AI Infrastructure news and analysis on Web Pulse: 32147 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Taiwan's Zhen Ding Technology has joined Nvidia's MGX ecosystem to supply cable-free PCB-based interconnect technology that replaces hundreds of individual copper cables in AI server racks. The approach uses high-layer-c…
North American PCB bookings surged 25.5% year-over-year in April 2026, pushing the industry's book-to-bill ratio to 1.24—its strongest reading in over a year, according to the Global Electronics Association. The data sig…
Lightrun and Cursor have integrated autonomous error remediation that enables AI agents to detect, diagnose, and fix production errors without human intervention. The system uses Lightrun's runtime instrumentation to cap…
A developer built a search and Q&A system called "Uma Busca de Gelo e Fogo" over the entire *A Song of Ice and Fire* series, processing approximately 66,000 paragraphs across 10 books. The system combines a full-text sea…
Cross-organization agent delegation requires three distinct identities—the delegating agent, the receiving agent, and the end user—to be verified by three separate authorities that do not share a common trust root. The b…
Salesforce's engineering output grew 151.3% year over year in April 2026 after restructuring teams around AI agents, with PRs merged per developer jumping 79% and work items completed per developer rising 50.8%. The shif…
Broadcom lost over $250 billion in market value after its third-quarter fiscal 2026 revenue forecast of approximately $29.4 billion, an 84% year-over-year increase, fell short of Wall Street's elevated expectations. The …
Samsung Electronics and SK hynix are introducing new heat-dissipation technologies for their upcoming eighth-generation HBM5 memory chips, as thermal management becomes a critical challenge for AI accelerators. The memor…
SoftBank Group's concentrated investments in OpenAI have driven a 70% share rally this year but also raised borrowing and credit concerns, with S&P Global Ratings revising the conglomerate's credit outlook to negative. S…
Google released the Google AI Edge Gallery for macOS, enabling Mac users to run Gemma models locally. The company also launched a new Gemma 4 12B variant and the Google AI Edge Eloquent dictation app for Mac. The Edge Ga…
A United Nations University report projects that global data centers could consume nearly 3% of the world's electricity by 2030, rising to 935 trillion watt-hours from 448 trillion watt-hours. The report attributes rough…
Nvidia has acquired Kumo AI, a four-year-old startup that builds predictive foundation models for business data, for at least $400 million according to The Information. The three Kumo co-founders moved to Nvidia last mon…
Bosgame launched the VTA-439 mini PC powered by the AMD Ryzen AI 9 HX 470 processor, delivering 86 TOPS of aggregate AI compute with 55 TOPS from the NPU and 31 TOPS from CPU/GPU. The system supports up to 256 GB of DDR5…
CrowdStrike reported a 15% increase in first-quarter operating expenses to $1.07 billion, while total revenue reached $1.39 billion, exceeding analysts' estimates of $1.36 billion. The cybersecurity firm issued 2027 reve…
A new open-source benchmark, "apple-silicon-llm-bench," reveals that Google's LiteRT-LM runtime outperforms MLX-Swift on the iPhone 17 Pro for Gemma 4 E2B inference, achieving 55.4 tok/s with 4.5x less memory usage. The …
Data center critic Andy Masley argues that most local concerns about data centers are based on misinformation, citing only CO2 emissions, noise pollution, and specific cases like Elon Musk's facilities as valid issues. H…
MCP Gateway, a new tool and capability gateway, allows AI clients to access unlimited backend tools without loading every tool definition into the context window, reducing token overhead by up to 89%. The system replaces…
Apple silicon's integration of CPU, GPU, Neural Engine, and unified memory has turned Mac minis and Mac Studios into the dominant hardware platform for running personal AI agents, driving backorders of four to five month…
With Angular v22, the MCP server and Angular Skills stack enables deterministic, verifiable agent-assisted development. The setup requires configuring MCP servers for Angular CLI and Chrome DevTools, installing Angular s…
BoxAgnts has implemented a capability-based security model for AI agents that replaces root-level access with granular, explicit permissions. The system's WASM execution model requires agents to declare specific capabili…