BIP NYC

collapse
Daily News Analysis
    THORChain halts swaps after block reorg bug, slowing Binance hackers

    THORChain halts swaps after block reorg bug, slowing Binance hackers

    Nov 02, 2026  Twila Rosenbaum
    THORChain node operators paused cross-chain swaps on ten networks after a one-block Ethereum reorganization exposed a refund bug that paid out 10 ETH the vault never received. The emergency halt also blocked North Korean-linked launderers from swapping assets stolen in the record Binance hack. Node operators will re-enable each chain individually once it is deemed safe, while trading continues on Avalanche, Base and XRP.
    Zoom adds ‘proof of human’ badges to show you’re not a deepfake

    Zoom adds ‘proof of human’ badges to show you’re not a deepfake

    Nov 02, 2026  Twila Rosenbaum
    Zoom enterprise customers worldwide can now display 'proof of human' badges during video meetings, powered by World's biometric identity system. The verification cross-checks an iris scan from World's Orb, a live selfie and the participant's video feed before showing a Human Badge to everyone in the call. The rollout follows costly deepfake incidents, including a $25 million transfer triggered by fake executives on a video conference and North Korea-linked hackers using fabricated Zoom feeds to spread malware.
    Cisco, Nvidia push ultra-dense networking for Ethernet AI fabrics

    Cisco, Nvidia push ultra-dense networking for Ethernet AI fabrics

    Nov 01, 2026  Twila Rosenbaum
    Cisco and Nvidia have introduced USFP, a high-density pluggable optical module designed to speed Ethernet AI fabric deployment. It supports 400 Gb/s per lane and scales to 12.8T of aggregate capacity while preserving field-serviceable pluggable operations. Cisco also expanded Silicon One with the G302 top-of-rack switch and new air- and water-cooled systems.
    It’s not your model — AI has a data movement problem

    It’s not your model — AI has a data movement problem

    Nov 01, 2026  Twila Rosenbaum
    Organizations are investing heavily in AI models, GPUs and licenses, yet many deployments stall because the network cannot carry machine-speed, bursty traffic across distributed environments. Bandwidth issues and latency have risen sharply, but leaders often respond by buying more compute instead of fixing the data path. The article outlines why end-to-end connectivity, routing and security inspection matter more than raw model capability, and what executives should assess before scaling AI.
    ICANN set to expand gTLD: Brands, AI, and crypto all get new domains

    ICANN set to expand gTLD: Brands, AI, and crypto all get new domains

    Nov 01, 2026  Twila Rosenbaum
    ICANN has opened its first new generic top-level domain application round in 14 years, drawing 1,615 proposals. Corporate brands, AI terms, cryptocurrency names, network jargon and geographic strings are all in the mix. The most contested string is .agent, with 13 applicants, while .bit follows with 11. The final list is due on Nov. 17, 2026, before objections and a prioritization draw.
    Gartner warns $3.5T in AI infrastructure spending could be at risk by 2030

    Gartner warns $3.5T in AI infrastructure spending could be at risk by 2030

    Nov 01, 2026  Twila Rosenbaum
    A leading research firm warns that AI infrastructure spending could outpace revenue by nearly $5 trillion through 2030, putting more than $3.5 trillion at risk of impairment. Neocloud providers face the sharpest pressure as hyperscalers dominate demand and financing costs rise. Industry analysts and Elon Musk both predict consolidation and failures among AI cloud startups.
    Oracle adds Google TPUs to OCI to expand enterprise AI compute options

    Oracle adds Google TPUs to OCI to expand enterprise AI compute options

    Nov 01, 2026  Twila Rosenbaum
    Oracle has partnered with Google to make Google Tensor Processing Units available through Oracle Cloud Infrastructure, marking the first time the chips will be offered by another hyperscaler. Oracle will supply the data center infrastructure, including power, cooling and networking, while Google provides the TPU technology. The companies say the option will expand enterprise AI compute choices as customers look beyond GPUs for training and inference. Pricing, specific configurations and initial capacity have not been disclosed, and availability is planned for calendar 2027.
    Google is making Gemini and Search work better with the apps you love

    Google is making Gemini and Search work better with the apps you love

    Nov 01, 2026  Twila Rosenbaum
    Google is expanding Gemini and Search with a new Connected Apps Platform designed to make third-party services easier to integrate. More than 50 connected apps are already available, and partners including Adobe, Experian, Localiza, Peloton, and Calm are joining or expanding across the two platforms. Developers can now build one integration and choose whether it appears in Gemini, Search, or both, with a dashboard for testing, review, and performance tracking. The platform is launching in preview, so broader access and new app support will roll out gradually.
    Samsung’s first AI glasses could arrive in December, but only in two countries

    Samsung’s first AI glasses could arrive in December, but only in two countries

    Nov 01, 2026  Twila Rosenbaum
    Samsung is reportedly preparing to launch its first AI smart glasses, called Intelligent Eyewear, in mid-December. The debut is expected to be limited to the US and South Korea, with no confirmed price or wider rollout date. The Android XR glasses use Gemini AI, cameras, and microphones for translations, navigation, and note capture, but the first generation lacks a display.
Your experience on this site will be improved by allowing cookies Cookie Policy