Section
Infrastructure – Page 5
The latest Infrastructure news, analysed daily by the i10X newsroom.
FilterCompanies & models
More stories
-
AI Boom Drives Record Data Center Capex Amid Power Constraints
Generative AI is driving unprecedented data center capital expenditure as bottlenecks shift from GPUs to power infrastructure. Learn how speed-to-power defines the new AI competitive moat.…
-
Industrialized AI Research: Agentic Pipelines and Compute Limits
Explore how AI research has shifted from artisanal methods to industrialized, agent-assisted pipelines driven by compute constraints and automation. Discover the impact on labs, researchers, and…
-
Real-Time Voice AI Stress-Tests Mobile Edge Computing
OpenAI Advanced Voice Mode and Google Gemini Live are pushing mobile networks and edge infrastructure to their limits. Discover the under-reported challenges for travelers and hardware.
-
Compute-Optimal Training: AI Scaling Laws Hit Data Wall
Chinchilla findings show optimal AI training balances parameters and data. Learn how the data wall is forcing shifts to synthetic data, test-time compute, and new infrastructure…
-
Data Centers Are Becoming Intelligence Factories for AI
AI training clusters demand 30-80 kW racks, obsoleting air cooling and forcing liquid systems plus gigawatt-scale power. Learn how physical constraints now limit LLM progress beyond…
-
Model Fatigue: Enterprises Shift to Stable AI Infrastructure
Rapid AI model releases are causing integration chaos and hidden TCO for enterprises. Discover why stability, LTS models, and abstraction layers now matter more than benchmark…
-
Synchronized AI Outage Exposes Cloud Concentration Risk
The September 3 outage hit ChatGPT, Claude, and Grok simultaneously, revealing shared infrastructure dependencies. Learn why multi-LLM failover strategies fail and how to build resilient AI…
-
GPT-6 Astra Myth: Project Astra vs GPT-4o Race
The viral ‘GPT-6 Astra’ search stems from confusion between Google’s Project Astra and OpenAI’s GPT-4o. Discover the real shift to real-time multimodal AI agents and infrastructure…
-
AI Data Centers: Power Density, Cooling & Scaling Limits
AI data centers are shifting to 30-100kW racks and liquid cooling as power and grid constraints limit LLM scaling. This analysis covers hyperscale builds, sovereign AI,…
-
Moonshot AI IPO: $50B Valuation on HKEX Explained
Moonshot AI, behind Kimi chatbot, targets a $50B IPO on HKEX via Chapter 18C. Discover how it navigates compute constraints and sets benchmarks for China’s AI…
-
Gemini Nano 4: Hardware-Gated AI Arrives on Android
Google mandates Gemini Nano 3 for core AI features while rolling out Nano 4 on Pixel 11. Discover the impact on developers, OEMs, and on-device inference.…
-
Gemini 3.5 Transcribe: Enterprise Speech-to-Text from Google
Explore Gemini 3.5 Transcribe, Google’s new speech-to-text model with 2.6% WER, streaming support, and Vertex AI security for regulated industries. Learn more.