AI Infrastructure news and analysis on Web Pulse: 32396 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
NVIDIA today announced new software, open source models and partnerships with leading software platform providers to build autonomous AI agents for industries and enterprises. Cadence, Dassault Systèmes, Siemens and Syno…
NVIDIA introduced the Vera CPU, designed specifically to accelerate agentic AI and reinforcement learning workloads in AI factories by improving CPU execution speed and task throughput. The chip combines 88 Olympus cores…
NVIDIA today announced the launch of Vera, its first CPU designed specifically for AI agents, now in full production and enabling 1.8x faster task completion than x86 CPUs. Global AI labs including Anthropic, OpenAI and …
NVIDIA released DSX OS, an open and modular software platform designed to operate and scale AI factories, aiming to improve efficiency and lower token production costs. The software provides open-source components that c…
Intel announced Xeon 6+ processors, Ethernet E835 controllers, and updates on its Crescent Island GPU at Computex 2026, positioning the CPU as the control plane for agentic AI workloads. The new chips, built on Intel 18A…
Intel announced at Computex that over 130 customers are using its Series 3 processors for edge AI and robotics designs, including Sensory AI's Ella multi-agent Physical AI store. The company also introduced OpenVINO Phys…
The open-source LLM inference landscape in 2025–2026 has split into three specialized tiers: throughput-oriented serving engines for production GPU clusters, portability-focused runtimes for consumer hardware and edge de…
Anthropic confidentially filed a draft S-1 with the SEC for a proposed public offering, marking the first initial public offering from a frontier AI lab. NVIDIA opened its Cosmos 3 physical-AI model, ramped Vera Rubin in…
LoreConvo offers a local-first AI memory tool that stores data on the user's machine without requiring cloud infrastructure or multiple services. Unlike competitors that need Docker, a relational database, and a vector d…
PostHog has released a guide for small teams and solo founders on setting up minimum viable AI observability after shipping their first AI feature. The guide recommends starting with just three components—tracing every L…
An AI agent deployed by PostHog during a Lisbon team offsite identified a three-year-old bug in the company's ClickHouse query engine that prevented timestamp filters from using the primary key correctly. The fix reduced…
Short interest across AI-related stocks has risen 24% in the last quarter, with GPU data center and neocloud companies facing the heaviest bearish bets at a median 16.8% of float shorted. The skepticism is concentrated i…
Developers seeking reliable JSON output from large language models can move beyond simple prompt instructions and retry loops by implementing constrained sampling techniques that mask invalid tokens during generation. By…
SigNoz has released a new guide for monitoring DeepSeek API calls using OpenTelemetry, enabling developers to track model latency, error rates, and token usage in unified dashboards. The integration allows users to instr…
Google has released a guide for setting up observability and monitoring for its Agent Development Kit (ADK) using OpenTelemetry, enabling developers to capture traces, logs, and metrics from AI agent applications. The in…
SigNoz has released a guide for setting up OpenRouter observability using OpenTelemetry, enabling developers to track LLM API requests, latency, and usage across multiple models in unified dashboards. The integration all…
OpenTelemetry (OTel) has become the second most active project in the Cloud Native Computing Foundation by standardizing how applications generate and transmit telemetry data across traces, metrics, and logs. The open-so…
An AI agent experiment earned $0 on its third day after shipping five public software packages and updated installation documentation for its OpenTrust and Hands Body and Feet toolchain. The agent's creator shifted strat…
At Computex 2026, Nvidia CEO Jensen Huang announced the Vera CPU, purpose-built for agentic AI workloads, with Anthropic, OpenAI, and SpaceX as anchor customers. The chip delivers 1.8x faster task completion than x86 CPU…
Cockroach Labs' David Joy warns that choosing a database for AI products based on experimentation-stage performance leads to production failures, urging teams to evaluate seven criteria including elastic horizontal scali…