Archive, page 18
Microsoft will make in-house Azure Maia chips its main AI inference system starting 2027 at multi-gigawatt scale; Maia 300 should deliver 40% more tokens per megawatt than leading GPU systems.
At OCP 2026, Cisco announced G302 silicon and USFP optics developed with Nvidia for AI networking.
The newsletter Import AI's 476th issue, with Clark flagging China's open-model advantage, covers the latest AI developments for 2026.
Pallix added Grok and Claude tracking, query fan-out, and ready-to-paste fixes to its AI visibility monitoring platform.
Agent Role Contracts v0.5.0 is a tool for checking AI-agent workflows against defined role contracts to catch misbehavior before deployment.
GUDEA raised $7 million in seed funding for AI narrative intelligence, technology that analyzes how stories and narratives spread.
Samsung is reportedly considering AI-powered widget design tools for its One UI 9.5 software update.
Nvidia's Arm-based Vera CPU has started shipping, targeting the server CPU market as part of its Vera Rubin platform for AI datacenters.
Nvidia added support for Google's Falcon transport protocol in its networking stack, aiming to address growing diversity in datacenter transport demands for AI workloads.
Users reported possible OpenAI service disruptions, according to outage tracking. The scope and cause were not immediately confirmed.
An APK teardown reveals Google Maps is building code for Gemini-powered route-impact comparisons in navigation alerts, letting the app explain how incidents affect a route. Not yet live to users.
A Reddit post notes Claude Code's Projects update splits one message into parallel tasks, each on its own branch, running while the laptop is closed, though users focus on usage limits.
Google's Gemini Notebook, formerly NotebookLM, released iPhone and iPad version 1.61.7 with bug fixes and improvements to its research and study tools.
NVIDIA opens its Vera Rubin architecture to help scale AI factories from megawatts to gigawatts, as the industry adopts open standards at OCP Global Summit.
Nvidia and Cisco introduced USFP, a new pluggable optics form factor for AI data centers with up to 12.8 Tb/s capacity, alongside Cisco ICs for AI fabrics, aimed at scaling AI factory connectivity.
DG Matrix unveiled Interport 360 SST and Interport OS, power infrastructure based on NVIDIA MGX, for AI factories.
Nvidia contributed its Vera Rubin rack and compute tray designs to the Open Compute Project, stripping cabling, hoses and fans, as part of broader open rack and networking work with AMD, Google and Meta.
Staffing firm Frontline Source Group says its AI chat assistant has been promoted to a full AI recruiter role. Few concrete details on capabilities or availability were given.
Microsoft unveiled its third-generation in-house AI chip, with deployment potentially exceeding 1 million units next year, aiming to cut costs.
Developer RafaelNGP released a native Xbox Game Pass app for jailbroken PS5s, enabling Xbox Cloud Gaming with four controllers and 4K upscaling via the PS5 GPU, built using Claude.
GWI announced a new agentic portfolio of offerings. The source gives no concrete product details, pricing or availability.
Report says DeepSeek is accelerating into embodied AI, with Liang Wenfeng courting robotics startups. No product or deal confirmed.
AlphaSense launched SuperAnalyst, a next-generation AI agent aimed at high-value financial and strategic workflows. The company says it powers research and analysis tasks for finance teams.
DeepMind's AlphaProof Nexus reportedly solved nine open Erdos problems and 44 number-sequence conjectures, some over 50 years old. First announced in a May 2026 arXiv preprint.
Dice article covers FinOps cost-per-outcome practices for engineers building AI agents. Advice content without a new product or event.
Microsoft's senior director of AI systems hardware architecture has left to join OpenAI, per Data Center Dynamics. The move signals OpenAI's continued buildout of custom AI hardware capability.
ASML expects AI-driven chip demand to keep booming past 2028, while warning that financing risk is rising across the sector.
IT services firm HCLTech is partnering with Anthropic to expand its enterprise AI play, per Mobile World Live. Financial terms and scope were not disclosed.
Microsoft is challenging Nvidia with its new Maia 300 AI chip, which it says delivers 40% greater efficiency.
Elon Musk said uploading a credit card photo lets Grok shop for you, while Visa flags rising fraud risks in agentic payments. The claim highlights gaps in agent commerce safeguards.
Visorway raised an angel-backed pre-seed round at a CHF 3 million valuation to scale AI-native procurement services.
Intel will discontinue SGX after Xeon Diamond Rapids, shifting focus to TDX for cloud, enterprise and agentic AI security needs.
The Information details OpenAI's plan to revive its Sora video generation product. Specifics are behind a paywall.
Peter Mo, formerly of Plaud, raised $26 million for Ambi, an always-listening AI assistant.
Thurrott.com piece heralds the arrival of Nvidia's RTX Spark era, following Microsoft's new AI chip news.
A young entrepreneur named Aidan Guo is building an AI desktop assistant, and his startup has already attracted backing from investors, according to The Economic Times.
Hardware.co announced custom AI hardware and private AI infrastructure solutions, powered by DEV.co, aimed at organizations that want to run AI on their own systems.
Fedstack, a Smoothstack company, said it has been named an OpenAI Select Partner, a designation within OpenAI's partner program for companies building on its models.
Elon Musk promoted how easily Grok can complete shopping purchases for users, saying you can just take a picture of your credit card to have the AI buy items.
A News Anyway piece assesses ChatGPT tasks aimed at small businesses, describing real productivity gains alongside caveats about reliability and oversight.