD EPYC for Google Cloud's TPU 8tx host systems is important, but the rest of the AMD...
Google’s chip business is on a faster track than Nvidia’s own AI boom
finance.yahoo.com — As Nvidia investigates its CEO, Google is rolling out more chips built to win Nvidia's customers on price.
Why Alphabet Stock Was Moving Higher Today
finance.yahoo.com — Alphabet introduced two new chips this morning.
FedEx adds four weekly China-US freighter flights as AI hardware demand keeps cargo busy
investinglive.com — FedEx is adding four weekly Boeing 777 freighter flights linking Shanghai and Guangzhou with its US hubs, targeting AI infrastructure, chips and advanced…
India's electronics firms face PCB squeeze as AI demand drives up costs
asia.nikkei.com — Impact is being passed on to customers as prices for materials rise as much as 350%
AI Data Centers Raid Used Turbine Market as New Units Queue to 2031
www.tftc.io — Global turbine orders top 100 GW against 60-70 GW annual capacity. Crusoe just ordered 650 MW of refurbished jet-engine generators. The energy scarcity thesis is no longer theoretical.
In Focus | Escaping Russia’s web: Central Asia caught between Chinese networks, US curbs
www.scmp.com — Powers including Beijing, Washington and Moscow are competing to supply the region with the energy infrastructure needed to power AI.
Announcing Long-Context Training on TPUs with MaxText Context Parallelism- Google Developers Blog
Google Cloud Makes Pathways Runtime GA for Large-Scale TPU Workloads
www.unite.ai — Google Cloud on November 9, 2026, announced that Cloud Pathways is generally available on Google Kubernetes Engine (GKE), bringing a single-controller runtime for large-scale training and inference to customers. Google s...
E2B Tests NVIDIA Vera for High-Density Agent Sandbox Fleets
www.unite.ai — E2B published pre-production test results on November 9, 2026, showing that four tuned runs on one dual-socket NVIDIA Vera server sustained 1,400 to 1,550 loaded sandboxes, while separate microbenchmarks measured higher...
Google claims its new AI software cuts inference latency by over 90% — but there's a catch
thenewstack.io — Google’s new open-source TPU Sync aims to speed AI inference by moving KV-cache data between memory tiers, cutting time to first token by up to 90%.
How to run open models in production without owning the serving stack
Google Cloud's new TPUs were designed for context and security
www.thestack.technology — They tackle two enterprise AI concerns: scaling agents eats up memory, and private data needs confidential computing.
Agentic AI Will Hit an Infrastructure Wall Before It Hits an Intelligence Wall
www.unite.ai — The Models Are Improving Faster Than Our Ability to Serve What They’re Being Asked to Do. Most of the argument about where AI agents are headed is an argument about capability. Can the model plan reliably? Can it recover...
Fearless Engineering: The CS-4 is our first engineering documentary.
The rise of sovereign AI factories: Why the C-suite is rejecting cloud dependency
www.cio.com — National data sovereignty and security demands are driving enterprises away from public clouds and toward on-premises supercomputing. Here is how organizations are converging HPC and AI to maintain control over their critical data and workloads.
The industrialization of AI: 8 critical considerations for building an AI gigafactory
www.cio.com — Building a giga-scale AI facility requires rethinking traditional data center design. Here is how technology leaders can navigate power, cooling, and networking demands to sustain next-generation AI workloads.
Dismiss email signup
Subscribe to 
The top AI news in your email delivered each morning. Choose categories