AI Infrastructure news and analysis on Web Pulse: 35124 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Nvidia has transferred its Open Secure AI Alliance to the Linux Foundation, a move announced on LinkedIn, to provide a neutral home for open-source AI security collaboration. The alliance, founded last month with members…
A developer has built a neural GPU texture block compression system that uses Evolution Strategies to train a small MLP decoder, achieving 5.0 bpp of amortized latent data across a texture stack. The 1,731-weight network…
NVIDIA released Personal AI Router (PAIR), an open source virtual inference router that distributes local AI requests across RTX, DGX Spark, and Mac nodes on a home network, proxying existing Ollama and LM Studio endpoin…
NVIDIA's latest DGX Spark firmware update added a Windows UEFI CA certificate, signaling imminent native Windows support for the device, the most requested feature. The company also confirmed RTX Spark Windows PCs from L…
Equinix Inc. announced at its Horizon event two new services, Equinix Fabric One and Equinix Inference Exchange, to turn its network into the control plane for enterprise AI inference, with Fabric One entering beta later…
A new architectural guide proposes a concrete cancellation contract for stopping AI agents on on-premises infrastructure, separating deadline enforcement from cancellation requests and tracking distinct states such as ru…
Telnyx has launched Canada as a data storage location in its Mission Control Portal, enabling customers deploying voice and AI workflows to keep call recordings, transcripts, AI conversation data, and usage logs within C…
Google's 2019 open-sourcing of Live Transcribe and MediaPipe illustrates why AI tools that package complex capabilities into reusable building blocks outlast those offering only end-user features, according to an analysi…
Daytona, the AI code execution platform, has published a guide for monitoring AI-generated code running in its sandboxes using OpenTelemetry and SigNoz, detailing four non-overlapping telemetry paths: in-sandbox app inst…
Monitoring of the mistralai/mistral-nemo endpoint on deepinfra/fp8 was retired on 2026-09-06 after the B3IT border-input detector could not find enough border inputs, and the endpoint returns no logprobs, making logprob …
Anshvardhan Shetty, a first-year EECS undergraduate at Imperial College London and research intern with Professor Juncheng Yang at Harvard SEAS, presents a study of KV cache admission on high-bandwidth flash using a 222-…
Microsoft introduced Project Zenith on Sept. 4, a new Windows 11 experience that pairs developer-class hardware with a preconfigured, distraction-reduced software environment, first available on devices powered by AMD's …
AC2, an end-to-end infrastructure platform for training and inferencing open weight models, has launched with SDK, runtime, and tracing packages that support coding agents like Claude Code, Cursor, and Codex. The platfor…
AWS launched Lambda MicroVMs on June 22, 2026, to securely execute AI-generated code in isolated Firecracker-based environments, addressing risks such as unintended filesystem operations and excessive permissions. The se…
Cisco Systems Inc. has introduced Unified Edge, a converged hardware platform combining computing, networking, and storage for AI inferencing at the edge, which earned the 2026 Tech Innovation CUBEd Award for the most in…
AMD has unveiled its Threadripper Halo workstation, designed for local AI workloads, offering up to 576 GB of HBM3e memory and 16 TB/s of memory bandwidth, targeting researchers with substantial budgets.
Mistral announced Agentic Search, a feature enabling autonomous search workflows with a data opt-out guarantee for training. The company also highlighted inference optimization for agentic use cases, reporting that Qwen …
Meta researchers developed Proteus, a system that uses AI agents to generate specialized GPU kernels, achieving 1.8–5.2× faster inference for Qwen 3.5 122B models compared to the best kernels in vLLM. The system emphasiz…
Enterprises face a 'token reckoning' in 2026 as usage-based AI contracts lead to bill shock, with Uber's 5,000 engineers exhausting their annual development budget in months. A Stanford Digital Economy Lab study found to…
Micron, in partnership with Tirias Research, highlights that AI inference workloads require rearchitected memory and storage infrastructure to balance performance, efficiency, and scalability. Jim McGregor, founder and p…