AI Infrastructure news and analysis on Web Pulse: 35641 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
A new analysis warns that autonomous prior-authorization agents in healthcare can fail when vector databases retrieve expired payer policies, causing denials. The article describes a case where an agent cited a retired 2…
SK hynix Inc. broke ground on its first U.S. High Bandwidth Memory (HBM) advanced packaging production facility in West Lafayette, Indiana, with mass production targeted for Q2 2029. The ceremony at Purdue University was…
Salesforce and Anthropic announced Claudeforce, a partnership integrating Salesforce data, workflows, and governance into Anthropic's Claude and Slack, with Claude becoming the default model for Slack and a reasoning mod…
InferCrane, an open-source inference operations platform, released v1.0.0-rc.1, a public beta CLI and SDK that deploys, operates, and safely evolves self-hosted AI inference behind a stable OpenAI-compatible endpoint. Th…
Apple's 2026 Mac Studio lineup, powered by the M5 Max and M5 Ultra chips, offers significant upgrades over the 2025 M4 Max and M3 Ultra models, including faster CPUs, enhanced GPUs with Neural Accelerators, and higher me…
ThunderPhone's v2 release abandons the traditional STT-LLM-TTS voice AI pipeline in favor of a multi-model, audio-aware architecture that runs multiple transcription models simultaneously and pipes raw audio into the LLM…
Carlos Cortez released Tinkuy 0.1.0, a minimalist agent framework written in 467 lines that provides a provider-agnostic loop for LLM tool calls. The framework, published on npm as @carloscortezcloud/tinkuy-agent, includ…
A critical vulnerability in vLLM's tool-call parser, CVE-2025-9141, allows model-generated arguments to be passed directly to eval(), enabling arbitrary code execution. The flaw highlights a broader pattern in AI inferen…
Amazon Web Services plans to deploy 2 million additional NVIDIA GPUs in 2027 and 2028, on top of more than 1 million GPUs previously planned beginning in 2026, according to an announcement at NVIDIA GTC 2026. The expansi…
At least 75 data center projects representing roughly $130 billion in investment have been blocked or delayed due to local community opposition, prompting Microsoft, Amazon, Google, and Meta to shift toward onsite power …
Oxmiq Labs presented high-bandwidth flash (HBF) as a capacity tier for AI inference at Hot Chips 2026, claiming it delivers 8 to 16 times the capacity of HBM at the same cost. The company detailed HBF specifications acro…
Deepgram released Enhanced Metrics and Prometheus/OpenTelemetry support for its speech-to-text and text-to-speech models on Amazon SageMaker AI, enabling customers to monitor billing, feature usage, and per-GPU engine me…
A user reports running Gemma4:26b on a GPU at 2000+ tokens/s pre-fill and 80+ tokens/s evaluation, and Laguna S 2.1 118B on a CPU server at 13-14 tokens/s, expressing frustration at the lack of models between 30B and 120…
Data centers worldwide consumed 788 terawatt-hours of electricity in 2025, up 19.7% from 658.2 TWh in 2024, according to the Energy Institute's 2026 Statistical Review of World Energy, which tracked data-center power dem…
Stripe has acquired an AI model routing company, a move that signals the payments giant's push to integrate intelligent model selection into its infrastructure. The acquisition aims to optimize AI-driven payment processi…
AWS, NVIDIA, and Heidi Health report that using NVIDIA CUDA Multi-Process Service (MPS) with NVIDIA Triton Inference Server on Amazon EC2 GPU instances cuts ASR inference GPU infrastructure requirements by 75%, from 16 i…
A developer built a local knowledge base on Actian VectorAI DB, emphasizing that the database does not embed text but stores vectors and runs nearest-neighbor searches. The project separates concerns: an embedding model …
SK hynix Inc. held a groundbreaking ceremony on August 27, 2026, for its first U.S. High Bandwidth Memory (HBM) production facility in West Lafayette, Indiana, with an investment of over $4 billion. The fab is scheduled …
A new report warns that AI agents are outpacing human patching, with 80% of organizations reporting agents operating beyond their intended scope and AI-discovered zero-day exploits mathematically outrunning human remedia…
A developer reports that automated scanners and LLM-based agents in production can be detected by their overly predictable timing, citing a MediSys sandbox where a scanner maintained a 4.2-second scan interval and 200ms …