AI Infrastructure news and analysis on Web Pulse: 35646 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Databricks will present four papers and a keynote at VLDB 2026, introducing Lakebase, a serverless Postgres over open lake storage, and LTAP, a paradigm unifying transactional and analytical processing, to meet AI agent …
Alibaba's Qwen team released Qwen3.8-Flash-Next, an open-weight multimodal mixture-of-experts model previewing the architecture for the upcoming Qwen4. The model features 125 billion total parameters with only 6 billion …
OpenAI has added WebMCP support to the built-in browser of its ChatGPT desktop app, enabling websites to expose tools that ChatGPT and Codex can use directly from the page. The feature, which requires GPT-5.6 Sol or Terr…
SK Hynix broke ground on August 27 on a $4 billion semiconductor facility in West Lafayette, Indiana, its first US manufacturing site, which will package high-bandwidth memory (HBM) chips for AI workloads. The project, b…
SK Hynix broke ground Thursday on a $4 billion advanced chip packaging and research facility in West Lafayette, Indiana, with cleanroom operations slated for the second half of 2028, aiming to localize AI memory producti…
OpenAI's first in-house AI chip, Jalapeño, outperformed Nvidia's GB300 in inference efficiency and speed per unit of power in initial benchmarks, signaling a growing challenge to Nvidia's dominance in AI chips. Developed…
Nvidia CEO Jensen Huang and OpenAI CEO Sam Altman will speak at a technology-focused Group of 20 ministerial meeting in Raleigh, North Carolina next week, with SpaceX CEO Elon Musk, Meta executive Dina Powell McCormick, …
Jedify, a context graph technology company, announced that its approach can cut AI token costs by up to 75 percent, citing research showing its context graph architecture averaged 25,036 raw tokens per SQL generation cal…
An OpenAI researcher warned that state-of-the-art AI models running 50 times faster could infiltrate systems before human teams can react, urging autonomous shutdown systems instead of simple monitoring. The warning coin…
Cisco Systems Inc. highlighted Virtual Port Channel (vPC) and Ethernet Segment Identifier (ESI) multi-homing as complementary technologies for scalable, resilient data centers, with ESI multi-homing defined by RFC 7432 a…
Nvidia reported third-fiscal-quarter revenue above expectations and forecast a 70% revenue increase for fiscal 2028, reinforcing its position as the AI boom's bellwether. Separately, AWS acquired DuckLabs, the team behin…
Google Cloud has launched AI-powered Database Operations Agents, including an Onboarding Agent and an Observability Agent, to simplify database lifecycle management. Integrated with Gemini Cloud Assist, these agents supp…
A new survey from the University of Pennsylvania's Annenberg Public Policy Center found that opposition to local data centers jumped 12 points in four months, from 49% in March to 61% in July, while an August Heatmap Pro…
Lovable, the AI app-building platform, argues that the common practice of forcing users to pick a model from a dropdown is a dead end, advocating instead for 'model independence'—a system where a control plane dynamicall…
A hobbyist quantizer known as Aesedai produced a GLM-5.2 GGUF quant using a 512GB RAM workstation with 2x56-core Emerald Rapids Xeons and a single RTX 6000, achieving about 15 tokens per second at 16k context and 400 tok…
NVIDIA reported $96.2 billion in revenue for fiscal Q2 2027, up 106% year-over-year, with net income of $59.7 billion and data center revenue of $89 billion, up 117%, undercutting speculation about a near-term default. T…
Apple Inc. is cutting more than 200 positions across its Vision Pro and Siri-related software organizations, including roughly 100 from Siri, machine learning, and software development teams, according to reports. The cu…
Automaton's durable-state redesign cut GrokBot inference calls by 95%, with 19 of 20 turns served from a local SQLite store instead of replaying model context. The architecture treats context as a cache, persisting claim…
IBM announced a dual-ISA processor built on a 2nm node with 11 high-performance cores running above 5.7 GHz, each natively executing both Arm and IBM Z or LinuxONE instructions, including full hardware AArch64 v9.3 with …
Google announced two new Android app quality requirements this week, including performance thresholds to reduce memory usage and code optimization, in response to industry-wide memory chip shortages driven by the AI data…