AI Infrastructure news and analysis on Web Pulse: 35683 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
The Cloud Security Alliance's Top Threats to Cloud Computing 2026 report separates AI-Enhanced Attacks and AI System Compromise into two distinct categories, ranked #2 and #6 respectively, according to the CSA. AI-enhanc…
Zhipu AI revealed that the anonymous model "Ox Alpha" is GLM-5.3-Flash, an MIT-licensed open-weights model released on Hugging Face on August 26 after a six-day stealth run that accumulated 44 trillion tokens and 503,000…
Google Cloud's Sensitive Data Protection and Model Armor can help enterprises balance protecting sensitive data with maintaining LLM functionality, according to a blog post by Leo Y. The post argues that aggressively blo…
Delaware Governor Matt Meyer signed a first-of-its-kind bill requiring data centers to power their operations with clean energy, pay their fair share of electricity costs, and fund grid infrastructure upgrades, a move ce…
The NVM Express consortium released NVMe 2.4, adding post-quantum cryptography support, persistent lockdown settings, voltage monitoring, and idle I/O exit latency limits. Mike Allison, chair of the NVMe Errata Task Grou…
Amazon is shutting down Mechanical Turk (MTurk) on September 30, 2026, after 21 years, ending the crowdsourced digital labor platform that helped build the human-data pipeline for AI. The company stopped accepting new cu…
Telnyx has released an open-source AI Call Campaign Orchestrator built with Python, Flask, Telnyx Call Control, the Telnyx Agent SDK, SQLite, and SMS. The tool provides a backend template for outbound call campaigns, han…
A developer argues that agentic AI systems are often designed backwards, starting with technology rather than product constraints. They propose a constraint-first approach structured around latency, cost, failure, and ev…
A developer tested the Qwen3.8-Flash-Next model on an NVIDIA DGX Spark (GB10, 128 GB unified memory) using llama.cpp with quantized KV cache, achieving 128K context and 414 tok/s prefill. They identified a fix for a llam…
Z.ai released GLM-5.3-Flash, a 320B-total-parameter mixture-of-experts model with 18B active parameters, a 1,048,576-token context window, and native image and video input, under an MIT license with weights on Hugging Fa…
Researchers introduced Maia 200, an AI accelerator delivering 10,145 Tflop/s FP4 and 5,072 Tflop/s FP8 performance within a 750W TDP and 7 TB/s HBM bandwidth, as detailed in a paper submitted to arXiv on August 25, 2026.…
Z.ai revealed that the mystery model 'ox-alpha' seen on OpenRouter and OpenCode last week is actually GLM-5.3-Flash, now available in Kilo across IDE extension, CLI, and Cloud Agents. The 320B-parameter model with 18B ac…
Meta Platforms Inc. unveiled its MTIA 400 accelerator at the Hot Chips conference, designed for both LLM training and ad-serving inference, delivering 12 petaFLOPS of MXFP4 compute at 1.7 GHz, which is about 20 percent f…
Imageat has released Qwen 3.8 27B Uncensored as a hosted chat API, enabling developers to integrate the long-context model without managing infrastructure. The model supports a 262,144-token context budget, per-request t…
Cloudflare's Kitesurf, a browser built for AI agent workloads that runs inside V8 isolates on Workers without Chromium, cuts CPU and memory use 3-7x versus Chromium on common agent tasks, signaling a shift away from head…
Amazon Web Services (AWS) and NVIDIA announced an expansion of their strategic collaboration, with AWS planning to deploy 2 million additional NVIDIA GPUs across its global infrastructure in 2027–2028. The companies will…
AWS and Nvidia announced plans to supply an additional 2 million GPUs to enhance infrastructure for agentic and physical AI systems, building on AWS's existing deployment of over 1 million Nvidia GPUs across global cloud…
NVIDIA introduced NVLink Fusion and NVHBM, a custom HBM base-die technology validated with leading memory vendors, to help hyperscalers and AI-native companies deploy custom XPUs into NVIDIA's AI infrastructure platform.…
NVIDIA expanded its NVLink Fusion platform with NVHBM, a custom high-bandwidth memory technology that integrates NVIDIA's memory controller into the HBM base die, delivering up to 30% greater memory bandwidth, 15% lower …
AWS and NVIDIA announced a major expansion of their strategic collaboration, planning to deploy 2 million additional NVIDIA GPUs across AWS's global infrastructure in 2027-2028, including 100,000 GPUs for U.S. government…