AI Infrastructure news and analysis on Web Pulse: 25737 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Anthropic launched Claude Science, an AI workbench for scientists, in beta on Tuesday. The tool runs locally on Mac, Linux, or remote machines and aims to accelerate scientific research by integrating tools and automatin…
Anthropic's Claude Sonnet 5 is now available on Snowflake Cortex AI in private preview, offering enhanced reasoning, coding, and agentic capabilities within Snowflake's secure perimeter. The model approaches Opus-level p…
Schneider Electric acquired Norwegian industrial AI company Cognite for $3.1bn in cash, aiming to integrate its data and AI platform into its Aveva software unit. The deal positions Schneider to lead in industrial AI, wh…
Cloudflare launched Temporary Accounts on June 19, enabling AI agents to deploy Cloudflare Workers without signing up or providing credentials. The feature provisions a temporary preview account with a 60-minute window f…
Moon Equity Holdings Corp. licensed its IndexR AIVROS system to Revvim for commercialization, targeting over 300 enterprise customers and 100 digital marketing agencies. The system consolidates 18 fragmented tools for SE…
DeepSeek launched DSpark, an open-source speculative decoding module that boosts AI inference speed by up to 400% on existing chips, potentially reducing demand for Nvidia's high-end accelerators. The software, designed …
AI chip startup Etched announced it has booked $1 billion in contract orders and raised $800 million to date, including a $500 million round at a $5 billion valuation, as it emerges from stealth with a specialized chip f…
Trix Cyrus introduces context engineering as a discipline that goes beyond prompt engineering to curate everything a language model sees during inference, including system prompts, retrieved documents, conversation histo…
Amazon Web Services committed $1 billion to a new Forward Deployed Engineering program, sending small teams of engineers to work directly with clients for 45-day engagements to build custom AI systems. The initiative, an…
Databricks co-founder Reynold Xin explains the limitations of traditional monolithic OLTP databases and introduces Lakebase, a serverless Postgres database that externalizes storage and compute. The architecture separate…
Bending Spoons, the Milan-based acquirer of AOL, Evernote, Vimeo, and WeTransfer, priced its Nasdaq IPO above the $26-$28 range at a valuation of roughly $19 billion, driven by strong demand from Wall Street. The company…
Giles Thomas built a JAX training loop for an LLM from scratch, using NNX and Optax, as part of a project to replicate GPT-2 small quality without referencing Sebastian Raschka's book. He trained a base model on 3.2B tok…
A new framework reveals that most companies claiming AI transformation have only added tools without changing how work gets done, creating a dangerous gap between hype and real operating-model change. The six-stage ladde…
Phantom has acquired the team behind Ventuals, a platform for permissionless perpetual futures on private company valuations, to bolster its on-chain trading capabilities. The Ventuals team, led by CEO Alvin Hsia and CTO…
Pump.fun has deprecated its Tokenized Agent launch option as of June 30, 2026, after community backlash over toxic player-versus-player dynamics. The Solana-based token launchpad removed the feature, which allowed AI age…
Anthropic is preparing to release Claude Sonnet 5, codenamed "Fennec," on OpenRouter, according to reports. The model, which was rumored for a February 2026 launch that did not materialize, is now nearing availability on…
US power grid operators warn that a dangerous heatwave could push electricity demand near record levels this week, coinciding with surging energy consumption from AI data centers. PJM Interconnection forecasts a record s…
Rob Hanna, CEO of Precision Content, argues that enterprise AI initiatives fail because organizations treat language like structured data while neglecting content governance. He says technical publications teams already …
RadixArk released Miles, an open-source PyTorch-native framework for large-scale LLM reinforcement learning post-training. The framework integrates SGLang for rollout, NVIDIA Megatron-LM for training, and Ray orchestrati…
A developer accidentally built a closed-loop self-healing system for AI configuration after noticing repeated rule-writing without behavior change. The system uses three files—config-health.py, pending-verifications.md, …