AI Infrastructure news and analysis on Web Pulse: 25309 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
NVIDIA released the Nemotron-3-Ultra-550B-A55B-NVFP4 model, a 550-billion-parameter large language model with 55 billion active parameters using NVFP4 quantization, under the OpenMDW-1.1 license. The model, hosted extern…
The AI ecosystem's rapid pace is causing friction with slower-moving societal layers, leading to backlash and instability. Mike Migurski introduced Stewart Brand's Pace Layers framework to analyze how fast-moving AI comp…
Amazon's Project Kuiper satellite broadband constellation is nearing 400 satellites in orbit as the company prepares for commercial service launch. The milestone brings Amazon closer to competing with SpaceX's Starlink i…
Pilot Photonics, a Dublin-based integrated photonics firm, has been awarded €10.4 million from the European Innovation Council Accelerator to scale its patented photonic chip technology for AI data centers, satellite com…
Vulkan 1.4.356 introduces the VK_EXT_shader_ocp_microscaling_types extension, adding support for Open Compute Project Microscaling MX data types (MXFP4, MXFP6, MXFP8, MXINT8) to improve machine learning efficiency. Devel…
Andy Burnham, expected to become UK prime minister as soon as July 20, 2026, is reportedly set to end the NHS's 330-million-pound data contract with Palantir, according to multiple UK outlets. The Palantir-built Federate…
The semiconductor industry group SEMI warned the Trump administration against intervening in the memory chip market, arguing that price or production controls would worsen the ongoing shortage driven by AI demand. The gr…
The Raspberry Pi 5 now comes in a 16 GB version for $120, enabling local AI model inference with MoE models like Qwen3.5 at 7–8 tokens per second, though real-world costs reach $180–220 with necessary accessories. The bo…
Google and Microsoft co-developed WebMCP, a proposed W3C browser standard now in a public origin trial in Chrome 149, to make websites natively compatible with AI agents. The standard lets developers declare structured, …
Jamesob published a guide on building a local system to run state-of-the-art large language models, detailing hardware configurations ranging from $2k to $40k. The setup uses multiple RTX GPUs and PCIe switches for peer-…
China's Ministry of Human Resources and Social Security reported over 200,000 AI-related job vacancies posted by more than 5,000 internet firms during a nationwide recruitment campaign, with hiring expanding into manufac…
Cloudflare detailed its internal unified data platform, Town Lake, reporting that billing workloads account for 53% of all platform queries. The system, built alongside an AI-powered analytics agent called Skipper, unifi…
Microsoft Dynamics partner Volt Technologies published an architecture guide on July 3, 2026 detailing how to integrate Microsoft Business Central with Azure AI services for AI-powered forecasting and cash-flow predictio…
Canada Pension Plan Investment Board (CPP Investments) is investing $2.4 billion alongside EQT and its EdgeConneX data-center platform to build AI infrastructure, increasing its exposure to a sector with durable long-ter…
Oxmiq, founded by Raja Koduri, raised $35 million to develop OxCore, a licensable GPU IP built on RISC-V that enables CUDA and PyTorch code to run on non-NVIDIA hardware without modification. The company's OxPython porta…
South Korean memory chipmaker SK hynix announced a 1,100 trillion won ($713 billion) investment plan to expand domestic manufacturing capacity, including new fabs for NAND and HBM, and will list on the Nasdaq to help fun…
Kubernetes Horizontal Pod Autoscaler (HPA) automatically adjusts pod replica counts based on real-time demand, addressing the industry-wide issue of average CPU utilization of just 8% across production clusters. By dynam…
TwelveLabs raised $100M in Series B funding on July 1, with Amazon as an investor and AWS as its preferred cloud provider. The company offers video understanding APIs that index visual, speech, and text content, with mod…
Condense.chat launched a public context-compression proxy for AI coding agents that reduces API bills by up to 72%. The proxy sits between agents and upstream models, compressing repeated context in long sessions, and re…
Amazon's AI data center expansion drove a record 16% rise in greenhouse gas emissions and a 34% jump in electricity use in 2025, according to its latest environmental report. The company's total emissions hit an all-time…